S K Dogra - Molecular Spectroscopy-McGraw-Hill Education (2012)

Molecular Spectroscopy
About the Author
S K Dogra received a PhD degree in Chemistry from The University of British Columbia, Vancouver, BC
Canada, in 1970, followed by post-doctorate work in USA and Canada, before moving to the Department of
Chemistry, IIT Kanpur in 1973, where he stayed until his retirement, up to 2004. At IIT Kanpur, he was involved
in both undergraduate engineering and science and post-graduate courses in Chemistry and Physical Chemistry
in particular. His main area of research involved proton transfer and electron transfer reactions in the excited
singlet state. He has authored over 160 research articles in national and international journals, wrote three books
and authored chapters in a couple of books. He has been involved in the curriculum development of NCERT and
is writing books for school teaching. He is also writing chapters for the CSIR e-learning program. Dr Dogra is
a fellow of National Academy of Sciences and Indian National Academy of Sciences and was elected the Vice-
President of Indian Chemical Society. He was on the editorial board of Indian Journal of Chemistry and editorial
board of INSA Proceedings until now. He is a recipient of the T R Seshadri 70th Birthday Memorial Award from
INSA and Lifetime Achievement Award from the Indian Chemical Society.
H S Randhawa has retired as Assistant Professor, Department of Chemistry and Biochemistry, Punjab Agricultural
University, Ludhiana, after a distinguished career spanning four decades. He graduated from Punjab University,
Chandigarh, in 1962 securing the second position and was Gold Medalist in MSc (Chemistry) in 1965 from the
University of Roorkee. He completed his PhD thesis in 1970 at the University of Roorkee. He worked as a Lecturer
in the University of Roorkee and Punjabi University, Patiala, and was a Research Associate with Prof. C N R Rao
at IIT Kanpur. He was Alexander Von Humboldt Fellow at the Institute of Organic Chemistry and Biochemistry,
University of Hamburg, Germany. He has contributed more than hundred research papers in various national and
international journals of repute. His main areas of research are molecular spectroscopy including normal vibrations
analysis. Physico-chemical studies of interactions of importance to biology including charge transfer and hydrogen
bonding by spectroscopic and nonspectroscopic techniques, molecular orbital calculations on complex molecules.
He has two books and two booklets to his credit.
S K Dogra
Former Professor
Department of Chemistry
Indian Institute of Technology
Kanpur, Uttar Pradesh
H S Randhawa
Former Assistant Professor
Department of Chemistry and Biochemistry
Punjab Agricultural University
Ludhiana, Punjab
Tata McGraw Hill Education Private Limited

NEW DELHI
McGraw -Hill Offices

New Delhi New York St Louis San Francisco Auckland Bogotá Caracas
Kuala Lumpur Lisbon London Madrid Mexico City Milan Montreal
San Juan Santiago Singapore Sydney Tokyo Toronto
Published by the Tata McGraw Hill Education Private Limited,
7 West Patel Nagar, New Delhi 110 008.
Copyright © 2012 by Tata McGraw Hill Education Private Limited.
No part of this publication may be reproduced or distributed in any form or by any means, electronic, mechanical,
photocopying, recording, or otherwise or stored in a database or retrieval system without the prior written
permission of the publishers. The program listings (if any) may be entered, stored and executed in a computer
system, but they may not be reproduced for publication.
This edition can be exported from India only by the publishers,
Tata McGraw Hill Education Private Limited
ISBN (13): 978-0-07-107267-0
ISBN (10): 0-07-107267-5
Vice President and Managing Director: Ajay Shukla
Head—Higher Education Publishing and Marketing: Vibha Mahajan
Publishing Manager (SEM & Tech. Ed.): Shalini Jha
Asst. Sponsoring Editor: Smruti Snigdha
Development Editor: Renu Upadhyay
Executive—Editorial Services: Sohini Mukherjee
Senior Production Manager: Satinder S Baveja
Proof Reader: Yukti Sharma
Marketing Manager—Higher Education: Vijay Sarathi
Product Specialist: Sachin Tripathi
Graphic Designer—Cover: Meenu Raghav
General Manager—Production: Rajender P Ghansela
Production Manager: Reji Kumar
Information contained in this work has been obtained by Tata McGraw-Hill, from sources believed to be
reliable. However, neither Tata McGraw-Hill nor its authors guarantee the accuracy or completeness of any
information published herein, and neither Tata McGraw-Hill nor its authors shall be responsible for any errors,
omissions, or damages arising out of use of this information. This work is published with the understanding
that TataMcGraw-Hill and its authors are supplying information but are not attempting to render engineering or
other professional services. If such services are required, the assistance of an appropriate professional should
be sought.
Typeset at Mukesh Technologies Pvt. Ltd., Puducherry.

Contents
Preface xv
Greek Alphabet xvii
1. INTRODUCTION 1
1.1 The Nature of Electromagnetic Radiations 1
1.2 Molecular Spectroscopy and Spectral Regions 2
1.3 Born–Oppenheimer Approximation 3
1.4 Schrödinger Wave Equation 3
1.5 General Condition of Resonance 3
1.6 Boltzmann Distribution Law and Population in Energy States 4
1.7 Widths and Shapes of Spectral Bands 4
1.8 Intensity of the Spectral Bands (or Quantum Mechanical Treatment of Transition
between two States) 8
1.9 Spectrum and Basic Elements of a Single-Beam and Double-Beam Absorption Spectrometer 11
1.10 Signal-to-Noise Ratio 12
1.11 Resolving Power of a Spectrometer and Its Relation to Slit Width 12
1.12 Limit of Sensitivity of the Spectroscopic Method of Identification of Substances 13
1.13 Fourier Transform (FT) and Computer Average Transient (CAT) 13
1.14 Laser 15
1.14.1 What is Stimulated Emission? 15
1.14.2 Wavelength Range and Power Output of Lasers 17
1.15 Plane-Polarised Radiation and Dichroism 19
1.16 Liquid Crystals 19
1.16.1 Classification of Liquid Crystals 20
1.16.2 Liquid Crystalline States in Living Nature 20
1.16.3 Liquid Crystals as Anisotropic Solvents 21
1.17 Fitting of a Straight Line and Principle of Least Squares 21
Problems 22
Appendix 1A: General Physical and Chemical Constants and Their Values 24
2. MICROWAVE SPECTROSCOPY 25
2.1 Introduction 25
2.2 Microwave Spectrometer 25
2.3 Moments of Inertia of Molecules 26
2.4 Diatomic Molecule as a Rigid Rotator 33
2.4.1 Spacing between two Adjacent Allowed Energy Levels 34
2.4.2 Selection Rules and TransitIon between Permissible Energy Levels 35
2.5 Diatomic Molecule as a Nonrigid Rotator 40
2.5.1 Spacing between two Adjacent Allowed Energy States 41
2.5.2 Transition between two Adjacent Energy Levels 41
2.6 Hyperfine Structure 45
2.6.1 Designation of Molecular Orbital Corresponding to the Value of Λ 46
2.7 Rotational Spectra of Polyatomic Molecules 48
2.7.1 Moments-of-inertia Defect 48
2.7.2 Rotational Spectra of Linear Polyatomic Molecules 49
2.7.3 Structural Parameters of Linear Triatomics of the Type XYZ 51
2.7.4 Rotational Spectra of Spherical-Top Molecules 53
2.7.5 Rotational Spectra of Symmetric-Top Molecules 54
2.7.6 Transition between two Adjacent Energy States (Prolate/Oblate Tops as Rigid Rotators) 54
2.7.7 Nonrigid Rotator Model of Prolate/Oblate Symmetric-Top Molecules 55
2.7.8 Rotational Spectra of Asymmetric-Top Molecules 57
vi Contents
2.8 Thermal Distribution of Population Among the Rotational Levels 58

2.9 Stark Effect in Relation to Dipole Moment Determination from Rotational Spectrum 63
2.9.1 Stark Transition between Adjacent Levels in Symmetric-Top Molecules 63
2.9.2 Stark Transition between Adjacent Levels in Linear Molecules 65
2.10 Inversion in Ammonia and Its Analogues 67
2.11 Barrier to Internal Rotation 70
2.11.1 Selection Rules for Transition between two Adjacent Rotational Energy Levels in Slightly
Asymmetric Molecules of the Type CH3OH 72
2.12 Rotational Sum over the States and Rotational Constants 74
2.12.1 Sum over States for Diatomic and Linear Polyatomic Molecules 74
2.12.2 Sum over States for Symmetric-top and Asymmetric-Top Molecules 75
2.12.3 Partition Function for Spherical-Top Rigid Molecules 77
2.12.4 Sum over States for Molecules with Internal Rotation 77
2.13 Applications of Microwave Spectroscopy 77
Problems 78
Appendix 2.1A: Energy Levels of a Diatomic Rigid Rotator 79
Appendix 2.2: Energy Levels of a Restricted Rotator 81
3. INFRARED SPECTROSCOPY 83
3.1 Introduction 83
3.2 Mechanism of IR Absorption and IR Activity 84
3.3 Instrumentation 85
3.3.1 Working of a Double-Beam Spectrometer 85
3.3.2 Sources 86
3.3.3 Monochromators 86
3.3.4 Detectors 86
3.4 Sample Preparation and Sample Cells 86
3.5 Calibration of Spectra 89
3.6 Bandwidth 89
3.7 Polarisation of Infrared Bands 89
3.8 Vibrational Spectra of Diatomic Molecules 89
3.9 Diatomic Molecule as a Harmonic Vibrator 89
3.9.1 Spacing between two Successive Vibration Levels 92
3.9.2 Selection Rules of Harmonic Vibrator 92
3.9.3 Transition between two Successive Vibrational Levels 93
3.10 Diatomic Molecule as an Anharmonic Vibrator 94
3.10.1 Spacing between two Adjacent Vibrational Energy Levels 96
3.10.2 Selection Rules 96
3.10.3 Transition between two Adjacent Vibrational Levels 96
3.10.4 Determination of Electrical Anharmonicity 98
3.10.5 Maximum Value of Vibrational Quantum Number and Energy 100
3.10.6 Bond Energy of Diatomic Molecule 101
3.10.7 Determination of Dissociation Energy of Diatomics 103
3.11 The Vibrational Partition Functions of Molecules 107
3.12 Population Distribution Among the Various Vibrational States 107
3.13 Rotation–Vibration Spectra of Diatomic Molecules 108
3.13.2 Transition between two Adjacent Rotation−Vibration States 109
3.13.3 Intensity Distribution in Rotation–Vibration Absorption Spectrum 114
3.14 Nature and Number of Vibrational Motions in Polyatomic Molecules 117
3.14.1 Fermi Resonance 119
3.14.2 Coriolis Force 120
3.15 Rotation–Vibration Spectra of Polyatomic Molecules 124
3.15.1 Structure and Characteristics of Rotation–Vibration Bands in
Polyatomic Molecules 125
3.15.2 The Band Types 126
Contents vii
3.15.3 Theoretical Aspects of PR-separation and IQ /ITotal in Different

Types of Molecules 127
3.16 Applications of IR Spectroscopy 130
3.16.1 Qualitative Applications 130
3.16.2 Quantitative Applications 143
3.16.3 Biological Applications 186
3.17 Far IR Absorption Spectroscopy 192
3.18 Near IR Absorption Spectroscopy 192
3.19 Mid IR Reflectance Spectroscopy (MIRS) 192
3.20 Near IR Reflectance Spectroscopy (NIRS) 195
3.21 Photoacoustic or Optoacoustic IR Spectroscopy (PAIS/OAIS) 195
3.22 IR Emission Spectroscopy (IMS) 197
Problems 198
Appendix 3.1A: Harmonic Vibrator 201
Appendix 3.2A: Anharmonic Vibrator 204
Appendix 3.3A: Energy Levels for the Diatomic Rotator-Vibrator or Energy
Levels of Diatomic Nonrigid Rotator-Anharmonic Vibrator 206
Appendix 3.4A: Molecular Constants of Diatomic Molecules 211
Appendix 3.5A: Reciprocal Table for the Conversion of Microns(m)
to Wavenumbers (cm−1) 211
Appendix 3.6A: Program in C++ for the Computation of Statistical a Thermodynamic
Functions from Spectroscopic Data 213
4. RAMAN SPECTROSCOPY 224

4.1 Introduction 224
4.2 Classical Theory of Raman Scattering 225
4.3 Quantum Mechanical Picture of Raman Scattering 227
4.4 Characteristic Parameters of Raman Lines 228
4.5 Equivalence of Beer–Lambert Law of Absorption in Raman Scattering 230
4.6 Second-order Raman Spectrum 231
4.7 General Selection Rule for Raman Scattering 231
4.8 Raman Spectra of Diatomic Molecules 231
4.8.1 Pure Rotational Raman Spectra of Diatomic Molecules—Selection Rules for
Pure Rotational Spectra 231
4.8.2 Pure Vibrational Raman Spectra of Diatomic Molecules 235
4.8.3 Rotational-Vibrational Raman Spectra of Diatomic Molecules 237
4.9 Vibrational Raman Spectra of Polyatomic Molecules 238
4.10 Various Forms of Raman Scattering 239
4.10.1 Electronic Raman Effect/Scattering (ERE/ERS) 239
4.10.2 Resonance Raman Scattering/Effect (RRS/RRE) 239
4.10.3 Stimulated Raman Scattering/Effect (SRS/SRE) 241
4.10.4 Spin-flip Raman Scattering/Effect (SFRS/SFRE) 241
4.10.5 Continuous Wave Stimulated Raman Gain Spectroscopy (CWSRGS) 241
4.10.6 Inverse Raman Effect/Scattering (IRE/IRS) 241
4.10.7 Hyper Raman Scattering/Effect (HRS/HRE) 241
4.10.8 Coherent Anti-stokes Raman Scattering/Effect (CARS/CARE) 242
4.10.9 Coherent Stokes Raman Scattering/Effect (CSRS/CSRE) 242
4.10.10 Higher-order Raman Spectral Excitation Scattering/Effect
(HORSES/HORSEE) 242
4.10.11 Surface-Enhanced Raman Scattering/Effect (SERS/SERE) 242
4.10.12 Time-Resolved Raman Spectroscopy (TRRS) 243
4.10.13 Raman Optical Activity (ROA) 243
4.10.14 Raman-Induced Kerr Effect SpeCtroscopy (RIKES) 243
4.11 Basic Principles of a Raman Spectrometer 243
4.12 Applications of Raman Spectroscopy 246
4.12.1 Study of Environmental Effects on Molecular Systems 247
viii Contents
4.12.2 Mechanism of Tautomerism and Polymerisation 248

4.12.3 Conformational Equilibria 248
4.12.4 Study of Ionic Equilibria 248
4.12.5 Study of Hydrogen-Bonded Equilibria 249
4.12.6 Nature of Chemical Bond 249
4.12.7 Molecular Structure 249
4.12.8 Energy Difference between Rotamers 250
4.12.9 Structure of Water 251
4.13 Biological Applications 252
Problems 255
5. ELECTRONIC SPECTROSCOPY 256

5.1.1 Multiplicity of Electronic States 258
5.1.2 Electronic Transition Energy 258
5.2 High–Resolution Ultraviolet/Visible Spectroscopy and Selection Rules for
Electronic Transitions in Light Diatomics 259
5.2.1 Electronic Spectra of Diatomic Molecules 262
5.3 Vibrational Coarse Structure of Electronic Spectra of Diatomics 263
5.3.2 Sequence and Progression 263
5.3.3 Deslandres System of Bands of a Diatomic Molecule 264
5.3.4 Intensity Distribution in Vibrational Electronic Spectra from Franck–Condon Principle 269
5.4 Rotational Fine Structure of Electronic-Vibration Structure of a Diatomic Molecule 272
5.4.2 Transition between Rotational Levels Associated with a Particular Vibrational Level in the
Electronic Spectrum of a Diatomic Molecule 272
5.4.3 The Fortrat Diagram 275
5.4.4 Simplification of the Analysis of Rotational Fine Structure in the Vibrational Electronic
Spectrum 277
5.5 Electronic Absorption Spectra of Polyatomics 278
5.6 The Isotope Effect in Molecular Electronic Spectra 279
5.6.1 Vibrational Isotope Effect 279
5.6.2 Rotational Isotope Effect 282
5.7 Nuclear Spin and Intensity Alternation in Electronic Band Structure 283
5.8 Continuous Absorption and Emission Spectra 288
5.8.1 Continuous Absorption Spectra 288
5.8.2 Continuous Emission Spectra 289
5.9 Predissociation/Diffuse Spectra 290
5.9.1 Mechanism of Predissociation in Diatomics 290
5.9.2 Predissociation in Polyatomic Molecules 291
5.10 Dissociation Energy and Its Determination 292
5.10.1 Band-Convergence Method 293
5.10.2 Predissociation Limit Method 299
5.11 Low Resolution UV Visible Spectroscopy 300
5.11.1 Introduction 300
5.11.2 Terms and Symbols Associated with UV/V Absorption Measurements of Molecules in
Solution / Vapour Phase 300
5.12 Intensity of Absorption Bands 303
5.12.1 Factors Affecting Position and Intensify of Absorption Bands 305
5.13 Symmetry Selection Rules for Polyatomic Molecules 306
5.14 MEchanism of Absorption and of Colour 307
5.15 Electron Transitions in Organic Molecules 307
5.15.1 Characteristics of Various Types of Electronic Transitions 309
5.15.2 Colour 311
5.16 Inorganic Electronic Absorption Spectroscopy 316
Contents ix
5.16.1 Energy Levels in Transition Metal Complexes 317

5.16.2 Crystal Field Stabilisation Energy 319
5.16.3 Charge Transfer Inter-Ligand Transitions in Transition Metal Complexes 321
5.16.4 Ligand-Field Stabilisation Energy 321
5.16.5 Selection Rules for Transitions between Energy States of Transition Metal Complexes 322
5.16.6 Bandwidths 323
5.17 Electronic Absorption Spectra of Transition Metal Complexes with d1, d2, d3, . . . d9 Configurations in
Octahedral and Tetrahedral Fields 323
5.18 Racah Parameters 347
5.18.1 Determination of Dq and b for Transition-Metal Ions in Octahedral and Tetrahedral
Fields 348
5.19 Basic Principles of a Double Beam UV-Visible Spectrophotometer 354
5.19.1 Spectrographs 356
5.20 Applications of Low Resolution UV-Visible Spectroscopy 357
5.20.1 Qualitative Applications 357
5.20.2 Woodward–Fieser Rules for Predicting the Wavelength of Absorption Maximum
in Conjugated Dienes 359
5.20.3 Fieser–Kuhn Rules 361
5.20.4 Woodward–Fieser–Scott Rules for Predicting the Position of Absorption Maximum
(p-p * transitions) in Enones and Dienones 365
5.20.5 Quantitative Applications 372
5.20.6 Biological Applications 396
5.21 Photoacoustic UV/ V Spectroscopy (Pauvs/Pavs) 407
5.21.1 Modern Photoacoustic Spectrophotometer 407
5.21.2 Applications of UV/V PhotoacouStic Spectroscopy 408
5.22 Optical Rotatory Dispersion (ORD) and Circular Dichroism (CD) 411
5.22.1 Origin of Absorption and Circular Dichroism 414
5.22.2 Types of Optically Active Chromophores 415
5.23 Applications of ORD and CD 416
5.24 Fluorescence and Phosphorescence 419
5.24.1 Quantum Yield or Quantum Efficiency of Photophysical Processes 421
5.24.2 Relationship between Luminescence (Fluorescence/ Phosphorescence) Intensity and
Concentration 422
5.24.3 Apparatus for Detecting Fluorescence (Fluoremeter) 424
5.24.4 Measurements of Radiative Lifetimes of Excited States from the Absorption Spectra 427
5.24.5 Measurements of Lifetimes of Excited States from Fluorescence or Phosphorescence
Intensity 427
5.24.6 Rate Constants of Fluorescence and Phosphorescence 429
5.24.7 Comparison of Absorption, Fluorescence and Phosphorescence Spectra 431
5.24.8 Factors Affecting Fluorescence and Phosphorescence 432
5.24.9 Intermolecular Deactivation (or Quenching) of Excited States 435
5.24.10 Kinetics of Intermolecular Processes 436
5.24.11 Chemical Reaction 440
5.24.12 Fluorescence Quenching 441
5.24.13 Applications of Fluorescence and Phosphorescence 444
Problems 448
6. PHOTOELECTRON SPECTROSCOPY 457

6.2 Types of Photoelectron Spectroscopy 457
6.3 Photoelectric Effect 458
6.4 Theory of Photoelectron Spectroscopy 458
6.5 Photoelectron Spectrum 463
6.6 Bandwidth and the Factors Affecting It 465
6.7 Resolution of Photoelectron Spectral Peak 466
6.8 Intensities of the Photoelectron Spectral Peaks 466
x Contents
6.8.1 Rules to Predict the Intensities of Ionisation Peaks 466

6.8.2 Measured Intensities of Photoelectron Spectral Peaks 467
6.9 Sensitivity and Detection Limit of Photoelectron Spectroscopy 469
6.10 XPS Chemical Shifts or Core Electron Shifts 470
6.11 Atomic Photoelectron Spectra 471
6.12 Rules for the Prediction of Atomic Photoelectron Spectrum 474
6.13 Photoelectron Spectrometer 474
6.14 Applications of Photoelectron Spectroscopy 479
6.14.1 Ionisation Energies 479
6.14.2 Surface or Interelectronic Structure Studies 480
6.14.3 Analysis of Gas Mixtures 485
6.14.4 Molecular Orbital Models 486
6.14.5 Anharmonicity Constant and Changes in Molecular Geometry on Ionisation 489
6.14.6 Molecular Charge Distribution 496
6.14.7 Structural Elucidation 497
6.14.8 Nature of Chemical Bonding 498
6.14.9 Substituent Effects 499
6.14.10 Nonbonded Interactions 501
6.14.11 Study of Biological Systems 504
Problems 505
7. NUCLEAR MAGNETIC RESONANCE SPECTROSCOPY 507

7.2 Characteristics of Magnetic Nuclei 507
7.3 Concept of Magnetic Energy States 508
7.4 Larmor Theorem 508
7.5 Energy of the Magnetic States 509
7.6 General Selection Rules for the Transition between Magnetic States 511
7.7 Spacing between two Consecutive Magnetic Levels 511
7.8 Transition between Magnetic Levels 511
7.9 Nuclei Population in Different States 514
7.10 Line Shape and Line Width 516
7.11 Factors Influencing the Intensity of NMR Lines 516
7.12 Measurement of NMR Signal Intensity 517
7.13 Spin-Relaxation Process 517
7.13.1 Types of Relaxation Processes 517
7.14 Width of NMR Signal and Relaxation Times 518
7.15 Fourier Transform NMR (FTNMR) Spectroscopy 519
7.16 Time Periods Involved in Multiple Pulse FT 520
7.17 Theory of Measurement of Relaxation-Rate Constants by Multiple Pulse FT 521
7.18 Measuring of S-L Relaxation Rates 522
7.19 Spin Echo 524
7.20 Chemical Shift and Shielding Constant 524
7.21 Estimation of Proton Chemical Shifts in Compounds of Known Structures 528
7. 22 Display of NMR Spectrum 530
7.23 Dipole–Dipole (D-D) InteraCtion 531
7.24 Spin–Spin (S-S) Coupling and Spin–Lattice (S-L) Coupling Constants 531
7.25 Types of S-S Coupling Constants and Factors Affecting Them 534
7.25.1 Vicinal Coupling (Jvic) 534
7.25.2 Geminal Coupling (Jgem) 536
7.26 Concept of Magnetically Equivalent Protons 537
7.27 Mechanism of S-S Interaction 538
7.28 Quantum Mechanics of S-S Interaction 541
7.29 Intensity of NMR Lines in AB Spin System 547
7.30 Second-order Effects 550
7.31 Effect of Chemical Environment on Shape of NMR Signal 551
Contents xi
7.31.1 Effect of Temperature 553

7.31.2 PMR Spectra of Cyclohexane (Operating Frequency: 100 MHz) 553
7.32 NMR Spectrometers 554
7.32.1 Basic Principle of CWNMR Spectrometer 554
7.32.2 Basic Principle of a Fourier Transform NMR (FTNMR) Spectrometer 556
7.33 Simplification of NMR Spectra 557
7.33.1 Shift Reagents 557
7.33.2 Superconducting Magnets 558
7.33.3 Double Resonance 559
7.34 Two-Dimensional Fourier Transform NMR (2DFTNMR) 561
7.35 13
C-NMR Spectroscopy 564
7.36 19
F-NMR Spectroscopy 571
7.37 31
P-NMR SpecTroscopy 572
7.38 NMR Spectra of Nuclei with I > 1/2 574
7.39 Applications of NMR Spectroscopy 575
7.39.1 Mole Fraction of a Non-ideal Solution from NMR Spectrum 575
7.39.2 Analysis of Mixtures 576
7.39.3 Study of Tautomeric Equilibria 577
7.39.4 Determination of Hydration Number 578
7.39.5 Determination of Optical Purity 578
7.39.6 Determination of Molecular Mass 579
7.39.7 Degree of Unsaturation in Natural Fats 581
7.39.8 Study of Conformational Equilibria 581
7.39.9 Study of H-bonded Equilibria 583
7.39.10 Dissociation Constant of an Acid 587
7.39.11 Structural Elucidation 588
7.39.12 Kinetic Studies 589
7.40.1 Determination of Oil Contents in Oilseeds 596
7.40.2 Dry Rubber Content in Natural Rubber Latex 597
7.40.3 Study of Fate of Inorganic Phosphates (Pi) in Developing Seeds 598
7.40.4 Analysis of Mixtures of Amino Acids and of Peptides 598
7.40.5 Optical Purity of Dipeptides and Configuration of Diketopiperazines Isolated from Natural
Sources 598
7.40.6 Protein–Drug Interactions 600
7.40.7 Nuclear Magnetic Resonance Imaging (NMRI) 601
7.40.8 Liquid Crystals in High Resolution NMR Spectroscopy 603
7.40.9 Recent Developments 606
Problems 607
8. NUCLEAR QUADRUPOLE RESONANCE SPECTROSCOPY 610

8.2 Nuclear Electric Quadrupole Moment 610
8.3 Electric Field Gradient (EFG) 611
8.4 Nuclear Electric Quadrupole Coupling 613
8.5 Origin of Nuclear Electric Quadrupole Coupling 613
8.6 Combined Magnetic and Quadrupole Interactions 613
8.6.1 Shift in Zeeman Energy Levels Due to First-order Nuclear Quadrupole Coupling 614
8.6.2 Transition between Zeeman Energy Levels Perturbed by First-order Quadrupole
Interaction 616
8.7 Quadrupole Splittings in Liquid Crystals 619
8.8 Nuclear Quadrupole Relaxation 621
8.8.1 Quadrupole Relaxation Models 622
8.9 Pure Quadrupole Resonance 625
8.10 Equivalence of Larmor’s Theorem of NMR in NQR Spectroscopy 631
8.11 NQR Spectrometer 632
xii Contents
8.12 Effects of Various Factors on Nuclear Quadrupole Resonance Frequency 633

8.12.1 Effect of Temperature and Pressure 633
8.12.2 Conjugative Effects 633
8.12.3 Inductive Effects 634
8.13 Applications of NQR Spectroscopy 635
8.13.1 Applications of NQR Through NMR 635
8.14 Applications of Pure NQR 643
8.14.1 Study of Chemically Inequivalent Quadrupole Nuclei in the Unit Cell 644
8.14.2 Nature of Chemical Bond 644
8.14.3 Structural Information from NQR Spectrum 645
8.14.4 Study of Charge-Transfer Complexes 645
8.14.5 Study of Phase Transitions 646
8.14.6 Study of Ionic Character of Chemical Bonds in the Gaseous and Solid States 646
8.14.7 Partial Double-Bond Character and Sigma Electron Population for Chlorine Atom Bonded to
an sp2 Hybridised Carbon Atom 646
Problems 648
9. ELECTRON SPIN RESONANCE SPECTROSCOPY 650
9.2 Similarities between ESR and NMR 650
9.3 Behaviour of a Free Electron in an External Magnetic Field 651
9.4 Intensity of ESR Lines and Factors Affecting it 653
9.5 ESR Line Width and Factors Affecting it 654
9.6 g-Value and Factors Affecting ESR Lines 657
9.7 Hyperfine Interaction 661
9.8 Types of Hyperfine Interactions 661
9.8.1 Contact or Fermi Hyperfine Interaction 661
9.8.2 Anisotropic Hyperfine Interaction 668
9.9 Zero-Field Splitting (Fine Structure Terms) and Kramer’s Degeneracy 669
9.10 Nuclear Quadrupole Effects in ESR Spectra of Solids 672
9.11 Rules for the Prediction of Number of Hyperfine Lines and their Relative Intensities 674
9.12 Basic Principle of an ESR Spectrometer 678
9.13 Determination of g-Value 680
9.14 Fourier Transform ESR Spectroscopy (FTESRS) 682
9.15 Electron Nuclear Double Resonance (ENDOR) 682
9.16 Electron Double Resonance (ELDOR) 683
9.17 Applications of ESR Spectroscopy 683
9.17.1 Determination of Concentration of Free Radicals 683
9.17.2 Study of Electron Transfer Reactions 684
9.17.3 Determination of Rotational Correlation Time 685
9.17.4 Study of Ion Pairs 687
9.17.5 Kinetics of Electron-Exchange Reactions 688
9.17.6 Determination of Exchange Integral 689
9.17.7 Determination of Unpaired Electron Spin Density and the Molecular
Shape of Free Radical 689
9.17.8 Absolute Dating 693
Problems 698
10. MÖSSBAUER SPECTROSCOPY 701

10.2 Pre- and Post-Mössbauer era 701
10.3 Line Position 705
10.4 Intensity of Mössbauer Line and Parameters Affecting It 705
10.4.1 Einstein Model 706
10.4.2 Debye Model 706
Contents xiii
10.5 Line Width of Mössbauer Line and Factors Affecting It 707

10.6 Hyperfine Interactions 712
10.6.1 Centre Shift 712
10.6.2 Second-order Doppler Shift (dSOD) 713
10.6.3 Nuclear Isomer Shift and Factors Affecting It 714
10.6.4 Partial Isomer Shift 722
10.7 Internal or Effective Magnetic Field 723
10.7.1 Effect of Temperature on Internal Magnetic Field 723
10.7.2 Effect of Pressure on Internal Magnetic Field 723
10.7.3 Effect of External Magnetic Field 724
10.8 Magnetic Hyperfine Interaction 724
10.8.1 Intensity of Hyperfine Lines 726
10.8.2 Sign of Nuclear Zeeman Interaction and Internal Magnetic Field 731
10.8.3 Relaxation Effects in Relation to Magnetic Hyperfine Splitting 731
10.9 Electric Quadrupole Interaction 733
10.9.1 Intensity of Nuclear Quadrupole Doublet 734
10.9.2 Magnitude and Sign of Quadrupole Interaction 735
10.10 Combined Magnetic and Quadrupole Interactions 736
10.10.1 Effect of Pressure on Quadrupole Splitting 739
10.10.2 Effect of Temperature on Quadrupole Splitting 739
10.11 Sign of Quadrupole Interaction 740
10.12 Goldanskii–Karyagin (G-K) > (G-K) Effect (Lattice Vibration Anisotropy) 741
10.13 Partial Quadrupole Splitting (PQS) 743
10.14 Mössbauer Spectrometer 744
10.14.1 Mössbauer Spectrum 750
10.14.2 Data Computation 755
10.15 Applications of Mössbauer Spectroscopy 755
10.15.1 Lifetime of the Excited State 756
10.15.2 Determination of Quadrupole Moment Ratio Qe/Qg 756
10.15.3 Determination of g-Factors and Nuclear Magnetic Dipole Moments 758
10.15.4 Determination of Internal Conversion Coefficient 759
10.15.5 Determination of d R/R or Nuclear Radius Changes from Mössbauer Isomer Shift 760
10.15.6 Determination of Gravitational Red Shift 762
10.15.7 Diffusion in Solids and Liquids 762
10.15.8 Study of Surfaces 763
10.15.9 Structure of Coordination Compounds 767
10.15.10 Spin-State Equilibria 772
10.15.11 Structure of Organometallic Compounds 772
10.15.12 Electronic Structure and Bond Nature of Charge Transfer Complexes 773
10.15.13 Studies of Lunar Soil 775
10.16.1 Studies of Haemoglobin and Its Derivations 779
10.16.2 Studies of Peroxidase and Its Derivatives 783
10.16.3 Studies of Iron–Sulphur Proteins 783
10.16.4 Studies of Cytochrome−C 786
10.16.5 Ferritin 787
Problems 789
11. MASS SPECTROMETRY 790

11.2 Comparison of Mass Spectrometry with other Spectroscopic Techniques 791
11.3 Various Types of Ions Encountered in Mass Spectrometry 792
11.4 Basic Principles of Mass Spectrometry 796
11.4.1 Theory 796
11.4.2 Basic Principle of a Simple Mass Spectrometer 797
11.5 Various Forms of Mass Spectrometry on the Basis of Methods of Separation of Ions 802
xiv Contents
11.5.1 Magnetic Field Deflection Mass Spectrometry 802

11.5.2 Ion, Resonant/Cyclotron Resonance, Mass Spectrometry (IRMS/ICRMS) 804
11.5.3 Time of Flight Mass Spectrometry (TOFMS) 806
11.5.4 Radio Frequency Mass Spectrometry (RMS) 807
11.5.5 Quadrupole Mass Spectrometry (QPMS) 808
11.6 Various Forms of Mass Spectrometry on the Basis of Ionisation Processes other than Eims 810
11.6.1 Chemical Ionisation Mass Spectrometry (CIMS) 810
11.6.2 Desorption Ionisation Mass Spectrometry (DIMS) 815
11.7 Isotope Ratio Mass Spectrometry 822
11.8 Tandem Mass Spectrometry (Tandem MS) 822
11.9 GLC Mass Spectrometry 824
11.10 Merits and Demerits of Various Types of Ionisation—Mass Spectrometry 827
11.11 Intensities of the Signals in the Mass Spectrum 828
11.12 Intensity of the Parent Peak and Factors Affecting It 828
11.13 Rules for the Mode of Fragmentation of Molecular Ions 829
11.14 Interpretation of Mass Spectra of Unknown Compounds 844
11.15 Computers in Mass Spectrometry 867
11.16 Applications of Mass Spectrometry 867
11.16.1 Nonbiological Applications of Mass Spectrometry 868
11.16.2 Biological Applications of Mass Spectrometry 912
11.17 Negative-Ion Mass Spectrometry 919
Problems 934
Suggested Readings 939

Nobel Prizes 943
Preface
Overview of the subject

As academicians with years of experience, we feel there are two types of students: those who learn how to learn
and those who learn how to think. Hence, we must design books in such a fashion that both categories of students
benefit by them and are able to read and understand on their own without undue reliance on formal classroom
instructions. Students must play an active role in the learning process and the teacher a passive one, serving only
as a guide. Merely reading about experiences and observations of others is not stimulating, instead a hands-on
interactive experience not only enriches a learner’s knowledge base but also helps him/her grasp and remember
concepts more easily.
Molecular spectroscopy plays a major role in understanding the various physical and chemical phenomena
in-vivo and in-vitro. Spectroscopy/Spectrometry is often used in physical and analytical chemistry for the iden-
tification of substances through the spectrum emitted from or absorbed by them. An application-based subject,
it finds use in Astronomy, Remote-sensing, Telescopes, Microwave, NMR and MRI, Nuclear Medicine, X-Ray,
Optic Fibers, Mineralogy, and Structure Determination of Biological Compounds.
The new developments and improvements in precision of spectroscopic measurements laid the basis for new
applications of molecular spectroscopy in chemistry, chemical engineering, biology, biochemistry, space chemis-
try and other areas. In these fields, complex molecules or mixtures are accessible to molecular specific investiga-
tion under difficult experimental conditions as well as in environmental and industrial-process analysis. If students
are to understand and enjoy molecular spectroscopy, they must have some of these experiences in reality.
Objective
The objective of this book is to present molecular spectroscopy in an interesting, understandable and enjoyable
manner. An appreciation, rather than only a working knowledge of molecular spectroscopy, is what this book aims
to impart to students. Therefore, we have laid strong emphasis on the aspects or techniques. There are clear expla-
nations of the physical meaning of the mathematics used, thus enabling students to appreciate the importance of
mathematics, the queen of all sciences, in the development of molecular spectroscopy. There is stress on practi-
cal applications throughout the book. The introduction of each new concept is followed by problems. It is our
earnest hope that these examples will enable students master a concept or technique before they move on to the
next. The physical concepts are amply illustrated by means of diagrams and graphs wherever possible. The text
in discussion is the comprehensive and in-depth treatise encompassing all aspects of the subject and coordinated
with needs of various universities. The beauty of the book is the way it starts from fundamentals of the subject as
required by a beginner and goes further to detailed explanations and latest advancements in the field.
Salient Features
Briefly, the highlights of the book are given below:
• Comprehensive in-depth explanation to all important topics: Microwave Spectroscopy, Infrared
Spectroscopy, Raman spectroscopy, Electronic Spectroscopy, Photoelectron Spectroscopy, NMR
• Appendices on other advanced forms of spectroscopy: Nuclear Quadrapole Spectroscopy, Electron Spin
Resonance Spectroscopy, Mossbauer Spectroscopy
• Selected multidisciplinary spectroscopy applications given at chapter end as per relevance: will be
beneficial for students of Life Sciences such as B. Pharma, B.Sc. Biotechnology, B.Tech. Biotechnology, M.Sc.
Biotechnology and M.Sc. Chemistry
• Pedagogy
� Over 400 Figures
� 150 Tables to explain concepts
� Solved Problems interspersed throughout the chapters
� Chapter-end Review Questions
xvi Preface
Acknowledgements
We are highly indebted to our teachers, both at undergraduate and postgraduate levels, who motivated us to join
this fascinating field of chemistry. This fascination was further accentuated during our stay with teachers with
whom we spent our time as research associates at different universities or institutes.
One of us (HSR), who is responsible for the basic draft of this book, wishes to express his deep sense of grati-
tude to his colleagues, especially professors B S Sekhon, R S Dhillon, Joginder Singh and G S Saroa, whose
encouragement and cooperation have been a source of inspiration. Professor B K Puri (IIT Delhi) also deserves
a special note of thanks for his valuable criticism and suggestions. Mr Jaswant Singh Dhillon carried out most of
the art and drawing work, and we are very thankful to him. We are also deeply indebted to the authors of the books
and research papers, which we consulted while preparing the manuscript for this book. We thank all the following
reviewers for taking out time to go through the manuscript and give important suggestions:
Sankar Chakravarty
Indian Association for the Cultivation of Science (IACS), Jadavpur, Kolkata
Satyen Saha
Banaras Hindu University (BHU), Varanasi, Uttar Pradesh
V K Saxena
Dr H S Gour University, Sagar, Madhya Pradesh
Saptarshi Mukherjee
Indian Institution of Science, Education and Research (IISER), Bhopal, Madhya Pradesh
Aloke Das
Indian Institution of Science, Education and Research (IISER), Pune, Maharashtra
Ms Hardeep Kaur developed the software for use in Infrared and Mass Spectrometry, and both she and Mr Harcharan
Kanwal (children of HSR) helped us in checking the solutions of the numerical problems and making helpful sug-
gestions. We are extremely thankful to them. At this juncture, we would also like to thank the Tata McGraw-Hill
team of Vibha Mahajan, Shalini Jha, Smruti Snigdha, Renu Upadhyay, Sohini Mukherjee, Satinder Singh and
Yukti Sharma for their help and cooperation at each stage of the project. Finally, we are grateful to our respective
wives [Dr (Mrs) Sulekha Dogra and late Mrs Mohinder Kaur] whose patience, dedication and encouragement
made this book possible.
Suggestions for the improvement of this book would be most welcome.
S K Dogra
H S Randhawa
Publisher’s Note
Remember to write to us. We look forward to receiving your feedback, comments and ideas to enhance the quality
of this book. You can reach us at [email protected]. Please mention the title and author’s
name as the subject.
Greek Alphabet
Letter Name
Α α Alpha
Β β Beta
Γ γ Gamma
Δ δ Delta
Ε ε Epsilon
Ζ ζ Zeta
Η η Eta
Θ θ Theta
Ι ι Iota
Κ κ Kappa
Λ λ Lambda
Μ μ Mu
Ν ν Nu
Ξ ξ Xi
Ο ο Omnicron
Π π Pi
Ρ ρ Rho
Σ σ Sigma
Τ τ Tau
Υ υ Upsilon
Φ φ Phi
Χ χ Chi
Ψ ψ Psi
Ω ω Omega
CHAPTER 1 INTRODUCTION
A teacher should produce less heat and more light.

—Unknown
1.1 THE NATURE OF ELECTROMAGNETIC RADIATIONS

Electric and magnetic fields of radiation oscillate sinusoidally in mutually perpendicular planes at right angles to
the direction of propagation of radiation as shown in Fig. 1.1. The electric (E) and magnetic (H) field components
of an electromagnetic wave in the X-direction are
⎛x ⎞
E = E0 sin 2p ⎜ vt ⎟
⎝λ ⎠
⎛x ⎞
H = H0 sin 2p ⎜ vt ⎟ (1.1)
⎝λ ⎠
Here, v and l represent the frequency and wavelength of the radiation; c is the velocity of light and is given by
c = v l.
Direction of Wave Motion
E H E H
H E H E
z
Fig. 1.1 Representation of propagation of an electromagnetic wave.
(a) Wavelength (l) The distance between one particle in a wave and the corresponding next particle in a
wave with which it is in phase is known as wavelength, i.e. the distance between two similar points on a wave is
called wavelength. As we know, there is a maximum and minimum in a wave, so in practice, the distance between
two minima or between two maxima is also a wavelength. The pictorial view of a wave is shown in Fig. 1.2. The
distance between particles a and b or a′ and b′ is the wavelength. The greatest distance a wave moves up and down
is called the amplitude, A.
a l b
A A
A A
a⬘ λ b⬘
Fig. 1.2 Representation of an electromagnetic wave.
2 Molecular Spectroscopy
Units Metre (m), centimetre (cm), millimetre (mm), micrometer (μm), nanometre (nm), Angstrom (Å); 1 μm =
10−4 cm = 10−3 mm, 1 nm = 10−3 μm, 1 Å = 10−8 cm = 0.1 nm.
(b) Wave Number The reciprocal of wavelength is called wave number.
Units Number of waves per centimetre (cm−1), Kayser (K), kilokayser (kK), 1 kK = 1000 K = 1000 cm−1.
(c) Frequency (n) It is a measure of the number of waves passing through a given point in a unit time. It is inter
linked with wave number through the following equation.
c
v= = cv (1.2)
λ
Units Cycles per second (cps, C/S) or Hertz (Hz), kilocycles per second (kcps or kHz), megacycles per second
(Mcps or MHz); l MHz = 103 kHz = 106 Hz.
(d) Energy Energy in the form of electromagnetic radiations comes in discrete units called photons. Each pho-
ton contains an amount of energy determined by the frequency of electromagnetic radiation. The German scientist
Max Planck first proposed the equation relating to frequency and energy,
c
E = hn = h = hc v (1.3)
λ
where c is the velocity of light (2.988 × l010 cm/s) and h is the Planck’s constant with the value 6.6259 × 10−27 erg s.
According to Einstein, hv = mph c2. Here, mph is the mass of the photon (~10−33 g). Pressure of light and bent light
rays in the strong gravitational field are evidences for the mass of the photon.
Units The units and the conversion factors of energy units are given in Table 1.1, while the values of general
physical constants and chemical constants are listed in Appendix 1A.
Table 1.1 Conversion factors of energy units.
Units cm-1 Ergs/molecule Cal/mole eV
1 cm−1 1 1.9865 × 1016 2.859 1.2398

1 erg/molecule 5.0348 × 1015 1 1.43956 × 1016 6.2421 × 1011
1 cal/mole 0.34975 6.946 × 10−17 1 4.3361 × 10−5
1 eV 8065.8 1.6020 × 10−12 2.306 × 104 1
1.2 MOLECULAR SPECTROSCOPY AND SPECTRAL REGIONS

The study of electromagnetic radiation with matter in all of its forms, viz., solids, liquids/solutions and gases is
called spectroscopy. The radiations are characterised by their frequencies, wave numbers, wavelengths and the
energy associated with the electromagnetic waves. Depending upon the energy of radiations but not the intensity
(energy per unit time per unit area), which is a measure of the amount of radiations; spectroscopy has been clas-
sified into different branches, summarised in Table 1.2.
Table 1.2 Various branches of spectroscopy and the molecular phenomenon associated with them.
Branch Phenomenon Wavelength Information

g-rays Nuclear transitions 0.3–0.003Å Oxidation states
x-rays Electronic transitions (inner electrons) 100–0.3 Å Core electron energies
Visible/UV Electronic transitions (outer electrons) 1μm–300 Å Electron configuration
Infrared Molecular vibration, molecular rotation 300–1μm Stiffness of bonds
Microwaves, far infrared Electron spin, molecular rotation 30–0.3 m Bond lengths and bond angles
Nuclear magnetic resonance, nuclear Electronic structure near the
Radio waves 300–3 m
quadrupole resonance nucleus
Introduction 3
1.3 BORN–OPPENHEIMER APPROXIMATION

The nucleus of an atom is much heavier as compared to its electrons and, therefore, the motion of the two can be
considered independent of each other, e.g. the mass of the hydrogen nucleus is 1840 times than that of an electron.
The total energy (E) of the system is thus the sum of energies due to electronic (Eelec ) and nuclear (Enuc ) motions.
E = Enuc + Eelec (1.4)
where
Enuc = Etrans + Evib + Erot + Eintrot (1.5)
Here, Etrans, Evib, Erot, and Eintrot stand for translational, vibrational, rotational and internal rotational energies,
respectively. In general, we can say that each degree of freedom can be treated independently. In the present
studies, Etrans is omitted since it is not quantised.
1.4 SCHRÖDINGER WAVE EQUATION

This is a second-degree differential equation and is the most appropriate method of providing us the solution of
one-body problems. The time-independent forms of this equation in one and’ three dimensions are given by the
expressions (1.6) and (1.7), respectively.
d 2ψ 8π 2 m
+ 2 ( E U )ψ = 0 (1.6)
dx 2 h
8π 2 m
∇2ψ + ( E − U )ψ = 0 (1.7)
h2
where y is the wave function which describes the wave property of a particle with mass m, ∇2 is the Laplacian
operator and is expressed as follows:
∂2 ∂2 ∂2
Cartesian coordinates: ∇2 = + +
∂x 2 ∂y 2 ∂z 2
1 ∂ ⎛ 2 ∂⎞ 1 ∂ ⎛ ∂⎞ 1 ∂2
Polar coordinates: ∇2 = ⎜ r ⎟ + ⎜ sin θ ⎟ + ;
r 2 ∂r ⎝ ∂r ⎠ r 2 sin θ ∂θ ⎝ ∂θ ⎠ r 2 sin 2 θ ∂ϕ 2
where x = r sin θ cos j, y = r sin θ sin j, z = r cos θ; dt = r2 sin θ dr dθ dj.
Rearranging Eq. (1.7), we get
⎛ − h2 2 ⎞
⎜⎝ 8π 2 m ∇ + U ⎟⎠ ψ = Eψ (1.8)
Hy = Ey (1.9)
−h 2
where H = ∇2 + U is known as the Hamiltonian operator.
8π m
2
A wave function y is associated with a state of physical system corresponding to energy E, i.e., it represents
the energy state of the system.
The energy of various types of motions, viz., electronic, vibrational, rotational, etc., can be determined with the
help of Schrödinger wave equation and it is found that
Eelec > Evib > Erot
1.5 GENERAL CONDITION OF RESONANCE

Resonance phenomenon may have both useful and harmful effects. For instance, a company of soldiers was
marching in step across a bridge in St Petersburg and the bridge collapsed. Investigation showed that the period
of free oscillation of the bridge coincided with that of an ordinary marching step. When any atomic or molecular
system is exposed to electromagnetic radiation, the absorption or emission (in emission, the sample itself serves
as a source and emission is complementary to absorption) of a photon with frequency n will occur only if the
energy difference ΔE between the ground and excited states of the system matches the energy of the photon, i.e.,
± ΔE = E2 − E1= hv (1.10)
This is the resonance condition, and the pictorial view of it is shown in Fig. 1.3.
E2
E2
ΔE = hn ΔE = hn
hn hn
E1 E1
(a) (b)
Fig. 1.3 Sketch of (a) absorption, and (b) emission phenomena under resonance conditions,
O—molecule in the ground state, O—molecule in the excited state.
1.6 BOLTZMANN DISTRIBUTION LAW AND POPULATION

IN ENERGY STATES
The population in any energy state is governed by the Boltzmann distribution law, i.e., at equilibrium,
x 0 e − Ex / T
(1.11)
where Nx is the number of particles in a given state x with energy Ex. N0 is the total number of particles in the
zeroth energy level, T is the absolute temperature, k is the Boltzmann constant: k = 1.38044 × 10−16 erg degree−1;
N k = R, the gas constant. The degeneracy of a state increases the population of a particular energy state.
If the number of molecules in the state with energy E2 is N2, and the number in the states with energy E1 are N1
then the Boltzmann distribution law states that at equilibrium
−E / T
N1 e 1
= = e +ΔEE / T
(1.12)
N 2 e − E2 / T
The population difference between the two states may be deduced from the population ratio expression (1.12)
as follows:
N1 N 2 e ΔEE / T − 1
=
N1 N 2 e ΔEE / T + 1
x2 ΔE
Since N1 = N2 = N; ex = 1+ x + + … and << 1
2! κT
N ΔEE
Therefore, Δn = N1 − N2 = (1.13)
2κ T
Thus, the population difference Δn relates directly to the total population, the energy difference between the two
states and inversely to the absolute temperature.
1.7 WIDTHS AND SHAPES OF SPECTRAL BANDS

Theoretically, the transition, due to absorption of photons, occurs between two exactly defined energy levels
under the resonance condition, i.e. ±ΔE = E2 − E1 = ± hv. Therefore, the absorption bands should have no width
at all [Fig. 1.4(a)]. However, the observations are contrary to this hypothesis. Even with very high-resolving-
power spectrometers, it is not possible to get a widthless band since there is a minimum width, called the natural
bandwidth, inherent in any molecular or atomic transition. This width arises because energy levels of molecular
systems are not precisely determined except in the ground level. According to Heisenberg’s uncertainty principle,
if a system exists in any energy state for a time of Δt seconds, the energy of that state will be uncertain to the extent
h
ΔE where ΔE × Δt ≥ ≈ 10−34 Js,
2π
where h is Planck’s constant. The system will remain in the ground state for an infinite time, i.e., t = ∞ and
ΔE = 0, since the ground energy state of a system is well defined. However, the excited state has a finite lifetime.
Therefore, we cannot, represent the excited level by a sharp line which represents only one value of energy, i.e.,
E2 . Hence, the transition between the excited and ground levels will have an uncertainty in energy ΔE. In other
words, we can say that the energy spread ΔE of the excited state manifests itself in the form of a band in the
Introduction 5
transition between the two states. The width of the band (which actually is a contour of a bunch of lines) is called
natural bandwidth and is defined as the width of the band at half its height [Fig. 1.4(b)].
Δn1/2
(a) (b)
Fig. 1.4 Representation of an absorption band having (a) no width, and
(b) full width Δν 1 at half maximum.
2
In general, according to the uncertainty principle, the width of a band from one excited state to another is given by
h ⎡ 1 1 ⎤
Γ= ⎢ + ⎥, (1.14)
2π ⎣⎢ τ ex1 τ ex2 ⎥⎦
where τ ex1 , and τ ex2 are the lifetimes of the respective levels. If one of the states is well defined, i.e., τ ex1 = ∞, the
natural width Γn of the excited state is given by the general expression
h
ΔE = Γn = (1.15)
2π τ ex
The natural bandwidth of the absorption band or the source emission may also be expressed in terms of
half-life (t1/2 ) of the excited state. We know that t1/2 = tex In 2.
0.693 h
Therefore, ΔE = Γn = ⋅ (1.16)
t1/ 2 2π
The transition times for electronic, vibrational and rotational transitions are 10−15, 10−13 and 10−12 second, respec-
tively. Therefore, the order of natural bandwidths for the respective bands will follow the order: electronic >
vibrational > rotational. In general, the natural width of spectral bands is very small and is usually masked by
other effects which contribute to the broadening of spectral bands. Among these, the most important are pressure
or collision broadening effect and Doppler effects.
(a) Pressure Broadening Molecules of gases and liquids always collide with each other. As a result of such
collisions, the energy levels of the molecules get perturbed which result in broadening of the band. The mechanism
for pressure broadening may be explained as follows: The molecules/atoms emitting or absorbing radiation in gases
undergo collisions. In each collision, there is a certain probability that an atom/molecule in an excited state will
make a radiationless transition to the ground or lower state so that the lifetime of the excited state will be decreased.
If the number of collisions/second removing atoms from the excited state is Nc then the total number of transitions
out of the excited state are
⎛ 1⎞
⎜⎝ N +
c
τ ex ⎟⎠
where tex is the lifetime of the excited state. The total observed width of the band will then be
h ⎛ 1⎞
Γ = Γn + Γp = N + (1.17)
2π ⎜⎝ τ ex ⎟⎠
c
where tp is the contribution to bandwidth from pressure effect and is given by

1 8 RT
Γ= (1.17a)
π l0 M
Here l0 is the mean free path, M is the molecular mass and T is the absolute temperature.
Since the number of collisions/second depends upon the pressure of the gas/liquid, this effect is called pressure
broadening. In order to observe experimentally the natural line width, it is essential to reduce the pressure in the
spectral source. Thus, by changing the pressure and observing the corresponding change in bandwidth, one can
obtain information about the collisions occurring in a gas or a liquid. Solid-phase spectra are often sharp due to
a limited number of molecular collisions. However, the splitting of spectral bands into components is evidence
for molecular collisions in solids.
(b) Doppler Broadening This effect arises due to the relative motion of molecules of gases or liquids with
respect to the photons. Thus, depending upon the direction of motion of the molecules, with respect to photons of
frequency v and velocity (v), the observed frequency (vobs) is shifted either towards the lower or higher frequency.
The frequency shift Δv is given by
Δv v − vobs v
= =± (1.18)
v v c
The signs (+) and (−) indicate that the molecule is moving away from and towards the photons respectively.
Since the molecular systems under investigation have a large number of molecules moving with different
velocities relative to the photons, the broadening of bands results. The contribution of Doppler effect to line
broadening is given by the expression
2vobs 2 RT vobs 2 RT (1.18a)
ΓD = = 1.665
c M c M
1 2 RT
or in terms of wavelength Γ D = 1.665 (1.18b)
λ obs M
where the terms and symbols have their usual meanings. The Doppler broadening is least in high atomic mass
atoms, so 198Hg should be an ideal light source for sharp spectral lines. It is to be kept in mind that while colli-
sion broadening predominates in liquids compared to that in gases, Doppler effect often determines the natural
bandwidth in gases.
From the foregoing discussion, it is evident that the net effect of all of these phenomena results in imparting
a finite width to the spectral bands. There are two types of curves encountered in spectroscopy: ‘Lorentzian line
shape’ and ‘Gaussian line shape’. The main difference between the two types of line shapes is that the outer
wings of the Lorentzian lines are much longer and drop more slowly than those of Gaussian lines. The two
types of normal spectral curves and the first and second derivatives of the respective curves have been shown in
Figs 1.5(a) and (b) respectively. In practice, we obtain neither pure Lorentzian nor pure Gaussian type curves but
an admixture of these curves. The Lorentzian and Gaussian curves may be fitted into the mathematical expression
given by
p − p 2 ( x − a )2
IG = e (1.19)
π
a 1
IL = (1.20)
π a + x2
2
Normal Absorption Curve
First Derivative Curve
Second Derivative Curve
(a) (b)
Fig. 1.5 Types of absorption line shapes: (a) Lorentzian, and (b) Gaussian.
Introduction 7
1 Gaussian Line
Here, p = , and the constants a and b stand for standard
2b 2 Slope A
and mean deviations respectively.
These two types of curves can be identified by the following Slope B
Slope A
two methods: (i) the slope of the derivative curve shown in =
a 2.2
=
Slope B b
Fig. 1.6, and (ii) the normalisation plot presented in Fig. 1.7.
In this method, the baseline is divided into units of m, where
m is the point on the abscissa where the curve reaches a maxi- a b
mum and the ordinates are normalised to unity. For broad
absorption, it is preferred to record the absorption signal as the
Lorentzian Line
derivative curve. The latter has two advantages: (i) The point
of maximum absorption is difficult to measure accurately with Slope A
a broad absorption curve but is shown with great precision as Slope B
the intersection- of two lines in the derivative curve. (ii) Slope Slope A a 4
= =
is zero in the derivative curve corresponding to the maximum Slope B b
in the absorption curve.
The number of peaks which appear as shoulders that never a b
pass through maximum in an absorption curve correspond to
the number of maxima or minima in the derivative curve. The
maxima or minima do not cross the abscissa since slope in the
derivative curve corresponding to the shoulder in the absorp-
tion curve is nonzero (Fig. 1.8), and (ii) the intensity of deriva-
tive signal can be estimated more accurately as compared to Fig. 1.6 The slope ratio method for identification of
the corresponding broad absorption. Gaussian and Lorentzian derivative curves.
1.0
0.8
0.6
Lorentzian
0.4
0.2
a 2a 3a 4a
5a 4a 3a 2a a
0.2
0.4
0.6
Gaussian
0.8
1.0
Fig. 1.7 The normalization method for identification of Lorentzian and Gaussian derivative curves.
(a)
(b)
Fig. 1.8 Comparison of (a) absorption, and (b) derivative curves.

1.8 INTENSITY OF THE SPECTRAL BANDS (OR QUANTUM MECHANICAL

TREATMENT OF TRANSITION BETWEEN TWO STATES)
The intensity of spectral signal depends upon the transition probability, population of the atoms or molecules in
the state from which transition is occurring, and the concentration as well as the path length of the substance
through which the electromagnetic radiations are passing. The transition from a lower to a higher state takes place
according to certain rules, deduced quantum-mechanically and experimentally and are called selection rules.
While the transition probability for allowed or permis-
sible transitions is nonzero, it is zero for forbidden transi- E2, N2
tions. Now let us consider that transition takes place from an
initial state, with energy El and population N1, to an excited
B12 B21 A21
state, with energy E2 and population N2 under the influence
of electromagnetic radiation of frequency v having density
E1, N1
r(v) as shown in Fig. 1.9. According to the Boltzmann dis- hn
tribution law (Eq. 1.11), at equilibrium, the population ratio
Fig. 1.9 Transition between two states of a molecule.
in the two states may be expressed as
N1
= e hv12 / κ T (1.21)
N2
The number of molecules making a transition per unit time from the state 1 to 2, i.e., intensity of radiation due
to induced absorption
= B12 N1 r (v12) (1.22)
Here, B12 is called the Einstein transition probability of induced absorption, units of B12 are (J–1m3 s–1 number–1)
(number s–1) = J–1m3s–2 = kg–1m and that of r(v12) are Jm–3 s.
The number of molecules making a transition per unit time from the state 2 to 1, i.e., the intensity of radiation due
to induced emission
= B21 N2 r (v12) (1.23)
Here, B21 is known as the Einstein transition probability of induced emission which is due to the perturba-
tion caused by the field of electromagnetic radiation. In addition, the molecules in the state 2 may emit radiation
spontaneously such that ΔE = E2 – E1 = hn. The spontaneous emission occurs in the absence of an external field.
Thus, the number of molecules making spontaneous emission per unit time, i.e., the intensity of radiation due to,
spontaneous emission is
= A21 N2 (1.24)
A21 is called the Einstein coefficient of spontaneous emission.
At equilibrium, the rate of absorption is equal to the rate of emission, i.e.
B12 N1 r (v12) = B21 N2 r (v12) + A21 N2 (1.25)
We know that B12 = B21; then rearranging Eq. (1.25) we get
N 2 A21
ρ( ) = (1.26)
B21 ( 1 2)
Combining Eq. (1.21) and Eq. (1.26) we obtain
A21
ρ( ) = hv / κ T
(1.27)
B21 ( )
According to Planck’s law of radiation,
8π hv
h 3 1
ρ( ) = 3 hv / κ T
(1.28)
c ( )
Dividing Eq. (1.28) by Eq. (1.27), we obtain
A21 8π hv21 3
= (1.29)
B21 c3
Thus, with the aid of Eq. (1.29) and having a knowledge of B21, we can compute A2l and vice versa. These
quantities may be determined both experimentally and theoretically. A21 can be determined directly from the
Fermi–Dirac theory of radiation. The following relationship exists between the areas of the absorption band and
the corresponding probability of absorption, Bl2.
∞
1 c
N hv12 ∫0 v
B12 = A dv (1.30)
Introduction 9
Thus, the probability of quantum transition may be determined from the area of the absorption band, i.e.
∞
∫ A dv
0
v
which is determined from the dependence of the absorption coefficient, Av of the appropriate band on the
frequency, n. The area of the absorption curve may be measured most accurately by the paperweight method.
If N 20 is the number of molecules in the excited state at t = 0 and N 2t is the number at any time t then
− dN 2t
= A21 N 2t
dt
− A21 t
or N 2t N 20 e (1.31)
After a certain time
1 N0
τ= , N 2t = 2 (1.32)
A221 e
For allowed transitions, t ≈ 10−8 s; while for some excited states, t is of the order of 10–3 s, and are called
metastable states.
The spontaneous emission is always in all the 4p, solid angles, and thus its contribution in the direction of
induced emission will be very small and is, therefore, neglected. Mathematically, when the solid angle (dw) in the
direction of observation is very small, then dw/4p will be very small, i.e., when dw→ 0, dw/4p→0.
Since only the induced emission is coherent with the incident beam of radiation, the net absorption intensity
is given by
Ia = N1 rB12−N2 rB2l
= rB2l (N1–N2) = rB2l Δn (1.33)
That is, the intensity of radiation depends upon the population difference between the states with energies E1 and
E2, provided r is constant and transition between these states is permissible according to the selection rule, i.e.,
transition probability B21 = B12 ≠ 0.
Since by Eq. (1.13),
N ΔE
Δn =
2κ T
consequently, Eq. (1.33) reduces to,
N ΔEE
I a ρ B21 (1.34)
2κ T
i.e., the net intensity of the absorption line depends on the total population in the ground state, the energy difference
between the two states, absolute temperature and population difference between the two states at equilibrium, pro-
vided r is kept constant.
Now, if transition is occurring as per selection rules from two initial states with different populations to some
other state then the intensity of the transition from initial state with greater population will be higher as compared
to the state with a smaller population.
The net intensity of absorption is related to the concentration C of the sample and the path length l by the
Beer–Lambert law represented by
I
T = t = e − εCCl (1.35)
I0
I
A = ln 0 = εC Cl (1.36)
It
where It is the intensity of the transmitted radiation, I0 is the intensity of incident radiation, A is the absorbance, C
is the concentration of the absorbing species, l is path length, and e is the molar extinction or molar absorbance or
molar decadic extinction coefficient which is different for different transitions. e depends upon transition proba-
bility—the larger the value of e, the larger will be the transition probability. This law is strictly valid for radiation
of a single wavelength, i.e. for monochromatic radiation. Lambert’s law ignores the concentrations. A common
student version of the Beer–Lambert law is
“The taller the glass the thicker the brew, the less light that gets through.”
The terms and symbols for use in the Beer–Lambert law are given in Table 1.3. It should be noted that Beer’s
law does not hold good in all cases. The deviations from Beer’s law are accounted for by the different degrees
of association, solvation and dissociation of molecules in solutions of different concentrations. Therefore, in
Table 1.3 Terms and symbols used in Beer–Lambert law.
Current Symbol Definition Current Name Obsolete Symbol Obsolete Name

It
T Transmittance − Transmission
I0
I0
A ln Absorbance D, E Optical density, Extinction
It
A
a Absorptivity k Extinction coefficient, Absorbancy index
bC
E AM Molar absorptivity aM Molar (molecular) coefficient, Molar
bC absorbancy index
b − Path length l, d −
M is the molecular mass of absorbing species, C is the concentration in g/litre−1, l is the path length in cm. E 1%
1cm A / bC ′ where C ′ is in per
cent by weight and b = 1 cm. Note that when the Beer–Lambert law is written in terms of natural logarithm instead of the base ten logarithm,
the molar extension coefficient (e) is called the Napierian absorption coefficient (ee), and the absorption (A) is called the Napierian absorption
(Ae). ee = 2.3026e.
measuring absorption spectra in solutions of different concentrations, it is necessary to check whether conformity
of Beer’s law is observed, e.g. the double increase of concentration must be equivalent to the doubled thickness of
the absorbing layer at the same concentration.
Problem 1.1: (a) Prove that photometric accuracy is the maximum at 37 per cent transmittance and remains
within tolerable limit over a range of transmittances about 15 to 65 per cent (absorbances of 0.8 to 0.2).
(b) Plot a hypothetical curve between per cent transmittance and log C. Does this graph give any indication of the
relative precision at various levels of absorbance?
Solution (a) Even for a system which obeys Beer’s law, the concentration range over which photometric analy-
ses are useful is limited at both high and low values. Therefore, at both high (little radiant energy is transmitted)
and low (large amount of radiant energy is transmitted) concentrations of absorbing material, the uncertainty in the
measurement of A or T exceeds the permissible experimental limits. For the greatest accuracy in the measurement
of absorbance A, the increment ΔA, which corresponds to the intensity change ΔIt , must be as small a fraction as
possible of the actual absorbance A, i.e. the quantity ΔA/A should be minimum. This is proved as follows:
According to Beer’s law,
It = e − A I 0
I0
A = log = log I0 − log It (1.37)
It
Differentiating with respect to It
1
dA = − (log e)
dI [I0 is constant]
It t
Dividing both sides by A and substituting the value of log e (= 0.4343), we obtain,
dA 0.4343 1
= dI
A A It t
Since e− I0
A
It
dA 0.4343 1
Therefore, =− ⋅ dI t
A A I 010 − A
Replacing differentials by finite increments
ΔA 0.4343 ΔI t ⎛ 1 ⎞
=− ⎜⎝ ⎟
A I0 A × 10 − A ⎠
Differentiating with respect to A (Δ It is constant)
d ( A/A) .4343 ⎛ 10 A ln10 10 A ⎞
=− − 2⎟
dA I 0 ⎜⎝ A A ⎠
Introduction 11
ΔA
Δ d ( A/A)
For to be minimum, =0
A dA
This means
10 A ln10 10 A
= 2
A A
1
from which A= = 0.4343
ln10
This means that the optimum value for the abundance is 0.4343 which corresponds to transmittance T = 36.8
per cent. The relative error in an analysis resulting from 1 percent error in the photometric measurements for
varying transmittance and absorbance is shown in Fig. 1.10, which reveals that although the error is least at
37 per cent T, it will not be much greater over a range of transmittances about 15 to 65 per cent (absorbances
0.8 to 0.2).
(b) The conventional method of plotting a calibration graph for spectrophotometric analyses is either the
exponential curve or the straight line. The latter has the advantage of showing the region over which Beer’s law
is obeyed, but it fails to give any indication of the relative precision at various levels of absorbance. However, a
plot between per cent transmittance and log C shown in Fig. 1.11 gives some additional features. If a sufficient
range of concentrations are covered, an S-shaped curve called Ringbom curve always results. If the system does
not deviate from Beer’s law, the point of inflection occurs at 37 percent transmittance, otherwise, the inflection is
at some other value but the general shape of the curve is the same. The curve generally has a considerable region
which is nearly straight. From the straight-line portion, the optimum range of concentrations can be selected
for the particular spectrophotometric analyses. The precision of analyses can be estimated from the slope of the
curve. The steeper the curve, the more sensitive the test is.
It can be proved by a differentiation procedure that if the absolute photometric error is 1 percent, the percent
relative error in the analysis is given by 230/ms, where ms is the slope taken as the transmittance change (read from
the vertical axis) in percent corresponding to a tenfold change in concentration.
The commercial spectrometers record the percentage of light transmitted by the absorbing species and is given
by the expression
It
Per cent transmittance = 100 (1.38)
I0
20
Percent Transmittance
10
40
Percent Relative Error
6 60
4
80
2 36.8
100
1 4 10 40 100
0 20 40 60 80 100 Log C
Percent Transmission
Fig. 1.11 Percent transmittance as a function of
Fig. 1.10 Relative error as a function of transmittance. logarithm of concentration.
1.9 SPECTRUM AND BASIC ELEMENTS OF A SINGLE-BEAM AND

DOUBLE-BEAM ABSORPTION SPECTROMETER
We know that when a beam of radiation falls on a sample, it may induce, depending upon its energy, electronic,
vibrational or rotational transitions between the ground and excited states of the sample under resonance condi-
tion, i.e., ± ΔE = hv. The radiation falling on the sample may undergo absorption, transmission, scattering and
reflection. If I0 is the intensity of the incident radiation, then
I0 = Ia + It + Is + Ir (l.39)
where Ia Ip, Is and Ir are the intensities of the respective phenomena. At the glass– 100% 0
air interface, Ir is very small (0.2−0.5%) and hence can be neglected. Thus, Eq.
(1.39) may be written as
Transmittance
I0 = Ia + It + Is (1.40)
Absorbance
The intensity of the absorbed radiation is measured as a function of
energy with the help of an instrument called absorption spectrometer.
The curve between absorbance versus wavelength is termed the absorption spec-
trum. In general, we can say that the spectrum, which is a characteristic of any
electromagnetic radiation, is a graph on which the intensity is plotted in the ver-
0 100%
tical direction and the energy or wave number in the horizontal direction. The
hypothetical spectrum for a single transition is shown in Fig. 1.12. ν (cm−1)
The width of a beam of monochromatic radiation is controlled with the aid of Fig. 1.12 Hypothetical spectrum
a slit element of the spectrometer. This beam is made to fall on a sample where corresponding to a single transition.
it gets partially absorbed and causes transitions in the sample. The transmitted part of radiation is then detected
by the detector and analysed in an analyser. In a double-beam spectrometer, the monochromatic beam is split
into two components, namely, sample and reference beam, by an additional element known as splitter, and are
passed through the sample and the reference. The transmitted radiation from the sample and the reference is then
detected and analysed separately. In practice, in a double-beam spectrometer, the intensity of transmitted radia-
tion I0 through reference, but not the intensity of radiation I from the source, is taken as the intensity of incident
radiation. On the other hand, in a single-beam spectrometer, since no reference is involved, I automatically takes
the status of I0. The advantage of a double-beam over a single beam spectrometer is that the spectral signals due
to reference and source get automatically eliminated. The fundamental elements of single-beam and double-beam
absorption spectrometers are shown in Figs 1.13 (a) and (b) respectively.
Source Slit Detector Analyzer
Sample
I0 It
(a)
Source Slit Splitter Sample Detector Analyzer

I It
I I0
Reference
(b)
Fig. 1.13 Arrangement to observe absorption spectrum with
(a) single, and (b) double-beam spectrometer.
Emission spectrum, which is only recorded in special studies, is generally ignored due to the following reasons.
Firstly, since every sample to be studied has to be made an emission source for all the samples, emission studies
are not simple and economical. Secondly, during emission itself (especially in the case of g - and x-rays) there
could be radiation damage that may change the structure of the sample.
1.10 SIGNAL-TO-NOISE RATIO

In modem spectrometers, the signal produced by a detector is amplified by electronic amplifiers. Resulting random
fluctuations due to electronic signals, called noise, appear in the spectrum. Therefore, the real peak in the spectra
will easily be detected if its intensity is large in comparison to noise. The sensitivity of detection of the actual peak
is maximum when signal-to-noise ratio is of the order of 3–4.
1.11 RESOLVING POWER OF A SPECTROMETER AND ITS RELATION

TO SLIT WIDTH
The efficiency of a spectrometer in separating the bands in a spectra is called its resolving power. According to
Lord Rayleigh’s criterion, two wavelengths differing by Δl are said to be resolved when the central maximum of
one coincides with the first minimum of the other. Resolving power is then defined as
Introduction 13
Δλ
λ v
Rp = = (1.41)
Δλ Δv
where l or v is the average of two wavelengths or wave numbers, and Δl
or Δ ν is the smallest difference between two wavelengths or wave num-
bers that can be measured on a spectrometer (Fig. 1.14).
λ
As such, the quantity l/Δl is called the theoretical resolving power.
Fig. 1.14 Sketch showing the resolving
We know that molecular absorption does not take place at a single wave- power of a spectrometer.
length only but occurs over a range of wavelengths that depends upon the
width of the entry slits. This parameter, which reduces the theoretical resolving power and width of slits, is called
the finite or mechanical slit width, while the small region of energy isolated by the exit slits is called the spectral
slit width. The distribution of energy as a function of wavelength emitted from the exit slit is known as slit function.
Thus, resolution of an instrument depends upon the spectral slit width and is proportional to the mechanical slit
width. Hence, narrower slits result in better resolution. Further, the energy received by the detector is proportional
to the square of the mechanical slit width. Thus, narrowing of this width to improve resolution has a negative effect
on the efficiency of the instrument, i.e. narrow slit width allows less energy from the beam to reach the detector,
and as a result, the strength of a signal will decrease. At a certain value of the slit width, the signal-to-noise ratio
becomes less than 3–4. Thus, it is fixed at such a value where signal-to-noise ratio is maintained.
Since large slit widths broaden the band, so in any quantitative work the slit width should be less than the width
of the band under consideration. In high-resolution spectroscopy, which requires narrowing the slit, the pen output
response is slower since the detector is hard worked as less energy is received by it. Hence, fast scanning under
these conditions leads to a considerably erroneous spectrum. Deviations from Lambert–Beer’s law are observed
when the ratio of slit width to bandwidth is greater than an order of 0.4. The deviations are the greatest both at the
largest slit widths and where the extinction coefficients have approached a maximum. When slit width is equal to
half the value of the bandwidth, the difference between the true and observed maximum extinction coefficient is
found to be of the order of 20 per cent.
1.12 LIMIT OF SENSITIVITY OF THE SPECTROSCOPIC METHOD

OF IDENTIFICATION OF SUBSTANCES
The limit of sensitivity of the spectroscopic method of identification of substances is defined as
I −I I
δ= 0 t = a (1.42)
I0 I0
For diatomic molecules, it is approximated by
1 λ 03
δ= ⋅ A21
2 R p ln (1.43)
40 π c
where l0 is the wavelength of absorbed radiation. A21 is the probability of transition (emission), Rp is the resolving
power of the instrument, l is the path length, n is the number of gas molecules in cm−3, i.e. the gas concentration.
Let d = 0.05, i.e., absorption is 5 percent, l0 = 400 nm (boundary between ultraviolet and visible regions of
radiation), A = 108 s−1 and l = 1 cm and Rp = 104 (the theoretical resolving power of an average spectral device),
n = 1013 molecules cm−3 or p = 10−4 mmHg (at 300 K). At high temperatures and also in the case of polyatomic
molecules, which have large number of rotational and vibrational degrees of freedom, only a fraction of n mol-
ecules absorb the radiation of wavelength l0. In such cases, the limit of sensitivity rarely exceeds corresponding to
partial pressures of the order of 10−1−10−3 mmHg. Further, since A is small in the infrared region, the sensitivity in
the IR region is very small as compared to the electronic absorption spectrum of the molecules. Due to this very
reason, the infrared absorption spectral method is used for the identification of substances in condensed states,
i.e. the liquid and solid states. The limit of sensitivity of the spectral method of identification based on emission
is higher as compared to that of absorption.
1.13 FOURIER TRANSFORM (FT) AND COMPUTER AVERAGE

TRANSIENT (CAT)
The method of Fourier transform spectroscopy is nearly about one hundred and fifty years old and Michelson and
Rubens already applied it in principle. Some historical papers are: H Fizeau, Ann. Chim. Phys. (3) 66, 429 (1862),
A A Michelson, Phil. Mag. (5) 31, 256 (1891) and (5) 34, 280, (1892), H Rubens and R W Wood, Phil. Mag. 21,
249 (1911). In the last five decades or so, the technique though initially applied by chemists, for optical investiga-
tions, especially in the far-infrared region has now made inroads in other fields of spectroscopy also.
The spectroscopic data, say intensity is invariably collected in the time domain, i.e. the data are collected as
a function of time and stored. In other words, the Free Induction Decay (FID) signal is stored. Since the spec-
troscopists are interested in the frequency domain, the time domain is converted to the frequency domain by the
mathematical tool known as Fourier Transform (FT). FT relates the time domain data f (t) with the frequency
domain data f (w) by the expression +∞
1
∫ f (t ) e dt
+ iwt
f (w) = (1.44)
2π −∞
This is called continuous transform since the limits of integration extend from −∞ to +∞. When the integral
limits extend over a finite time, the discrete FT is defined as
1 +t
f (w) = ∑ f (t )e + iwt dt
2π − t
(1.45)
The inverse FT is expressed as
1 +w
f (t ) = ∑
2π − w
f ( w )e iwt dw (1.46)
A strong pulse (each pulse is actually a packet of radiation) is applied to the sample and the resultant spectral
response pattern is stored in a small computer. The spectrum, i.e., position and width of the frequency package
can be extracted from this information by taking the inverse FT with the help of a large digital computer. FT is
equally applicable to emission and absorption spectroscopy and FT spectra can successfully be recorded in the
NMR, ESR, UV and IR regions. The FT Raman spectrum can also be recorded.
FT of a time-domain decaying exponential and sinusoidal or cosinusoidal oscillation is a Lorentzian line as
shown in the Figs 1.15(a) and (b) respectively. In the latter case, the line shifts from the central zero frequency.
This shift is equal to the frequency of the oscillation of the sinusoidal or cosinusoidal. Shown in Fig. 1.16 is the
FT of time-domain signal of summed sine waves into frequency-domain spectrum.
The beat frequency is independent while the rate of decay of the overall time-domain signal is dependent on
the width of original spectral peak. The larger the width, the larger the decay, provided the spectral peak is at the
same central frequency. However, two spectral lines having the same width but at different central frequencies
in the frequency domain will have different beat frequencies in the time domain. The beat frequency in the time,
domain signal increases with the increase in the frequency of line of the spectral peak.
The advantage of FT spectroscopy over conventional spectroscopy is its speed. This is due to the fact that the
whole FT spectrum is recorded instantaneously and simultaneously while in the latter case, each point in the
spectrum is to be recorded as a function of frequency. So, slit-width adjustment and focusing are no problems in
the FT spectrum. The computing and plotting time is of the order of 10 seconds which is much shorter compared
to 10–15 minutes required by the conventional method to obtain similar resolutions. The other advantage is that
resolving power stays constant over the entire spectrum.
A time-averaging computer or Computer Average Transient (CAT) is used as an accessory with commercial spec-
trometers in order to enhance the weak signal. The use of CAT in conjunction with FT is useful for obtaining better
signal-to-noise (S/N) ratio. The narrow region of the frequency-domain spectrum in which the signal is expected is
FT
100 0 100
(a)
FT
Time FT
Frequency
100 0 100
Frequency
(b)
Fig. 1.15 FT of time domain signal of (a) a decaying exponential, Time
and (b) sinusoidal or cosinusoidal oscillation. Fig. 1.16 FT of time-domain signal of summed sine waves.
Introduction 15
scanned several times. The information obtained in each scan is fed to a small computer and the total information
is automatically stored in the main computer. After n summed scans, the signal will be n times enhanced in the
store, whereas the noise will accumulate less rapidly due to its random nature. It can easily be shown that n scans
will enhance the noise level in the store by n . Hence, the net gain in S/N is n/ n n . In other words, we can
say that an increase in the intensity of the signal is proportional to the square root of the number of scans.
In brief, we can say that in comparing conventional spectroscopy methods and explaining their principles, we
are concerned mainly with the problem of how to separate electromagnetic radiation into its spectral elements and
to determine the wavelength and frequency as a function of their intensity simultaneously. The problem of separa-
tion into spectral elements becomes trivial if a tunable source emitting monochromatic radiation, i.e. laser is used.
While with the techniques of optical FTS, the electromagnetic radiation is not separated into spectral elements,
rather a two-beam interferometer is used and the interference properties of the electromagnetic radiation as a
function of the path difference between the waves are studied; the results of the study, (i.e. the plot between
intensity of the constructive interference of waves and path difference (s) called interferogram) are then converted
mathematically to the spectrum (i.e. I versus ν ) on a computer. This conversion is a Fourier transform, which is
why the technique is called Fourier transform spectroscopy. This method employs mathematics, computers and
electronic data, perhaps new tools, and is hard to digest especially for organic spectroscopists!
1.14 LASER
Laser stands for ‘light amplification by stimulated emission of radiation’. Sometimes, lasers are also called ‘opti-
cal masers’. The word maser stands for ‘microwave amplification by stimulated emission of radiation’. However,
light is not the correct word to be used, because lasers also produce beams in the infrared, ultraviolet, g -ray and
x-ray regions of the spectrum. The use of lasers in science and technology is due to their three important proper-
ties: (i) laser radiation is extremely energetic, (ii) laser beams are highly monochromatic and possess an extremely
narrow line width, and (iii) laser radiation is highly directional.
In 1964, C H Townes of the Columbia University (USA) shared the Nobel prize with the Soviet physicists
A M Prokhorov and N G Basov, of the Lebedev Physical Institute of the USSR Academy of Sciences, in the field
of quantum electronics and specifically in the invention of the maser and the laser. The maser was invented prior
to the laser.
1.14.1 What is Stimulated Emission?

Consider a box with perfectly reflecting walls containing atoms of a gas and photons. Due to heat energy, the
atoms collide with each other and with the walls of the box, continuously interchanging energy which puts some
atoms in a higher energy level above the ground level. Einstein in 1919 postulated that in this ‘radiation field’, a
photon may interact with an atom and give rise to the following three distinct phenomena, namely (i) absorption,
(ii) spontaneous emission, and (iii) stimulated emission of photons.
If a photon strikes an excited atom which has a photon absorbed in it, it would stimulate the atom to emit its
absorbed photon, provided the energy of the absorbed photon is equal to that of the impinging photon. This pro-
cess of emission of radiation is called stimulated emission of radiation.
The essential requirement to get more stimulated emission is to have a larger population of excited atoms (not
less than five or so) in the system, so that a photon may have more encounters with the excited atoms, rather than
with those in the ground state. It is also essential that the atoms should stay in the excited state long enough, so
that interactions between photons and excited atoms may take place before the excited atom spontaneously emits
the absorbed photon. The process of maintaining maximum atoms in the excited state is called population inver-
sion and is achieved by (a) electrical pumping, (b) optical pumping, and (c) chemical pumping.
(a) Electrical Pumping It consists of exciting the gaseous atoms by electron impact and is achieved by passing
an electric discharge through the system.
(b) Optical Pumping It is also possible to produce a larger population in the excited state by an absorption
mechanism in which atoms in the ground state absorb photons of energy corresponding to the energy difference
between the ground state and the excited state of the atom. This is called optical pumping.
(c) Chemical Pumping It is based on breaking of chemical bonds. The lasers which are made to work by chemi-
cal reactions are called chemical lasers. The chemical reactions useful for laser activity are exothermic reactions.
Now, when the photon strikes an excited atom which R

has a photon absorbed in it, it would stimulate the atom 2
to emit its absorbed energy, provided the energy of the A1 + A2
A2 1
A1
absorbed photon is equal to that of the impinging photon.
The photons thus stripped may be in phase and constitute
1
a monochromatic beam of radiation with very large ampli-
tude. The emission of monochromatic radiation by two 2
photons (say ‘1’ with amplitude A1 and ‘2’ with amplitude (a)
R
A2; A2 >A1) in and out of phase is shown in Fig. 1.17. The

2
monochromatic radiation may be continuous or it may be
in the form of a pulse. The direction of the laser beam is A1 1 R
defined by optical resonant cavity which in its simplest A2 − A1

version is a pair of spherical mirrors (totally reflecting and A2
2 1
output mirrors) set on an optic axis. The active material is (b)
placed in between these mirrors. Solid active materials are
Fig. 1.17 Superimposition of two waves: (a) in phase, and
often in the shape of a cylinder whose axis is aligned with (b) out of phase.
the axis of the optic resonator, the length of the cylinder
being about ten times its diameters.
The dimensionless quality factor Q of the resonant cavity is expressed as
Q=2pnt (1.47)
Here, the parameter 1/t defines the rate of decay of the field energy in the passive resonant cavity. The smaller
the losses in the resonator, the smaller l/t and consequently, the higher the resonator’s Q. The quality factor Q is
used to show the ability of the cavity to store energy and is related to the loss factor a having the dimension of
reverse length by the expression.
1 λα
= (1.48)
Q 2π
If the causes of loss are many, say, absorption and radiation through the side surface, we may associate with
each type of loss its own a-factor, ai, and its own Q-factor, Qi so that
1 λα i
= (1.49)
Qi 2π
If these losses are independent of each other then
1 1
a = ∑ai and =∑ (1.50)
Q Qi
The loss factor due to radiant losses, a rad has the form
( r)
α rad = ln (1.51)
2L
where L is the resonator length and r is the reflectivity (reflection factor) of the output mirror (the ratio of the light
flux reflected by the mirror to the incident flux).
Diffraction losses increase as the mirror diameter ‘a’ decreases and the wavelength and resonator length
increase. In order that a resonator be low in diffraction losses at wavelength l, it should be so long that
a2
>> 1 (1.52)
4λ L
The dimensionless parameter on the left-hand side is called the Fresnel number
a2
NF = (1.53)
4λ L
This number is not the only criterion used in estimating diffraction losses. Other features of resonator geometry
should also be taken into account. These are determined by resonator length and mirror radii of curvature as laser
resonators normally use spherical rather than flat mirrors. So, two resonators of the same Fresnel number may
incur different diffraction losses due to different geometry.
Assume for generality that the mirrors have different apertures (a1 and a2) and different radii of curvature [rc(1)
and rc(2)]. Such a resonator is defined by three principal parameters:
NF = a1a2/4lL
G1 = (a1/a2) [1−L/rc (1)] (1.54)
G2 = (a2/a1) [1−L/rc (2)]
Introduction 17
Two resonators will have equal diffraction losses, if they are identical in these parameters. Such resonators are
called equivalent resonators.
The relationship for the line width of the laser at half maximum power in terms of the ‘angular frequency’ is
given below:
c 2 (1 − r ) 2 1 N2
Δ= ⋅ ⋅ hw (1.55)
L2
P N 2 − N1 0
where c is the velocity of light.
From the above equation, one can easily see that for a laser radiation of angular frequency w0:
(a) line width is inversely proportional to the square of the length of the resonant tube, i.e.
1
Δ∝
L2
(b) line width is also proportional to the reflectivity r, and larger values of r will give narrower line width;
(c) line width is inversely proportional to the power output, i.e.
1
Δ∝
P
(d) line width is also proportional to the number of atoms in the excited state.
If we substitute r = 0.98, L = 100 cm, N2/(N2− N1 ) = 2, P = 100 mW (l W = 1 Js−l) and the value of w0 for
6328 Å line of the He−Ne laser, we end up with a value of Δ which is far less than a wave number. Experimental
value of Δ, however, is found to be of the order of 0.05 cm−1.
1.14.2 Wavelength Range and Power Output of Lasers

The wavelength range for coherent radiation emission of gas lasers extends from about 2358 Å to 120.08 μm.
Table 1.4 gives the wavelengths of commercial lasers.
The present power output limits of the commercial gas lasers are He−Ne = approx. 125 mW; Ar = 10 W;
Kr = 5 W; Xe = 1 W; CO2 ~1000 W. Commercially available ruby and neodymium solid-state lasers provide pulses
ranging from fractions of a joule to 12 kJ. The lifetime of a pulse varies from a microsecond to a nanosecond.
The CO2 laser produces a beam that can be focused on an area of 0.001 cm2 with an intensity of 1000 kW/cm2
in a continuous operation and one million kW/cm2 in pulsed operation with the pulse time equal to one nanosec-
ond. The N2 in CO2 laser does not produce any emission but it transfers energy from the electron to the CO2 mol-
ecules and lifts it to an upper laser level as shown in Fig. 1.18. A molecule of CO2 gas has a lifetime in the upper
laser level (2) equal to about 0.001 second. The molecule has a good chance of encountering a photon of suitable
energy that will force it down to a lower level (1). He in CO2 laser helps in dissipating the thermal energy.
Table 1.4 Wavelengths of some commercial lasers.

Gas lasers
Gas Å
Ar 4880
Ar 5145
Kr 5208
Xe 5413
He−Ne 5940
Kr 6271
He−Ne 6328
Kr 6871
CO2 33910
CO2 10600
Solid-state lasers
Solid −
Ruby 3472
Ruby 6943
Neodymium 10600 (1012 W)
Pulse = 0.1 nanosecond
Note: l W = 1 Js−l
The lasing range covered by various dyes runs from CO2

Energy Transfer
0.3 to 1.3 μm. Rhodamine 6G is a practically important
lasing dye of the xanthene group. N2
V′′, J′′
The wavelength of photodissociation laser, i.e. iodine
10.6 mm Laser
laser (CF3I/C3F7I + hv (= 0.3 μm) → I* + CF3/C3 F7) is
1.315 μm. It is to be noted that after lasing, the iodine
molecules join up with the molecular residue, i.e. C3F7 + Electrical
I → C3F7I. The mechanism of lasing action of iodine laser pumping
is shown in Fig. 1.19.
Energy
Excimer or excited state dimer lasers are rare gas
(
dimer Arr2* Kr )
Kr2* Xe*2 , a rare-gas oxide (ArO*, KrO*,
XeO*) or a rare-gas atom in combination with a halide V′, J′
atom (ArF*, KrF*, XeCl*, ArCl*). The lasing for ArCl*
*
is at l = 175 nm while for Arr2 is at l = 126 nm. Fig. 1.18 Sketch showing the mechanism of carbon dioxide laser.
The available pulse plasma lasers are Mg− Ca, and Sr−He. The Sr−He system lases in the violet portion of the
spectrum at l = 0.416 and 0.431 μm.
The examples of semiconductor lasers pumped by electron beams are GaAs, CdS, CdSe, etc., CdS lases at
l = 0.49 μm; GaAs doped with tellurium (for degenerate p region) and zinc (for degenerate n region) called
injection laser (p−n junction) oscillates in the wavelength range from 0.82−0.9 μm in the infrared. Bidirectional
semiconductor lasers, have also been designed, i.e., a single experimental light source at two widely different
wavelengths depending on the direction of electrical current flowing through it. The direction is changed by
switching between negative and positive voltages applied across the device.
A ruby pulse laser lases at l = 6943 Å. The rubies used in lasers are synthetic gemstones. They are made by
fusing aluminium and chromium oxide to produce large crystals. The amount of chromium in a synthetic ruby
is about 0.05 per cent. The laser action depends upon the amount of chromium present. The ruby crystal is about
0.64 cm in diameter and 3.80−5.10 cm in length. The mechanism of a ruby laser is shown in Fig. 1.20.
4
F2
Iodine
4
F1 Radiationless
Dissociation
Transition
2
Energy
Pumping by 1.3-mm laster

1 2E Metastable
Absorption hν1 hν2 State
Photons from Xe 6943 Å Laser
Recombination Flash Lamp, 560 nm
4A
2
0
Fig. 1.19 Sketch showing the mechanism of iodine laser. Fig. 1.20 Sketch showing the mechanism of a ruby laser.
Chromium ions are particularly responsive to light having a wavelength of 5600 Å. A flash from a Xe lamp
(rich in yellow-green region) excites chromium ions from their ground state 4A2 to the excited states 4F2 and 4F1.
By radiationless transition, these states decay to the metastable state, 2E, and cause the crystal to heat up. The ions
remain in the metastable state for several milliseconds from where they begin to emit photons spontaneously as
they fall to the ground state 4A2 giving an emission line at 6943 Å. While the chromium ions are trying to get back
to ground level, the flash lamp keeps on radiating more chromium ions and reproduces a population inversion.
The photons at l = 6943 Å now start interacting with the excited ions in the inversion state 2E and produce the
lasing action. Laser output is of shorter duration varying from about 1–2 ms.
The continuous wave neodymium doped with yttrium aluminum garnet (Nd: YAG) laser emits at l = 1.0615 and
1.0642 μm in the infrared. YAG is doped with chromium ions in addition to Nd3+ lases at l = 0.43 and 0.59 μm.
In glass lasers, the active material is glass doped by ions of rare-earth elements, specifically by neodymium.
The Nd glass pulse laser emits at l = 1.06 μm.
Free-electron lasers which can be tuned even in the infrared and ultraviolet regions have also been produced.
Stimulated spin-flip transitions result in coherent, tunable Raman scattering and a source of laser-type radiation
in the far infrared. Tuning is achieved by a change of the magnetic field. Such a type of laser is called Spin-Flip
Raman Laser (SFRL) or Magneto Raman Laser (MRL).
Lasers with extremely sharp colours can now be constructed and with the frequency-comb technique, precise
readings can be made of light of all colours. It is now possible to measure frequencies with an accuracy of up to
fifteen digits.
Introduction 19
1.15 PLANE-POLARISED RADIATION AND DICHROISM

Optic
When unpolarised radiation is passed through Axis Plane of Vibration (Vertical)
specifically cut crystals of tourmaline or quartz
(called polarisers), the radiation splits up into
two linearly or plane-polarised components.
One component has displacements or vibra-
tions, i.e. electric vector parallel to the plane of Unpolarised
the paper, and is termed as parallel component Radiation
(p-component), while the other has electric
vector perpendicular to the plane of the paper Plane of Polarisation (Horizontal)
↔
and is called perpendicular component (s-com- Fig. 1.21 Polarisation of radiation; -electric vector parallel in the plane of
ponent). The intensity of these two components the paper; and •-electric vector perpendicular to the plane of the paper.
is the same while the wavelength and refractive indices are different. This process of production of linearly
polarised radiation components, i.e. s and the p, from unpolarised radiation, is called dichroism. When the optic
axis of the polariser is parallel to the electric vector in the plane of the paper of unpolarised light ( ) then only
↔
the p-component passes through the polariser while the s-component (•) gets eliminated. The vibration of the
p-component occurs in a vertical plane called the plane of vibration, while the horizontal plane is termed as the
plane of polarisation. The production of such a plane-polarised radiation is shown in Fig. 1.21.
Let us now examine the behaviour of plane polarised radiation in different natured media under different con-
ditions. The electric field vector E of linearly polarised radiation can be imagined to have right-(ER) and left-(EL)
hand components. In an isotropic medium such as water or glass, the right- and left-hand components will be
equal and in phase, i.e. they have the same speed of rotation and amplitude. In other words, we can say that the
indices of refraction and absorptivities of the medium corresponding to these two components are equal, i.e.,
nL = nR and aL = aR. Consequently, the direction of plane polarised radiation in such media remains fixed.
However, when polarised radiation is passed through an anisotropic medium such as quartz or sugar solu-
tion, the radiation will be resolved into two components with same amplitude and a phase difference of p/2. The
indices of refraction of the medium corresponding to two components will be different, i.e. nL ≠ nR, while the
absorptivities will be equal, i.e. aL = aR. Radiation having this property is said to be circularly polarised. In case
the amplitudes and hence absorptivities as well as the indices of refraction of the medium are not equal, i.e. nL ≠ nR
and aL ≠ aR, the radiation with such a property is said to be elliptically polarised. The quantity (nL – nR ) is called
birefringence, and (aL–aR), the circular dichroism of the medium. By right-and left-handed rotation, we mean
that the plane of polarisation rotates in the clockwise and anticlockwise directions respectively with respect to an
observer looking towards the incoming beam of the radiation.
Further, any medium that exhibits circular birefringence may also exhibit circular dichroism. This phenomenon
of combination of these two effects in the region in which the optically absorption bands are observed is termed
the Cotton effect. When a beam of plane polarised radiation is passed through an optically inactive substance which
is placed in a magnetic field, the emerging radiation will be rotated through a certain angle provided the direction
of magnetic field and the incident beam are parallel to each other. This phenomenon is known as Faraday effect
or Magneto-Optical Rotatory Dispersion (MORD). In addition to this, there is also Magneto-Circular Dichroism
(MCD). On the other hand, when an optically isotropic medium is placed in an external electric field E, the medium
becomes doubly refracting, i.e. the indices of refraction parallel ( ) and perpendicular (n⊥) to the direction of the
field are not equal ( n ≠ n⊥ ) and the phenomenon is known as Kerr electric-optic effect or simply, Kerr effect.
1.16 LIQUID CRYSTALS

The term ‘liquid crystals’ was first coined by Lehmann in 1890. Liquid crystals behave like ordinary isotropic liq-
uids and exhibit isotropic properties similar to those of solid crystals. In other words, we can say that though these
compounds have a crystalline arrangement of molecules, yet they flow like a liquid. Due to their intermediatory
nature, they are also called mesophases or mesomorphic phases.
About 3 per cent of all organic compounds will form the liquid crystalline phase when heated above their melt-
ing points, and will appear as turbid, more or less viscous liquids. Such liquid crystals are called thermotropic.
At higher temperature, the transition to other mesophases may occur in some cases, while others exhibit only one
mesophase. The turbidity suddenly disappears and a clear liquid is formed at some higher temperature, which
is characteristic of the system. The phase transition is reversible and of the first-order type, with a latent heat of
100 cal mole−1. Mesophases are also formed in some suitable solvents and are called lyotropic, e.g., long-chain
fatty acids in aqueous solutions. They play an important role in biological systems.
1.16.1 Classification of Liquid Crystals

Depending on the arrangement of the molecules in the mesophases, liquid crystals have been classified as
(a) nematic (thread), (b) semectic (soap), and (c) cholesteric.
(a) Nematic Liquid Crystals They exhibit one-dimensional arrangement in which the long molecular axes lie
approximately parallel to each other. The molecules move, almost unhindered, as in a normal liquid. The degree
of orientation in their arrangement is described by an order parameter with values lying between zero (liquid) and
one (crystal). The simple structure of compounds forming nematic liquid crystals is
R X R′
where R and R′ are the short chains or small groups. —X—, —N = N—, —N = NO—, —CH = CH—,
-C ≡ C-, —CH = N(O)—, —CH = N—, O—CO—, .p-azoxyanisole (PAA) is the best nematic liquid crystal. Its
nematic range is 118−136°C. The nematic range of MBBA (i.e. 4-methoxy-4′-n-butyl-benzylideneaniline nem-
atogen) is 21–48°C while that of EBBA (i.e. 4-ethoxy-4′-n-butylbenzylideneaniline) is 37–80°C. The nematic
range of MBBA can be extended by mixing it with EBBA.
(b) Semectic Liquid Crystals They display two-dimensional structure in which the molecules are arranged in
layers. The centres of gravity of the elongated molecules are arranged in equidistant planes and their arrangement
may be random or regular. Depending on the orientation of these molecules in their layers, they are designated as
A, B, C,... The long axes of the molecules are parallel to the preferred direction which may be normal to the planes
in semectic A or tilted by a certain angle in semectic C. Molecular ordering of seven types have been observed in
semectic liquid crystals on the basis of x-ray diffraction studies. The various semectic types can be distinguished
by their textures in a polarising microscope. p-( p′-ethoxybenzylideneamino)-ethyl cinnamate is a typi-
cal compound forming semectic phases. The structure of the compound is
C2H5O CH N CH CH COOC2H5
Phase Transition
Phase Transition:
81°C 118.5°C 156.5°C 118.5°C
Solid Semectic B Semectic A Nematic Isotropic
(c) Cholesteric Liquid Crystals Like nem-

atogens, they also display a one-dimensional
arrangement in which the long molecular axes lie
approximately parallel to each other. However, in
this case, the orientation alters in a regular manner Solid Liquid Crystal Isotropic Liquid
from location to location—e.g. many cholesterol es- Heating
Cooling
ters form this type of liquid crystals (cholesterol it- (a)
self exhibits no mesophase). A necessary condition L(Z)
for the formation of cholesteric mesophase is that L L
the molecules are chiralic. The pictorial view of Z
different mesophases is shown in Fig. 1.22.
1.16.2 Liquid Crystalline States

in Living Nature
This state is a precondition for life itself. Biological
membranes are the most important liquid Nematic Semectic Cholesteric
(b) (c) (d)
crystalline states in living nature. They consist of
phospholipid bilayers in which the ‘hydrophobic’ Fig. 1.22 (a) Molecular arrangement in the various condensed states
of long-chain molecules, (b) nematic orientation without periodicity,
fatty acid chains are oriented inwards and the (c) semectic orientation and arrangement in equispaced planes, but no
‘hydrophillic’ head groups outwards. The bilayers periodicity within planes, and (d) cholesteric-twisted orientation with
act as molecular barriers segregating various periodicity preferred direction/structural symmetry axis—optic axis.
Introduction 21
organelles (areas) of the cell from each other. The specific vital transport of materials through the bilayers is
regulated by proteins embedded in the phospholipid membrane. It is assumed that important biological processes
such as photosynthesis, the respiratory chain and sensory stimulation occurring in the membranes are due to
specific interactions between the phospholipids and proteins. Thus, the function of biological membranes at the
molecular level can be understood provided we have a thorough knowledge of liquid crystalline properties.
1.16.3 Liquid Crystals as Anisotropic Solvents

The mesomorphic guest molecules may be incorporated into liquid crystals up to fairly high concentration without
destruction of the distant order prevailing in the liquid crystalline matrix. The liquid crystalline solutions can be
oriented by electric, magnetic, or mechanical forces, i.e. bulk samples of highly oriented solute molecules can
be easily prepared. The molecular properties of the guest are not altered appreciably by the weak intermolecular
forces. This forms the basis for the applicability of liquid crystals as anisotropic solvents for spectroscopic
investigations of the anisotropic molecular properties. Organic thermotropic nematogens are most widely
applicable solvents. Some typical liquid-crystal solvents are recorded in Table 1.5.
Table 1.5 Typical liquid-crystal solvents.
Compound Structure Mesomorphic Applications

Range( °C)
N-(p-methoxybenzildene)-p-
CH3O – C6H5 – CH = N – C6H5 – C4H9 nem 20-42: NMR, ESR, IR
butylaniline (MBBA)
p-azoxyanisole CH3O – C6H4 – N(O) = N – C6H4 – OCH3 nem: 117-135 ESR
Equimolar mixture of N-(p-methoxy
CH3O – C6H4 – N(O) = N – C6H4 – C4H9 NMR
nitrosobenzene)-p-butylaniline
N-(p-butylnitroso benzene)-p-methoxy
CH3O – C6H4 – N = (O)N – C6H4 – C4H9 nem: -5-75 NMR, ESR
aniline
N-(p-ethoxy nitroso benzene)-p-
C2H5O – C6H4 – N(O) = N – C6H4 – C4H9 and
butylaniline
N-(p-butyl nitroso benzene)-p-methoxy
C2H5O – C6H4 – N = (O)N – C6H4 – C4H9
aniline
Butyl-p-(p-ethoxyphenoxy carbonyl) NMR, ESR, UV,
C4H9O – COO –C6H4 – COO – C6H4 – OC2H5 nem: 56-87
phenyl carbonate IR
4,4′-di-hexyloxyazoxy benzene C6H13O – C6H4 – N(O) = N –C6H4 – OC6H13 nem: 81-127 ⎫
⎬ NMR
sm: 72-81 ⎭
4,4′-di-heptyloxyazoxy benzene C7H15O– C6H4 – N(O) = N –C6H4 – OC7H15 nem: 92-123 ⎫
⎬ NMR
sm: 74-92 ⎭
Terephthal-bis-(4-n-butyl)-aniline C4H9 – C6H4 – N = (H)C – C6H4 – C(H) = N –
sm A: 199.6-172.5 ⎫
(TBBA) C6H4 – C4H9
sm B: 172.5-144.1 ⎬ NMR
sm C: 144.1-113.0 ⎭
4-methoxybenzylidene-4-amino-alpha- CH3O – C6H4 – CH = N – C6H4 – CH =
NMR
methyl cinnamic acid-n-propylester C(CH3) – COOC3H7
1.17 FITTING OF A STRAIGHT LINE AND PRINCIPLE OF LEAST SQUARES

Let us consider the fitting of a straight line
y = c′ = mx (1.56)
to a set of n points (xi, yi); i = 1,2, 3,..., n. The problem is to determine the intercept c′ and slope m so that the line
(1.56) is the line of best fit.
The term ‘best fit’ is interpreted in accordance with the Legendre’s of least squares. The principle consists in
minimising the sum of the squares of the deviations of the actual values from its estimated values as given by the
line of best fit. According to principle of least squares, we have to determine c′ and m so that
n
∑ (y mx i ) is minimum
2
E i −c u (1.57)
i =1
Differentiating Eq. (1.57) with respect to c′ and m and applying the condition of maxima and minima, we get
⎛ ∂E ⎞ n
⎜⎝ ⎟
∂c ′ ⎠ m
= 0 = − 2 ∑ ( − − )
i =1
⎛ ∂E ⎞ n
⎜⎝ ⎟⎠ = 0 = − 2∑ xi ( − − ) (1.58)
∂m c ′ i =1
Consequently,
n n
∑yi =1
i = nc + m∑ xi
i 1
i=
n n n
∑ x y i = c′
i =1 i 1
xi m ∑ x i2
i =1
(1.59)
Equations (1.58) and (1.59) are known as the normal equations for estimating c′ and m.
n n n n
All the quantities, ∑x , ∑x
i =1
i
i =1
2
i , ∑y
i =1
i and ∑x yi =1
i i
can be obtained from the given set of points (xi, yi );
i = 1, 2, 3,…, n and the Eq. (1.59) can be solved for c′ and m. With the values of c′ and m so obtained, Eq. (1.56)
is the line of best fit to a given set of points (xi, yi); i = 1, 2, 3 ..., n.
Problem 1.2: By the method of least-square fit, find the slope and the intercept of the line y = c′ = mx from
the following data:
x: 1 2 3 4 6 8
y: 2.4 3 3.6 4 5 6
Solution The line is y = c′ + mx.

x y x2 xy
1 2.4 1 2.4
2 3.0 4 6.0
3 3.6 9 10.8
4 4.0 16 16.0
6 5.0 36 30.0
8 6.0 64 48.0
∑ 24
i =1
24 130 113.2
Using normal Eq. (1.59), we get 6c′ + 24m = 24; 24c′ + 130 m = 113.2. Solving these equations for c′ and m,
we obtain, intercept c′ = 1.976 and slope m = 0.50.
PROBLEMS
1. Using the data in Table 1.2, compute the energy-, fre- 1
quency-and wave number range for the various regions (c) log = eCl (d) log T = eCl
T
of electromagnetic radiation.
2. In what units is the molar decadic coefficient measured? (e) T = − eCl (f ) e = D /Cl
(a) cm−1 (b) moles/cm2 (c) cm2/mole (d) moles (e) It is a (g) −log T = eCl (h) T × 10D = 1
dimensionless quantity. [Ans: (c)]
3. Which of the equations listed below gives the correct (i) D = log T [Ans: (b), (c), (g)]
relation between the transmittance T, the optical density 4. Which of the equations listed below reflect correctly
D, the concentration C, the molar extinction coefficient e the relations that exist between the variables if the
and the layer thickness l? Beer–Lambert law is obeyed?
1 (a) At l = const if Cl < C2 < C3 then
(a) T = − log (b) T = − log D
D D1 /e1 >D2 /e2 > D3 /e3
Introduction 23
(b) If C1 < C2 < C3 then 13. Suppose we have doubled all the linear dimensions of
e1 l1/D1 > e2 I2 /D2 < e3 I3/D3 the resonator, (i.e., cavity length and mirror curvature
radii and apertures). Will the new resonator be equiva-
(c) At I = const, e = const and D0 = const, if C1 > C2 > lent to the original one? [Ans: No, NF = 2]
C3 then 14. When a source emitting radiation of frequency n moves
It ′ It ′′ < It ′′′ with velocity v towards or away from the observer, the
observer detects the radiation if frequency nobs given by
(d) At I = const, e = const and I0 = const, if C1 > C2 >
1/ 2
C3 then ⎡ v⎤
⎢1 ± c ⎥
It ′ It ′′ > It ′′′ ν obs ν⎢ ⎥
⎢1 ∓ v ⎥
(e) At e = const and D = const ⎣ c⎦
C1 C2 C3 where the plus and minus signs have their usual mean-
= =
I1 I2 I3 ings. Calculate the frequency of the radiation source
detected by the observer if the source with a 500 cm−1
(f) If C1 < C2 < C3 then frequency is moving with a 300 m/s velocity away from
or towards the observer. [Ans: 499.99 cm−1, 500 cm−1]
e1l1/D1 > e2l2 /D2 > e3l3/D3
15. (a) What is the width of a transition from a state with a
(g) At l = const if C1 < C2 < C3 then lifespan of 10.0 ps? (b) A laser line occurs at 500 nm
whereas the observer detects it at 500.1 nm. In what
D1/e1 < D2 /e2 < D3 /e3 [Ans: (c), (f ), (g)] direction and with what speed is the laser moving relative
to the observer? (c) An observer is moving towards the
5. The absorbance of some unknown solution having a
red traffic light (660 nm). At what speed would it appear
concentration Cx is Ax. What will be the value of absor-
green (520 nm)? (d) Calculate the Doppler broadening
bance in terms of Ax if the (a) path length is reduced by
for Ha (6562.79 Å) and Hb (4861.33 Å) lines of helium
one half, and (b) path length is doubled? [Ans: (a) 2Ax
at 300 K.
(b) 0.5Ax]
[Ans: (a) 0.53 cm−1 (b) Away from the observer with
6. A sample of a solution having concentration Cx obeys a speed of 6 × 104 m/s (c) 6.36 ×107 m/s
Beer–Lambert’s law. The solution shows 80 per cent (d) 0.0576 Å; 0.0426 Å]
transmittance when measured in a cell 10 cm in length:
16. The energies of different coloured photons of visible
(a) Calculate the per cent transmittance for a solu-
light are given below.
tion half and twice the concentration in the same cell.
(b) What will be the length of the cell to give the same
Colour Energy × l0−19(J)
transmittance (80%) for solution of twice the original
concentration? (c) Compute the per cent transmittance Violet 4.7
of the original solution when contained in cells of 1, 2
Blue 4.2
and 5 cm in length.
Green 3.7
7. Sharp-line spectra are observed in gases while in case
of solids and liquids, they are broad and diffused. Yellow 3.4
Explain.
Orange 3.2
• Hint: In gases the atoms or molecules are at a distant Red 2.8
apart, so a single quantum transition independent of
other transitions can take place.
What would be the wavelengths of the respective
8. Light absorption may be registered as graphs in the colours? [Ans: 420, 470, 530, 580, 620 and 700 nm]
form of various quantities, (a) lt /I0 (b) 100.It/I0 (c) (I0 − It)
17. (a) A Nd|3+: YAG laser (lYAG = 1064.l nm) produces
/I0 and 100(I0 − It) I0 (d) D = A = ln I0/It (e) A = 100 In It/
pulses at a repetition rate of l kHz. If each pulse is 150
I0 (f ) e = A/Cl, which of the quantities listed above are
ps in duration and has a radiant energy of 1.25 ×10−6 kJ,
plotted as a function of frequency (in cm−1) or less often,
calculate P the radiant power of each laser pulse and
the wavelengths in microns (μ) in infrared spectroscopy?
<P>, the average radiant power of the laser, the radiant
[Ans: (b), (e)]
energy of the laser photon and the number of photons
9. What frequency range corresponds to the interval of in a single pulse.
light wavelengths from 0.1 to 10 μm?
[Ans: 3 × 1013 –3 × 1015 Hz] • Hint: Radiant power P = Radiant energy per unit
10. What is the photon energy for a light wavelength of time.
0.6 μm? [Ans: 2 eV] Average radiant power P = Total power emitted per second
11. The quality factor of an optical resonator is Q = 2 × 107 at = (Energy/pulse) × [Number of pulses per second.)
0.6 μm wavelength. Compute the loss coefficient for this Q. Number of photons in the laser pulse
[Ans: 0.5 m−1]
Radiant energy per pulse
12. Compute the peak power of a pulse, given that the laser =
Radiant energy of 1064.1nm photon
n
operates at a pulse repetition rate of f = 1 MHz yielding
pulses of t = 20 ns duration with an average power of (b) Calculate the number of photons in a 2.00 mJ light
train being P = 10 W. pulse at 1.06 μm, 537 nm and 266 nm.
• Hint: The peak power in this case is P/tf. [Ans: (a) 8.3 × 106 W, 1.867 × 10−19 kJ, 6.70 × 1015
[Ans: 500 W] (b)107 × 1014, 54 × 1014 and 268 × 1013 photons]
18. A CO2 laser operating at 9600 nm uses an electric power 1.4 W, calculate the radiant power of each laser pulse.
of 5 kW. This laser produces 100 ns pulses at a repetition How many photons are produced by this laser in one
rate of 10 Hz and has an efficiency of 27%. Calculate the second? [Ans: 560 kW, 549 × 1016 photons]
number of photons in each pulse. 20. The ratio of areas under the peaks in the ultraviolet, vis-
• Hint: Radiant power of pulse = 5 kW ible and infrared regions in the electromagnetic spec-
[Ans: 652 × 1019 photons per pulse] trum of the sun as measured in the upper atmosphere is
1:4.44:5.66. What might be the per cent composition of
19. A titanium sapphire laser (l = 780 nm) produces pulses
solar radiation emanating from the surface of the sun at
at a repetition rate of 100 MHz. If each pulse is of 25 fs
6000 K? [Ans: 9, 40, 51 per cent]
duration and the average radiant power of the laser is
APPENDIX 1 A: GENERAL PHYSICAL AND CHEMICAL

CONSTANTS AND THEIR VALUES
Avogadro’s constant NA = 0.60229 × 1024 mole−1
Velocity of light c = 2.997925 × 108 m s−1
Electron mass m = 0.91083 × 10−30 kg
Electron charge e = 0.106206 × 10−18 C
Faraday F = NA e = 96,490 C mole−1
Dalton D = 1.66033 × 10−27 kg
Planck’s constant h = 0.66252 × 10−33 J s
= 3.9901 × 1013 amu Å2 s-1
Quantum of angular momentum h = h/2p = 0.105443 × 10−33 J s
Proton mass mp= 1.67239 × 10−27 kg
Neutron mass mn = 1.67470 × 10−27 kg
Boltzmann constant κ = 1.3805 × 10−23 J K−1
= 8.3142 × 1023 amu Å2 s-2 k-1
Gas constant per mole R = NA κ = 8.3146 J mole−1 K−1
Gas constant R = 0.08206 litre atm deg−1 mole−1
Standard molar volume of gas at 273K and 1.013 × 105 = 22.415 litre
Centigrade temperature t 0 C = T K – 273.15
Atmospheric pressure 1 atm = 101.325 kN m−2
Electric dipole moment 0.1602 × 10−28 C m (4.8029 D)
1 D = 10−18 esu cm
Electron-volt 1 eV = 96,4905 kJ mole−1
gcm2 = 6.0226 × 1039 amu Å2

kg m2 = 6.0226 × 1046 amu Å2
1 amu = 1.6604 × 10−27 kg
1.6604 J = 1047 amu Å2 s−2
Dimensions of charge
[Q] = [Length]3/2 [Mass]1/2 [Time]−1
Q2 = L3 M1 T-2 = cm3 g s−2
Thus e2 r 2 = cm3 g s−2 cm2 = g cm5 s−2
CHAPTER 2 MICROWAVE
SPECTROSCOPY
To find out from theoreticians what methods they use, stick closely to one principle, do not listen to their words, fix
attention on their deeds.
— Einstein
2.1 INTRODUCTION
Rotation of molecules manifests itself in the heat capacity of gases and in the structure of the rotational spectra.
The discrete nature of the spectra reveals that the molecular motion is quantised. Classical mechanics fails
to explain the fine structure of the spectra. However, the solution of the Schrödinger wave equation for the
rotational motion of the molecule suggests that the rotational motion is quantised.
The rotational spectra of all molecules except the lightest of the diatomics fall in the microwave region, i.e.
0.2–200 cm−1, while those of light diatomics such as HC1 occurs in the far infrared region, i.e. < 400 cm−1.
Furthermore, depending upon the spectroscopic nature of the molecules, the rotational spectrum may also be
deduced from the infrared, electronic and Raman spectra of the molecules. The rotational spectra in the far
infrared and microwave regions are recorded on a spectrometer, the working of which is described here.
2.2 MICROWAVE SPECTROMETER

The block diagram of a spectrometer for recording pure rotational spectra in the microwave (0.2–200 cm−1)
and far infrared (< 400 cm−1) regions is shown in Fig. 2.1. The monochromatic beam of microwave radiation
(22–60 kMHz) generated in the Klystron tube are passed through the wave guide metallic pipe (~10 m long),
fitted with mica windows at the end, containing a gaseous sample. The modified radiation, i.e. the transmitted
radiation detected by the crystal diode detector (which is capable to provide a 100 kHz output), is ampli-
fied. The amplified signals are analysed with the help of an analyser. The frequency of the Klystron tube is
adjusted electrically by the power-supply unit which in turn is synchronised with the analysing system. Har-
monic generators are employed to provide higher frequencies, i.e. 250 kMHz. The microwave spectrometer
is generally fitted with a square wave Stark modulator
(100 kHz modulation). The function of the Stark modu-
VS and SI
lator is to study the effect of electric field on the rota-
tional spectra of molecules. When the spectrum is to be
recorded in the far infrared region, the source employed KT AC CD
is a ‘Nernst glower’ which is electrically heated rare
earth oxides.
The microwave spectra are not recorded in the solid
Applied Field
and liquid phases. In liquids, the time period of molecu- SWSM
lar collision (10 − 10 s) is very small as compared

−12 −13
to the time for complete molecular rotation (~10−10 s). 100 kHz
Time
There is also no free rotation in solids. In the light of
this, the rotational motion in liquids and solids is not
quantised and hence their rotational spectra cannot be KPSU
SS
AM
studied by this technique. On the other hand, in gases AN
at very low pressure (~10−4 mmHg), a number of free Fig. 2.1 Basic elements of a microwave spectrometer:
KPSU—Klystron power supply unit, KT—Klystron tube, VS
rotations result before any molecular collision occurs. and SI—vacuum system and sample inlet, CD—crystal detec-
Therefore, the rotational spectra of gases are recorded tor, SWSM—square wave stark modulator, SS—syncronising
by this technique. signal, AC—absorption cell, AM—amplifier, AN—analyser.
2.3 MOMENTS OF INERTIA OF MOLECULES

The moments of inertia of a molecule rotating about the principal axis passing through its centre of mass will be
Ix ∑m r
i
i ix Iy =
i
m i riy I z ∑m r
i
2
i iz (2.1)
where mi stands for the masses of atomic nuclei and rix, riy and riz are the equilibrium distances from the atomic
nuclei to the axis of rotation.
Assume that the origin of coordinates is the molecule’s centre of mass, whose ordinates will be determined
from the expression
∑ m i rix = 0 ∑ m i riyiy = ∑ m i riz = 0
i i i
(2.2)
Thus, in general we can write

I ∑m r
i
i i
2
(2.3)
∑m r = 0i
i i (2.4)
Problem 2.1: Determine the moment of inertia of a diatomic molecule about an axis passing through its centre
of mass and normal to the molecular axis if the equilibrium bond distance is r Å.
Solution Knowing the atomic masses of the nuclei in the diatomic molecule, we can determine the moment of
inertia of the molecule from Eq. (2.3). The geometrical configuration of a diatomic molecule is presented in Fig. 2.2.
Let m1 and m2 be the masses of the nuclei of the diatomic molecule and
r1 and r2 be the distances of the respective nuclei from the centre of mass
Cm of the molecule such that the bond length is
r = r1 + r2 (2.5)
r1 r2
By Eq. (2.4), i.e. taking moments about Cm, we get
m1r1 = m2 r2 (2.6) m1
Cm
m2
From Eqs (2.5) and (2.6) r

m2 r m1r
r1 = ; r2 = (2.7)
m1 + m2 m1 + m2
Fig. 2.2 Geometrical configuration of a
From Eq. (2.3), the general expression for the moment of inertia of a diatomic molecule.
diatomic molecule is
I m1r12 + m2 r22 (2.8)
In diatomics, the moment of inertia about the molecular axis is zero since all the masses concentrate on this
axis, i.e. for this particular rotation in diatomics; since r = 0, hence I = 0. Substituting the values of r1 and r2
from Eq. (2.7) into Eq. (2.8), we get
m1m 2 2
I= r μm r 2 (2.9)
m1 + m 2
m1m 2 mm
where μm = = 1 2 (2.10)
m1 + m 2 Mm
Here, mm is called the reduced mass of the molecule, and Mm is the molecular mass of the diatomic molecule.
For example, let us now calculate the moment of inertia of 35C137C1 if the equilibrium bond distance is 1.99 Å.
We know that m1 = 35 amu, m2 = 37 amu, and r = 1.99 Å.
From Eq. (2.9), we obtain
35 (amu) × 37 (amu)
I= × (1.99)2 ( ) 2
(35 + 37) amu
m
= 71.22 amu Å 2 = 71.22 au

Microwave Spectroscopy 27
Problem 2.2: Derive the general expression for the moment of inertia of (a) linear triatomic molecules of the
type XYZ, and (b) symmetric triatomic linear molecules of the type XY2.
Solution (a) The linear triatomic molecules such as ICN, COSe, HCN, OCS, etc., are of XYZ type. By
taking HCN as an example, we have to proceed as follows. The geometrical configuration of HCN is shown
in Fig. 2.3.
Here, mH, mC, mN are the masses of hydrogen, carbon and nitrogen atoms respec- Cm
tively; rCH and rCN are the bond distances of CH and CN bonds respectively and r is
the distance of the carbon atom from the centre of mass of the molecule. mH mC mN
Using Eq. (2.4), i.e. taking moments about Cm, we obtain
H C N
mH ( rCCHH r ) + mC r mN (r
( rCCNN r) = 0 (2.11)
Rearranging Eq. (2.11), we get Fig. 2.3 Geometrical configura-

tion of HCN.
( m N rCN − m H rCH ) ( m N rCN − m H rCH )
r= = (2.12)
mH + mC + m N Mm
where M m mH + mC + mN is the molecular mass of HCN.
3
Further by Eq. (2.3), i.e. I ∑m ri i
i i
2
, we get
I m H ( rCH + r ) 2 + m C r 2 + m N ( rCN − r ) 2 (2.13)
= Mm 2
+ 2r ( m H rCH − m N rCN ) + ( m H rC2H + m N rCN
2
)
Substituting the value of r from Eq. (2.12) into Eq. (2.13), we obtain
( mH rCCHH − mN rCN ) 2
I mH rC2H + mN rCN
2
− (2.14)
Mm
Thus employing Eq. (2.14), we can determine the moment of inertia of any molecule of type XYZ, e.g. let us
determine the moment of inertia of OCS. For OCS, the expression (2.14) may be written as
( mO rCCOO − mS rCS ) 2
I OCS 2
mO rCO + mS rC2S −
Mm
where Mm mO + mC + mS = 16 + 12 + 32 = 60 amu.
We know, rCO 1.163 Å and rCS = 1.558 Å.
Therefore,
(16 × 1.163 − 32 × 1.558) 2 = 83.04 amu Å 2
I OCS = 16 × (1.163) 2 + 32 × (1.558) 2 −
60
(b) CO2 and CS2 belong to XY2 type of molecules. The geometrical configuration of XY2 systems is given in Fig. 2.4.
Thus, the general expression for the moment of inertia of XY2 systems is
2
mr 2 + mr 2 = 2mr 2 ∑ mrr
i i
i
2
Cm
Problem 2.3: Describe the general method to determine the product of moments of
inertia of bent triatomic molecules such as H2S, H2O, SO2, etc., and apply the method to
SO2 if SO bond distance is 1.432 Å and OSO bond angle is ll9°21´. r r
X X
Solution Assume that the origin of coordinates is the molecule’s centre of mass; then m m
from Eq. (2.2),

∑ mi rix
i
mi riiyy ∑ mi riz = 0
i i
Fig. 2.4 Geometrical con-
figuration for XY2 systems.
The moments of inertia will be determined from the expression (2.1), i.e.
Ix ∑m r
i
2
i ix , I y = ∑ m i riy2 , I z
i
∑m r
i
2
i iz
The product of the principal moments of inertia of the molecule can be calculated using the third-order secular
equation,
+ I xx − I xy − I xz
I A I B I C = − I xy + I yy − I yz
−II xz I yz + I zz
= I xx ( I yy I zz − I yz I yz ) − I xy ( I xy I zz + I yz I xz ) − I xz ( I xy I yz + I yy I xz ) (2.15)
∑m
1
( ) 1
( )
2 2
where I xx i yi + zi − mi y i − mi z i
Mm Mm
∑m
1
( ) 1
( )
2 2
I yy i xi + z i − mi x i − mi z i
Mm Mm
∑ m (x
1
( ) 1
( )
2 2
I zz i
2
i + yi2 ) − mi xi − mi yi
Mm Mm
I xy ∑m x y i i i −
1
Mm
( mi xi ))(( mi yi )
I xz ∑m x z i i i −
1
Mm
( mi xi ))(( mi zi )
I yz ∑m y z i i i −
1
Mm
( mi yi ))(( mi zi )
Now, let us apply this method to determine the product of moments of inertia IA, IB, IC of the sulphur diox-
ide molecule. The equilibrium internuclear distance rS−O = 1.432 Å and the equilibrium bond angle is equal to
119.34°. The equilibrium geometrical configuration of SO2 is shown in Fig. 2.5.
O1 Y
1.432Å
59.67°
P
−X X
S
1.432Å
O2 −Y
Fig. 2.5 Equilibrium geometrical parameters
of molecule of SO2 for determining the product
of moment of inertia.
From right-angle triangle O1PS

x r cos θ = 1.432 cos 59.67° = 0.723
y r sin θ = 1.432 sin 59.67° = 1.236
mO 16 amu; ms = 32 amu
The data generated are recorded in tabular form as under.
Atomic masses and coordinates of the atoms.
Atom m, amu x, Å y, Å z, Å
S 32 0 0 0
O1 16 −0.723 1.236 0
O2 16 −0.723 −1.236 0
3
Mm ∑m
i =1
i = 64 amu
Terms involved in the secular equation.

3
Atom O1 O2 ∑i=1
2
x i 0.523 0.523
yi2 1.528 1.528
xi yi −0.894 0.894
mi xi −11.568 −11.568 −23.136
mi yi 19.776 −19.776 0
mi xi yi −14.298 14.298 0
(
mi xi2 + yi2 ) 32.804 32.804 65.608
m (x
i
2
i +z )
2
i
8.363 8.363 16.726
m (y
i
2
i +z )
2
i 24.443 24.443 48.886
Now from Eq. (2.15),
∑m (y ) 1
( ) 1
( )
2 2
I xx i i + zi − mi yi − mi zi = 48.886
886 au
Mm Mm
∑ m (x
1
( ) 1
( )
2 2
I yy i
2
i + zi2 ) − mi xi − mi zi
Mm Mm
( −23.136) 2
= 16.726 − = 16.726 − 8.363 = 8.363 au
64
∑ mi xi + yi − M
1 1
( ) ( )
2 2
I zz mi xi − mi yi
m Mm
( −23.136) 2
= 65.608 − = 65.608 − 8.363 = 57.245
245 au
64
I xy I xz = I yz = 0
I A I B I C = 48.886 × 8.363 × 57.245 (au )3 = 23.403 × 103 (au)3
Problem 2.4: Apply the secular Eq. (2.15) to determine the moments of inertia of molecules of the XY3 AB type.
Find also the reduced moment of inertia.
Solution CH3OH, CH3SH, etc., belong to XY3 AB type molecules. Taking CH3OH as an example, let us apply
Eq. (2.15) to work out the moments of inertia from the bond distances and bond angles of methyl alcohol. The
structural parameters of methyl alcohol are:
∠COH = 110°, ∠HCH = 09°28 ’, (O H) = Å,
r( ) 1. Å n r (C H)) .11 Å.
Let the carbon atom be the origin of coordinates. The X-axis is in the direction of the C—O bond. The equilib-
rium bond distances, bond angles and two projections in the XZ and YZ planes are presented in Fig. 2.6.
Z Z
H2
H2
H1 H1
109°28′ 110° O C
Y
C X
O
H3 H4
H3,H4
Fig. 2.6 Projections and geometrical configurations of the CH3OH
molecule.
The data required for Eq. (2.15) are generated and recorded in tabular form as follows:
Atom m, amu x, Å y, Å z, Å
C 12 0 0 0
O 16 1.43 0 0
H1 1 1.43 + 0.96 cos 70° = 1.76 0 0.96 sin 70° = 0.90
H2 1 1.11 cos 109° 28′ = − 0.37 0 1.11 sin 70° 32′ = 1.05
H3 1 −0.37 1.11 cos 240° = − 0.96 1.11 sin 240° = − 0.56
H4 1 −0.37 0.96 −0.56
6
Mm ∑m
i =1
i = 32
6
Atom C O H1 H2 H3 H4 ∑
i=1
x2 0 2.045 3.098 0.137 0.137 0.137

y 2
0 0 0 0 0.922 0.922
z 2
0 0 0.810 1.013 0.314 0.314
xy 0 0 0 0 −0.355 −0.355
xz 0 0 1.584 −0.389 0.207 0.207
yz 0 0 0 0 −0.538 −0.538
mx 0 22.880 1.760 −0.370 −0.370 −0.370 23.530
my 0 0 0 0 −0.960 0.960 0
mz 0 0 0.900 1.050 −0.560 −0.560 0.830
mxy 0 0 0 0 0.355 −0.355 0
6
Atom C O H1 H2 H3 H4 ∑
=1
mxz 0 0 1.584 −0.389 0.207 0.207 1.609

myz 0 0 0 0 0.538 −0.538 0
m (x + y )
2 2
0 32.704 3.097 0.136 1.057 1.057 38.051
m (x + z )
2 2
0 32.704 3.907 1.238 0.449 0.449 38.747
m (y + z )
2 2
0 0 0.810 1.102 1.234 1.234 4.380
Solution of the secular Eq. (2.15) gives the following moments of inertia.
0 + 0.688
I xx = 4 38 − = 4.38 − 0.021 = 4.359 au
32
553.660 + 0.688
I yy = 38.747 − = 38.747 − 17.323 = 21.424 au
32
553.660 + 0
I zz = 38.051 − = 38.051 − 17.301 = 20.750
750 au
32
0
lxy = 0 − = 0 au
32
23.53 × 0.83
lxz = 1.609 − = 1.609 − 0.610 = 0.999
999 au
32
0
l yz = 0 − = 0 au
32
Thus, the product of the principal moments of inertia is
IA IB IC = Ixx (Iyy Izz − Iyz Iyz) − Ixy (Ixy Izz + Iyz Ixz) − Ixz (Ixy Iyz + Iyy Ixz)
IA IB IC = 4.359 (21.424 × 20.750 − 0 × 0) − 0.999 (0 × 0 + 21.424 × 0.999)

= 1937.784 − 21.381 = 1916.403 (au)3
The group CH3 rotates relative to the OH group. The rotation of these groups is free since C—O is a sigma
bond. Hence the rotational energy exceeds that of the potential barrier. The reduced moment of inertia,
I CH3 I OH
I red =
I CH3 I CO
I CH3 = 3m (r sin70032′ )2 = 3(1.11 × 0.943)2 = 3.286 au.
H C−H
IOH = 1 × mH(rO−H sin700 )2 = 1(0.96 × 0.94)2 = 0.814 au.
3.286 × 0.814 2.674

Hence, I red = = = 0.652 au
3.286 + 0.814 4.1
The symmetry number of CH3OH is equal to the product of symmetry numbers of CH3 and OH groups,
i.e. 3 × 1 = 3
The principal moments of inertia of some important typical molecular systems can be determined directly from
the following simple expressions:
(i) For symmetric triangular molecules of the type of XY2, i.e. H2O, H2S, etc., the expression for the moments
of inertia are
m
I x = r12
2
(2.16)
2mm ′ ⎡ 2 ⎛ r1 ⎞ ⎤
Iy = ⎢r − ⎥
2m + m ′ ⎢⎣ 2 ⎜⎝ 2 ⎟⎠ ⎥⎦
Iz = Ix + Iy
where r1 is the distance between identical atoms, r2 is the distance between different atoms, m is the mass
of each of the identical atoms and m′ is the mass of the third atom. The validity of the relation Iz = Ix + Iy
will be discussed else where in this chapter.
(ii) In case of symmetric tetrahedral molecules such as CH4, CCl4, etc., the moments of inertia are expressed as
m 2
Ix Iy =
r1 and I z = mrr12 (2.17)
2
where m is the mass of each of the end atoms and r1 is the distance between these end atoms. The distance
between the end atom and the central one, i.e. between the vertex and centre of the tetrahedron, is
3
r2 r (2.18)
8 1
For CH3D,
I B (CH 3
⎡5 (mD mH )
⎤ 2
) ⎢ mH mD ⎥ r 2 ( CH )
mC + 3mH + mD ⎥
(2.19)
⎢⎣ 3 ⎦
(iii) For planar XY3 type molecules, e.g. BF3,
IA 3
= m y rxy
x
2
IB = (2.20)
2 2
and for nonplanar XY3 type molecules, e.g. NH3, ND3,
rHH (
3 rNH − hp ) (2.21)
⎡ 3( 2 2
)⎤
cos α = ⎢1 −
p
⎥ (2.22)
⎢⎣ 2r 2 ⎥⎦
Here, a = HNH
IA 3m y rxy2 sin 2 β
3m y rxxy2 ⎡ ⎛ 3m y ⎞ 2 ⎤
IB = ⎢2 − 1 − sin β ⎥ (2.23)
⎛ 3m y ⎞ ⎣ ⎜⎝ m x ⎟⎠ ⎦
2 ⎜1 +
⎝ m x ⎟⎠
Height of pyramid hp = r cos b (2.24)

where b is the angle of X−Y with the threefold axis. y x
(iv) For linear X2 Y2 type molecules, e.g. C2H2. C2HD, the
expression for moments of inertia is deduced as follows: H C C H
The geometrical configurations of C2H2 and C2HD are Cm
given in Fig. 2.7. In C2HD the centre of mass Cm is shifted d
towards D. H C C D
Equation (2.3) yields
4
C¢m
I ∑m r
i =1
i i
2
2mC y + 2mH x for C2H2
2 2
(2.25) Fig. 2.7 Geometrical configurations of C2H2 and C2HD.
For C2HD, from Eq. (2.4) we write,

4
∑m r = m
i=1
i i H (x + d mC ( y d ) − mD (xx − d mC ( y d ) = 0 (2.26)
Rearranging Eq. (2.26) we obtain

(m D mH )x
d= (2.27)
M m*
where M m* 2m C + m D + m H
Thus, the moment of inertia of C2HD about its centre of mass is
I* 2m C y 2 + ( m + m ) x 2 − M m* d 2 (2.28)
From I and I* we can obtain x and y, i.e.
x=
1
rCC + rCH =
(
M m* I − I )
2 M m (m D − m H )
1 I 2mH x 2
y r = (2.29)
2 CC 2mC
Problem 2.5: Determine the principal moments of inertia of methane if its bond length is 1.09 Å.
Solution From Eq. (2.18), we get
8 8
r1 = r2 = × 1.09 =1.779Å
3 3
Thus, from Eq. (2.17)
1
Ix I y = (1.779) 2 = 1.582 au
2
I z = (1.779) × 1 = 3.164 au
2
and
2.4 DIATOMIC MOLECULE AS A RIGID ROTATOR

The mass of a nucleon is approximately 1840 times than that of an electron. The diameter of the lightest nuclei is
of the order of 10−13 cm while the bond lengths are of the order of 10−8
cm. The total mass is concentrated in the two nuclear regions. In the
light of these facts, we can assume that the diatomic molecule behaves
as a rigid dumb-bell, i.e. when the molecule rotates, the bond distance
remains unaltered. Let us now study the rotational motion of the rigid
diatomic molecule about an axis passing through its centre of mass Cm r1 r2
and perpendicular to the molecular axis as shown in Fig. 2.8.
Here, m1 and m2 are the masses of the nuclei of the diatomic molecule, m1 m2
Cm
r is the bond length, r1 and r2 are the distances of the nuclei from Cm such
r
that r = r1 + r2. Proceeding as in Problem (2.1), we obtain
I μm r 2 (Eq. 2.9)
Fig. 2.8 Rigid dumb-bell model of a diatomic
molecule.
Thus, we are able to reduce a two-body problem to a one-body
problem. According to classical mechanics, the energy of rotation is
expressed as
P2 (2.30)
E rot 2π I ω 2 =
2I
where w is the frequency of rotation in Hz and P is the angular momentum about an axis through the centre of mass
and normal to symmetry axis. According to this expression, all energies are possible with a frequency of rotation w,
i.e. the rotational spectra should be continuous. But this is contrary to the experimental observations. Actually, the
rotational spectra of diatomic molecules have a fine structure. In order to explain the fine structure of the rotational
spectra, let us describe the rotational motion of a diatomic molecule of mass mm by Schrödinger wave equation in the
polar coordinates, i.e.
8π 2 μ m
∇2 ψ + E rot = 0 (For solution see Appendix A 2.1A) (2.31)
h2
where ∇2 is the Laplacian operator in polar coordinates. The solution of Eq. (2.31) leads to the expression
h2
EJ = J ( J + 1) (2.32)
8π 2 I
or the rotational term value,
EJ h
FJ = = 2 J (J + 1), in cm −1
hc 8π Ic
where EJ is the energy of the allowed energy levels and J represents the total angular momentum of the molecule,
h
i.e. P J (J + 1) . . It is called the rotational quantum number and can take up any integral value including
2π
zero, i.e. J = 0, ±1, ±2, ±3,…. Equation (2.32) holds good if the molecular motion is considered due to the motion
of the nuclei only. Equation (2.32) may also be expressed as
c ( J + 1)
E J = BhcJ (2.33)
Here, B is called the rotational constant and is defined as
h 16.8575 −1
B= = cm (2.34)
8π Ic
2
I
The quantum mechanical derivation based on Schrödinger wave equation of energy levels of a restricted rotator
is dealt with in Appendix 2.2A.
2.4.1 Spacing Between Two Adjacent Allowed Energy Levels

The energies of the two successive levels may be expressed as
c ( J + 1)
E J = BhcJ (Eq.2.33)
E J +1 = Bhc ( J 1) ( J + 2) (2.35)
Subtracting Eq. (2.33) from Eq. (2.35), we obtain
ΔErot E J + − E J
Δd J = = = 2 B ( J + 1) cm −1 (2.36)
hc hc
From Eq. (2.36), we get
J Spacing, D dJ
0 2B
1 4B
2 6B
3 8B
… …………
Thus, the spacing between the two successive levels grows with the growth of J and the difference between the
adjacent spacings is 2B. The behaviour of ΔdJ with J is shown in Fig. 2.9.
8B
6B
2
4B
1
2B
0
Fig. 2.9 A sketch showing the influence of quantum number on the permissible
energy levels and the separation between the adjacent levels of a rigid rotator.
2.4.2 Selection Rules and Transition Between Permissible Energy Levels

The discrete nature of the spectrum suggests that not all transitions are permissible among all the energy levels.
The transition between any two adjacent levels occurs in accordance with some specific rules known as selection
rules. Let M be the permanent dipole moment of the molecule and Mx, My and Mz are its components in the direc-
tion of three axes such that M = M x2 + M y2 M z2. The selection rules have been deduced quantum mechanically
from the solution of the integral
M t = ∫ ψ *i M̂ t ψj dt (2.37)
where Mt is the transition dipole moment during the transition between the states j and i, M̂t is the transition dipole
moment operator and is related to the permanent dipole moment M. Ψi and Ψj are the wave functions of energy
states i and j respectively and are orthogonal to each other, and dτ is the volume element. The transition dipole
moment arises due to the interaction of the oscillating electric field of radiation (periodically reversing) with the
dipole moment of the molecule. It is discussed in Chapter 3 on infrared spectroscopy. The components of Mt along
three axes are
∧
M t ( x ) = ∫ ψ i M t (x )ψ j dτ (2.38)
∧
M t ( y ) = ∫ ψ i M (y )ψ j dτ (2.39)
M t ( z ) = ∫ ψ i Mˆ t ( z ) ψ j dτ (2.40)
The transition dipole moment operator in the three directions is

M t ( x ) = ex, M ( y ) = ey, M t (z) = e
∧ ∧ ∧
(2.41)
where e is the electronic charge and x, y, z are the Cartesian coordinates:

From Eqs (2.38) to (2.41), we obtain
t( x ) = e ∫ ψ i x ψ j dτ (2.42)
M t ( y ) = e ∫ ψ i y ψ j dτ (2.43)
M t ( z ) = e ∫ i ψ j dτ (2.44)
In polar coordinates, we may write

Mx M sin θ cos ϕ ⎤
⎥
M y = M sin θ sin ϕ ⎥ (2.45)
M z M cos θ ⎥
⎦
The transition dipole moment can now be expressed as

π 2π
M t (x ) = ∫ ∫ ψ i Mˆ t ( x ) i θ cos ϕ ψ j dτ
θ= 0 ϕ = 0
π 2π
M t (y ) = ∫ ∫ ψ i Mˆ t ( y ) i θ sin ϕ ψ j dτ (2.46)
θ= 0 ϕ = 0
π 2π
M t (z ) = ∫ ∫ ψ i Mˆ t ( z ) θ ψ j dτ
θ= 0 ϕ = 0
where dτ d θdϕ .
The wave functions for the two states may also be expressed as
ψi ψ j ′′ , M j ′′ ⎤
⎥ (2.47)
ψj ψ j ′ , M j ′ ⎥⎦
where double prime stands for upper level while single prime corresponds to lower level, and MJ is the component
of J along the direction of an external electric or magnetic field. Thus, in the absence of any external field, the
selection rules are as follows:
(a) M ≠ 0,
0 i.e. Molecules having Permanent Dipole Moment will be Microwave Active.
In other words, we can say that molecules whose electrical centre do not coincide with their centre of mass, will
be microwave active, e.g. CO, 16O18O, 35Cl37Cl, HD, etc. Thus, symmetric molecules such as H2, O2, Cl2, N2, etc.,
with M = 0 will be microwave inactive.
(b) Mt ≠ 0
Only those transitions are allowed for which at least one of the components of Mt, namely Mt (x), Mt (y) or Mt (z)
is nonzero, i.e. the permanent dipole moment should change during the transition between the two states. In non-
symmetric diatomics, we know that the electric centre of the molecule does not coincide with the centre of mass,
so that when the state of the molecule changes, i.e. J changes, its dipole moment changes.
(c) The Pauli–Fermi Principle Should be Followed, i.e. the Transitions for which Δ J 1, Δ M j = 0 , are
Permissible.
Let J' and J" be the rotational quantum numbers of two states. With the aid of Eq. (2.33), the energy difference
between the two states may be expressed as
EJ − EJ ′ ΔE rot
ν rot = = = B (J ″ − J ′ )(
) (J ″ + J ′ + 1) (2.48)
hc hc
According to the Pauli–Fermi principle, i.e. J" − J ′ = ±1, we obtain
ν rot = 2BJ ″ (2.49)
where J ″ = 1, 2, 3,…
or ν rot = 2 (J ′ + 1) (2.50)
where J ′ = 0, 1, 2, 3,…
Thus, according to Eq. (2.49) or (2.50), the rotational transitions 0→1, 1→2, 2→3,… will be positioned at 2B,
4B, 6B,… cm–1, respectively. Interestingly, the spacing between the adjacent lines in the spectrum and the differ-
ence between the successive spacings of the energy levels are the same, i.e. 2B. Since B is very small, i.e. in case
of HC1, B = 10 cm–1, the energy difference between any two adjacent rotational energy levels will also be very
small. Therefore, even at ambient temperatures, the higher rotational levels are also populated. Consequently, the
rotational spectra of diatomic molecules consist of a number of equally spaced lines as shown in Fig. 2.10.
Knowing 2B from the spectrum, the moment of inertia and hence, bond length of the diatomic molecule can be
determined with the aid of Eq. (2.34). The experimental data on the rotational spectra of diatomics reveal that the
lines are not equally spaced as expected on a theoretical basis. The difference between the observed and theoretical
value of 2B is significant. This discrepancy may be because of the assumption that the diatomic molecule behaves
as a rigid rotator. In reality, when the molecule rotates, due to centrifugal force, the bond length is slightly stretched
8B
6B
4B
1
2B
0
2B 4B 6B 8B 10B
ν– (cm−1)
Fig. 2.10 Allowed rotational energy levels and the spectrum of a rigid diatomic molecule.
and the equilibrium distance is dependent on J. In the light of this, a ‘correction’ corresponding to this effect should
be introduced in the rotational energy expression of the diatomic molecule, i.e. in Eqs. (2.32) and (2.33).
Problem 2.6: The lines in the far IR spectrum of HC1 are positioned at the following wave numbers:
Line No. 1 2 3 4 5 6 7
vrot, cm −1 85.384 106.730 128.076 149.422 170.768 192.114 213.466
Find the average value of moment of inertia and bond length of HC1.
Solution The average value of difference between the wave numbers of successive lines, 2B = 21.346 cm−1.
Thus B = 10.673 cm−1.
Using Eq. (2.34) we obtain
I=
3.9890 × 1013 ( )
8 × 3.14 × 3.14 × 3 × 10 10
( ) × 10.673(cm ) −1
16.8575
= amu Å 2 = 1.57945 au
10.673
Thus, from Eqs (2.9) and (2.10)

35 × 1 35
μm = = amu
35 + 1 36
1.57945 amu Å 2
r= =1.274 Å
(35 / 36) amu
Problem 2.7: The average value of spacing between the adjacent lines in the rotational spectrum of the diatomic
molecule CX is 3.84235 cm−1. The equilibrium distance is 1.131 Å. Find the atomic mass of X.
Solution Given 2B = 3.84235 cm−1.Thus, B = 1.921175 cm−1.
r = 1.131
131 Å
From Eqs (2.34) and (2.9) we get

16.8575 8.7745
I= = 8.7745 au, μm = = 6.859 amu
1.921175 (1.131)2
Let the mass of atom X in CX is mx. Then applying Eq. (2.10), we obtain
12mx
6.859 = or mx = 16 amu
12 + mx
Hence the CX molecule is actually CO.
Problem 2.8: Calculate the wave numbers of the lines in the rotational spectrum of CO for the 0→1, 1→2,
2→3 transitions if the equilibrium bond distance of CO is 1.131 Å.
Solution From the preceding problem,

I = 6.859 × (1.131) = 8.7737
2
7737 au
From Eq. (2.34), 16.8575

B= = 1.9213 cm −1
8.7745
Hence, vrot (0 → 1) 2 B 3.8426 cm −1

vrot (1 → 2) 4 B 7.6852 cm −1
vrot ( 2 → 3) 6 B 11.5278 cm −1
Problem 2.9: The average value of spacing between the adjacent rotational lines in the spectrum of NaCl is
0.432 cm−1. Show that ion pairs are present in the vapours of NaCl.
Solution Given 2B = 0.432 cm−1 .Therefore, B = 0.216 cm−1.

From Eq. (2.10) we obtain the reduced mass of NaCl, i.e.
23 × 35.5 23 × 35.5
μm = = amu
23 + 35.5 58.5
Now using Eq. (2.34),
16.8575 16.8575 × 58.5

r= = = 2.364 Å
B μm 0.216 × 23 × 35.5
The calculated value of bond distance, i.e. 2.364 Å, agrees well with the bond distance of NaCl determined
by x-ray diffraction technique. From this, it is inferred that in the vapours of NaCl, ions pairs are also present in
addition to the Na+ and Cl− ions.
Problem 2.10: Determine the rotational energy of CO on the quantum levels J = 1 and 2 if the equilibrium
nuclear distance of CO is 1.131 Å.
Solution From Eqs (2.10) and (2.9), we obtain
16 × 12 192
μm = = = 6.857
857 amu
16 + 12 28
From problem 2.8,
Therefore, I = 6.859 × (1.131)2 = 8.7737 au.
With the aid of Eq. (2.32), we can determine the energy of the rotational levels for J = 1 and 2.
(3.9890 × 1013 amu Å 2 s 1 )

E1 = ×1
1(1 + 1)
8 3.14 × (8.7737 amu Å 2 )
= 45.986 × 1023 amu Å2 s−2
= 45.986 × 1023 × 1.6604 × 10−47 J
= 0.763 × 10−22 J
E2 = 22.993 × 1023 × 2(2 +1)
= 137.958 × 1023 amu Å2 s−2
= 137.958 × 1023 × 1.6604 × 10−47 J
= 0.229 × 10−21 J
Problem 2.11: Predict and sketch the general behaviour of isotopic substitution on the position of the rotation
line as well as on the spacing between the adjacent lines in the rotational spectra of molecules.
Solution A slight change in bond length on isotopic substitution occurs because of the zero point energy which
depends upon the mass of the atoms. However, for all practical purposes, the bond length in the different isotopic
species of the same molecule is assumed to be the same. From Eq. (2.34), we write
h
B= (Eq. 2.34)
8π 2 Ic
h
B* = (Eq. 2.34*)
8π * c
2
where * stands for heavy molecule.

Dividing Eq. (2.34) by Eq. (2.34*), we obtain
B I * μ*m
= = (2.51)
B* I μm
Since B varies inversely with the moment of inertia of the molecule, the effect of isotopic substitution results
not only in shifting the rotational lines but also in decreasing the separation between
J
the adjacent lines in the spectra of the diatomic molecules and is shown in Fig. 2.11.
With the aid of Eq. (2.51), we can also determine the atomic mass of the isotope of
the molecule.
3
Problem 2.12: Determine the difference between the wave numbers corresponding
6B∗ 6B
to 1 → 2, transition in the rotational spectra of 12C16O and12C18O. The equilibrium bond
length in both the cases is 1.131 Å.
2
Solution From Eq. (2.9) we obtain 4B∗ 4B
12 × 16
I ( 12
C16 O = ) 12 + 16
× (1.131) 2 = 8.7713 au 2B∗ 2B
1
12 × 18
I ( 12
C18 O = ) 12 + 18
× (1.131) 2 = 9.2099 au
2B∗
4B∗
6B∗
2B
4B
6B
Thus, from Eq. (2.34) we get

ν– (cm−1)
B ( 12
C O =
16
)
16.8575
8.7713
= 1.9218 cm −1 Fig. 2.11 A sketch showing
the effect of isotopic substi-
tution on the energy levels
B ( 12
C18 O =) 16.8575
9.2099
= 1.8303 cm −1 and the spectrum of the rigid
diatomic molecule.
Hence, for the transition 1→2

vrot (12 C16 O) 4 B 7.6872 cm −1
vrot (12 C18 O) 4 B 7.3212 cm −1
Thus, the isotopic substitution shifts the wave number of the spectral line corresponding to 1 → 2 transition to
the lower side and the magnitude of shift is 0.3660 cm−1.
Problem 2.13: The average value of difference between the wave numbers of adjacent lines in the rotational
spectra of 12C16O and 12CX* are 3.8436 and 3.6606 cm−1 respectively. Determine the atomic mass of an isotope of
the atom X, i.e. X*.
Solution We know that isotopic substitution does not affect the bond length. Let atomic mass of the atom X*
be m*. Given B = 1.9218 cm−1 and B* = 1.8303 cm−1.
Now employing Eq. (2.51), we get
1.8303 12 + m * 12 × 16 16 × 12( * +12) 4( m * )
= × = =
1.9218 12 m * 16 + 12 28 × 12m * 7m *
12 × 4 1.9218 48 × 1.9218
m* = = = 17.999 am
mu
( .8303 − 4 1.
.8303 ) 5.1249
Thus, the atomic mass of the isotope of atom X, i.e. X* is 18 amu. The molecular formula of isotopically sub-
stituted molecule will be l2Cl8O.
2.5 DIATOMIC MOLECULE AS A NONRIGID ROTATOR

As we have already discussed, during the rotation of the molecule, the bond distance does not remain fixed. Thus,
taking into consideration the role of centrifugal force in the rotational motion of the molecule, the allowed energy
of the rotational level, as deduced from the solution of the Schrödinger wave equation, is expressed as
h2 h4
EJ = J ( J + 1) − J 2 (J + 1) 2 + (2.52)
8π 2 I 32π 4 I 2 r 2 k
= hcBJ (J+1) − hcDJ 2 (J+1)2 +….
where D is the centrifugal distortion constant and is defined as
4B 3
D= (2.53)
ω2
Here, D << B and D is 10−4 B, w is the vibrational frequency of the harmonic oscillator and is expressed as
k (millidynes//Å )
w(cm−1) = 1303.16 (2.54)
μ m (amu )
where k is the force constant and is defined as the restoring force to bring the molecule back to its original position.
An approximate value for k of the diatomic molecule can also be obtained from the expression
3/ 4
k = a Nbo ⎡ X 1 X 2 ⎤ b (2.54a)
⎢ r ⎥
⎣ ⎦
in which Nbo is the bond order (number of bonds between the two atoms), X1 and X2 are the electronegativities of
the atoms, r is the internuclear distance, a and b are constants with values 1.67 and 0.30 respectively.
The restoring force F = −k (r−re) = −kΔr (2.55)
where Δr = r − re is the displacement from the equilibrium position. The Potential Energy (PE) of a harmonic
oscillator is expressed as
1
U (r) = k (r−re)2 (2.56)
2
The PE curve of a harmonic oscillator will be a parabola as shown in Fig. 2.12.
The harmonic assumption is valid only in the lowest vibrational state of the
molecule and will be discussed in detail in Chapter 3 on infrared vibrational spec-
troscopy. However, if the molecule is not in the lowest vibrational state then the
correction due to molecular vibration which changes the moment of inertia of the
molecule must be applied. U(r)
The correction for the rotational constant BV in the Vth vibrational level is
2
BV = Be−ae ⎛ 1⎞ ⎛ 1⎞ (2.57)
⎜⎝V γe V + ⎟
2⎠ ⎝ 2⎠ re
r
where Be is the rotational constant for the theoretical equilibrium state of no vibra-
Fig. 2.12 The change in poten-
tion, ae, and γe are the rotational-vibrational coupling constants. ae, and ge are very tial energy with internuclear bond
small and distance of a diatomic oscillator.
ae = Be (2.58)
100
Consequently, the value of B, D, and r will be different in different vibrational states. B0, D0 and r0 thus cor-
respond to the ground vibrational state, i.e. V = 0 while Be, De and re stand for equilibrium state. Be, De and re can
only be determined theoretically but not experimentally. In the light of the above facts, the energy of the allowed
levels of a nonrigid rotator in the higher vibrational states should be expressed as
⎛ 1⎞
EJ = hcBJ(J + 1) − hcDJ2(J + 1)2−ae ⎜V + ⎟ (2.59)
⎝ 2⎠
The higher terms being smaller are ignored. Quantum mechanical derivation of this equation is given in Appen-
dix 3.3A of infrared spectroscopy.
2.5.1 Spacing Between Two Adjacent Allowed Energy States

The spacing between two successive energy levels, ΔdJ , may be deduced from Eq. (2.52) as follows.
EJ = hcB J (J+1) − hcD J2(J+1)2 (Eq. 2.52)

EJ+1 = hcB (J + 1)(J+2)− hcD (J+1)2 (J+2)2 (2.60)
Proceeding as in the case of a rigid rotator, we get from Eqs (2.52) and (2.60)
ΔdJ = E J EJ ΔErot
= = 2B(J+1) − 4D(J+1) (2.61)
3
hc hc
From Eq. (2.61), we obtain the spacing between the two adjacent levels and is recorded in tabular form as under.
J Spacing, D dJ
0 2(B − 2D)
1 4 (B − 8D)
2 6(B − 18D)
… …………
The results indicate that the spacing between the adjacent energy levels in comparison to rigid rotator model of
the diatomic molecule (where D = 0) is small. However, as expected, the spacing grows with the quantum number
J. The spacing between the adjacent levels, i.e. 2(B − 2D), 4(B − 8D), 6(B−18D),..., as well as the gap between
the adjacent spacings, i.e. 2(B − 14D), 2(B − 38D), 2(B − 74D),..., are not uniform as in the case of a rigid rotator
model of the diatomic molecule. The non-uniform behaviour may be due to centrifugal distortion which increases
as the rotational quantum number increases.
2.5.2 Transition Between Two Adjacent Energy Levels

The selection rules for a nonrigid rotator model are the same as for the rigid rotator model of the diatomic molecule,
i.e. M ≠ 0, Mt ≠ 0, and ΔJ = ±1. Let J ″ and J ′ be the rotational quantum numbers for the upper and lower energy
levels respectively. From Eq. (2.52), the energy expressions for the two states may be written as
EJ hcBJ ″ h D ″ 2 (J ″ +1
hcDJ
DJ + )2 (2.62)
2
EJ hcBJ ′ hhcDJ
D ′ (J ′ + 1)
DJ (2.63)
From Eqs (2.62) and (2.63), we obtain

EJ EJ ′ ΔE rot
v rot = =
hc hc
= B [(J″ − J′)2 + (J″−J′)] −D[(J″ + 1)−J′(J′+ l)][J″ (J″ + 1) + J′ (J′ + 1)]
= (J″− J′) (J″+J′ + 1) [B − D{J″(J″+ l) + J′ (J′+ l)}] (2.64)

Applying the selection rule, ΔJ = J″− J′ = + 1, Eq. (2.64) reduces to
vrot = 2BJ″−4DJ″3 (2.65)
ν rot
= 2B − 4DJ″2 (2.66)
J″
where J ′′ = 1, 2, 3,...
In terms of lower level, we may write
v rot
= 2B − 4D (J′+1)2 (2.67)
(J ′ + 1)
where J′ = 0, 1, 2, 3, ….
v rot v rot
The plot between and J″2/(J′+1)2 will be a straight line with slope = −4D and intercept = 2B.
J ′′ (J ′ + )
From slope and intercept, one can determine D and B and hence the bond length of the diatomic molecule.
The energies of the different transitions as worked out from Eq. (2.65) or (2.67) have been recorded in the
tabular form as under:
J Transition vrot ’ cm-1
0 0→1 2B − 4D
1 l→2 4B − 32D
2 2→3 6B − 108D
3 3→4 8B − 256D
Thus, according to nonrigid rotator model of the diatomic molecule, just like rotational energy levels, there
is also a decrease in the separation between the spectral lines as compared to the rigid model. The energy-level
scheme and the lines in the spectrum of rigid and nonrigid diatomic molecules are presented in Fig. 2.13.
Problem 2.14: The following lines have been identified in the high-resolution rotational spectrum of CO.
Line No. 1 2 3 4 5
J 0 1 2 3 4
v rot , cm−1 3.845 7.690 11.534 15.379 19.222
Determine analytically and graphically the values of ro and D0 for the molecule if the vibrational frequency of CO
is at 2300 cm−1. Analyse the data by the method of least-square fit also.
8B
8B – 256D
6B
6B – 108D
4B 4B – 32D
1
2B 2B – 4D
0
2– 4– 6– 8– 10B ν (cm−1) 2B – 4D 8B – 256D

(a) (b)
Fig. 2.13 The change in rotational energy levels and spectrum of a diatomic molecule in passing from
(a) a rigid to (b) a nonrigid rotator.
Solution Assuming the molecule to be in the ground vibrational state, the value of the B0 can be determined
from the average value of the difference of wave numbers between the adjacent lines. The average value of 2B0 =
3.844 cm−1. Hence B0 = 1.922 cm−1.
12 × 16 192
μm = = amu
12 + 16 28
Substituting B0 and mm into Eq. (2.34), we get
16.8575
ro = [Since I = mm r2]
μm B
= 16.8575 × 28
= 1.131Å
192 × 1.922
From Eq. (2.53),
4 (1.922)3
D0 = = 5 97 10 −6 cm−1
( 2300) 2
Graphically, the values of B0 and D0 can be obtained with the aid of Eq. (2.67), i.e.
v rot
= 2B 4 D (J ′ )2
(J ′ )
where J' = 0, 1, 2, 3….
v
The data to be employed in the plot between rot and (J' + 1)2 are generated and recorded in the tabular form
as follows: (J ′ )
The plot is shown in Fig. 2.14.

J′ (J′ + 1) (J′ + 1)2 v rot

, cm−1
( J ′ 1)
0 1 1 3.845
1 2 4 3.845
2 3 9 3.844
3 4 16 3.845
4 5 25 3.843
3.845
3.844
n–rot
3.843
(J ′ + 1)
3.842
3.841
3.840
0 2.5 5.0 7.5 10 12.5 15 17.5 20 22.5 25 27.5
(J ′ + 1)2
Fig. 2.14 Plot of v rot /( J ′ + 1) against ( J′ + 1)2.
From the graph,
Intercept 3.8452
B0 = = = 1.9226 cm−1 and
2 2
Slope 8.8 × 10 −5
D0 = = = 2.2 × 10 −5 cm
4 4
16.8575 × 28
Therefore, r0 = = 1.1307Å
1.922 × 192
Since the points on the graph are scattered, the results of graphical method can be improved upon by the
method of least-square fit rather than manual fitting. The method of least-square fit has already been discussed
in detail in Chapter 1. The data needed for the method are generated and reported in the tabular form as under:
v rot
J′ (J′+1) (J′+1)2 v rot /(J′+1),cm−1 (J′+1)4 (J′+1)2 ,cm−1
(x) (y) (x2) ( J ′ + 1)
(xy)
0 1 1 3.845 1 3.845
1 2 4 3.845 16 15.380
2 3 9 3.844 81 34.596
3 4 16 3.845 256 61.520
4 5 25 3.843 625 96.075
5
∑ = 55
i =1
19.222 979 211.4160
Using normal equations from Chapter 1, we get
10B0 −220D0 = 19.222

110B0 −3916D0 = 211.416
Solving for B0 and D0, we get
B0 = 1.922 and D0 = 1.73 × 10−5 cm−1
16.8575 × 28
Thus, r0 = = 1.131 Å
192 × 1.922
2.6 HYPERFINE STRUCTURE

The rotating molecule has a magnetic field associated with it and is normal to the axis of the molecule. The mag-
netic moments of the rotating nuclei perturb slightly the rotational energy levels which, in turn, result in splitting
of the rotational lines in the spectrum. This splitting is due to the interaction between the rotational motion of the
molecule and the motion of the electrons. In the light of this, the rotational quantum number, J, consists of two
numbers—one characterising the motion of the nuclei and the other, the motion of the electrons. J for the three
Hund’s cases framed from the vector model of diatomic molecule may be expressed as
(i) J J nnuc + Ω ⎫
⎪ (2.68)
(ii) J J nnuc + Λ + S = K +S ⎬
(iii) J J nnuc + Ω ⎪
⎭
2 2
The squares of the vectors J and K are equal to ⎛⎜ h ⎞⎟ J(J+1) and ⎛⎜ h ⎞⎟ K(K+1), respectively, whereas the
⎝ 2π ⎠ ⎝ 2π ⎠ 2
squares of Ω and Λ, which are projections of vectors L+ S and L of the molecule, are equal to ⎛⎜ h ⋅ Ω⎞⎟ and
2
⎛ h ⎞ respectively. ⎝ 2π ⎠
⎜⎝ Λ ⎟
2π ⎠
Accordingly, the energy of the rotational levels is expressed as
Bhcc ⎡⎣ J ( J + ) − Ω 2 ⎤⎦ ⎫
(i) E J = Bh
⎪
⎪
Bhcc ⎡⎣ K ( K + ) − Λ2 ⎤⎦ ⎬
(ii) E J = Bh
⎪ (2.69)
(iii) E J Bhc ⎡⎣ J ( J ) − Ω 2 ⎤⎦ ⎪⎭
where Ω = Λ ± ∑. A comparison of this expression with J = L + S indicates that Ω, Λ and ∑ are respectively
related to the numbers J, L and S where J, L and S are the total, orbital and spin angular momenta respectively. Ω,
Λ and ± ∑ are the projections of total, orbital and spin angular moments respectively on the molecular axis. The
quantum numbers Λ and ∑ characterise the orbital and spin motion of the electrons respectively. For every value
of S, there are 2S + 1 values of ∑, given by
∑ = Λ + S, Λ + S−l,…, Λ −S (2.70)
where Λ = 0, 1, 2, 3…..
The corresponding electronic molecular terms are designated as
∑, Π, Φ…..
The quantum numbers for the various molecular electronic terms are as follows:
For ∑ terms Σ( 0) Σ (S 1Λ 0) , 2 Σ ( 0) Σ (S 1 Λ = 0) ;
2 3
0 1 2
2.6.1 Designation of Molecular Orbital Corresponding to the Value of Λ

For one electron—Total: l = 0 (s); Λ = 0 (∑); l = 1 (π); Λ = 1 (Π); l = 2 (d); Λ = 2 (Δ). This is similar to
the designation of atomic orbitals (s, p, d, f…., for l = 0, 1, 2, 3…..) and total angular momentum (S, P, D, F…
for L = 0, 1, 2, 3……)
∑
−
Various Symbols in the Notation u
:
(i) ∑ corresponds to electronic state for which the orbital or electronic angular momentum Λ is zero.
(ii) 2 represents, the multiplicity of the state, i.e. 2S + 1 = 2, ∴ S = 1/2
(iii) Negative sign indicates that electronic wave function of a molecule changes sign on reflection through
a symmetry plane.
(iv) When the electronic wave function of a molecule changes sign on reflection through centre of symme-
try, it is termed as ungerade (u).
For Π terms: 2 Π ( S 1 2, Λ 1) , 1
(S 0, Λ = 1) ,3 (S 1, Λ = 1) ;
For Δ terms: 2Δ ( S / ,Λ ) , 1 Δ ( S = 0, Λ = 2 ) .
The complete molecular electronic terms can be worked out from the formula
2S 1
ΛΩ (2.71)
3
e.g. for Λ = 1, S = , the complete term may be worked out with the aid of Eqs (2.70) and (2.71) as follows:
2
5 3 1 −1 3
Ω= , , , and 2S+1 = 2 × + 1 = 4 .
2 2 2 2 2
Thus, the term 4Π : 4Π5/2, 4Π3/2, 4Π1/2, and 4Π−1/2,
Similarly, we can show that the components of 4Δ state are 4Δ7/2, 4Δ5/2, 4Δ3/2, 4Δ1/2.
The selection rules for the transition between the energy levels expressed by the expression (2.68) are
Δ Λ = 0, ± 1 (2.72)
ΔΩ = 0, ± 1 (2.73)
Δ∑ = 0 (2.74)
ΔS = 0 (2.75)
According to Rule (2.72), ∑ ↔ ∑, Π ↔ Π, Δ ↔ Δ, ∑ ↔ Π, Π ↔ Δ combinations are allowed while ∑ ↔ Δ
and Π ↔ Φ combinations are not allowed.
By Rule (2.73), 0 ↔ 0, 1 ↔ 1, 2 ↔ 2, 0 ↔ 1, 1 ↔ 2 combinations are permissible while 0 ↔ 2, 1 ↔ 3 are
forbidden.
Rule (2.74) is helpful in understanding the peculiar behaviour of fine structure of molecular spectra (Hund's
case ii). By Rule (2.75), intercombinations, i.e. combinations between terms of different multiplicities, are not
allowed.
Further, according to the important rule associated with the kind of symmetry of molecular terms, positive
terms (+) combine with negative (−) and even terms (g) combine only with odd terms (u), e.g. we may have the
rule ΔS = 0, g ↔ u and the like. Here g stands for gerade and u for ungerade.
Equation (2.69) is valid under the following conditions:
For EJ ≥ 0, J ≥ Ω, and K ≥ Λ must be satisfied. Thus, J = 0 (in the case of integral J) and K = 0 are possible
only if Ω = 0 and in turn Λ = 0. When Ω > 0, Λ > 0, the minimum values of J and K are determined by the values
of Ω and Λ.
Now coming to the fine structure which results due to the interaction of molecular rotation and the motion of
the electrons, it is found that the magnitude of splitting increases with increase in quantum number J (or K). This
increase is due to a build up of the intramolecular magnetic field with rotation of the molecule. The number of
sublevels into which a particular rotational level splits is given by 2S + 1.
The rotational level in the case of level 1Σ shows no splitting since in this case Λ = 0 and S = 0.
For the level Σ , the rotational level splits into two components since S = l/2 while into three components in
2
case of level 3Σ since S = 1. Further when Λ > 0, the rotational level splits up into two sublevels and the splitting
is independent of the multiplicities of the Π terms, i.e. 1 Π, 2 Π and 3 Π .However, the splitting in this case is due
to the interaction between Jnuc and Λ but not between Jnuc and S. The levels corresponding to Λ > 0 are doubly
degenerate since the same energy corresponds to the values of + Λ and − Λ. For this reason, when there is intra-
molecular magnetic field due to the molecular rotation, each such level splits into two components (+ and −) as
shown in Fig. 2.15. This type of splitting is termed as Λ type doubling.
In case of Π-, Δ-, and Φ-levels, there is strong interaction between S and Λ for small values of J. The splitting
in this case is independent of rotation. However, as J grows, the (S, Jnue) type interaction becomes stronger, con-
sequently splitted components become closer.
The properties of symmetry of the rotational terms as deduced from the wave function Ψ = ± Ψelec ± Ψrot that char-
acterises the rotational term belonging to corresponding electronic level of the molecule, have been recorded as
Ψelec J Ψrot Ψelec J Ψrot
+ Even + − Even −
Odd − Odd +
Thus, in the case of Σ terms, the corresponding rotational terms will alternately be positive and negative as
shown in Fig. 2.15 (a). In case of Π-, Δ-, Φ-, etc., terms, one of the components of Λ-type doubling of each rota-
tional level will be (+) and the other (−) and is shown in Fig. 2.15 (b).
In case of molecules having same nuclei, the symmetry properties are also summarised in the tabular form as
+
follows and are shown for ∑ - states in Fig. 2.16 (a) and (b).
(a) (b)
J −
− 5
(a) J (b)
+
−a 5 −s
+
+ 4 +s 4 +a
−
−a 3 −s
−
−
3 +s 2 +a
+
+ −a 1 −s
+ +s 0 +a
2
− Σ+g Σ+u
− −
1
+ + Fig. 2.16 Properties of symmetry of rotational terms of the iden-
0 +
Σ+(Δ = 0) Π(Δ = 1) tical nuclei in the case of Σ-states (Λ = 0) (a)
+
∑ − term is positive
Fig. 2.15 Positive and negative rotational terms in case of
(a) ∑ +, and (b) Π electronic terms. and symmetric, and (b) ∑ − term is negative and symmetric.
Ψelec J Ψrot Ψetec J Ψrot
Even (g) Even +s Odd (u) Even +a
Odd −a Odd −s
The components of Λ-type doubling Π-, Δ- Φ-, etc, terms have different symmetries (s and a). The selection
rules for J are as follows:
(i) For Λ ≠ 0, ΔJ = 0, ± 1 and for Λ = 0, ΔJ = ± 1.
(ii) The selection rules for combining positive and negative rotational terms are + ↔ − and − ↔ +.
This rule does not contradict the rule ΔJ = ± 1 (Λ = 0).
(iii) For Raman spectrum, the rule for the transition between positive and negative rotational terms are + ↔ +
and − ↔ − and is in agreement with the rule ΔJ = 0, ± 2 for Raman scattering of radiation (Chapter 4).
In case of molecules with identical nuclei, only symmetric–symmetric and antisymmetric–antisymmetric com-
binations of terms are possible: s↔s and a ↔ a. Hence according to this rule, molecules such as H2, N2, O2 cannot
exhibit rotational spectrum since only the transitions +s ↔ −a are permissible.
2.7 ROTATIONAL SPECTRA OF POLYATOMIC MOLECULES

Based on the moments of inertia of molecules in the three principal axes, the molecules have been classified as
follows:
(a) Linear Polyatomic Molecules of the type XY2 and XYZ, etc., have two rotational degrees of freedom
about axes perpendicular to that of the molecule. In such molecules, Ix = Iy , Iz = 0, e.g. CO2, CS2, OCS,
HCN, SCSe, etc. On the other hand, nonlinear molecules have three rotational degrees of freedom.
(b) Spherical Top Molecules of this type have several axes of symmetry of the order n ≥ 3. The moments
of inertia of these type of molecules about all axes passing through the centre of mass are exactly equal,
i.e. Ix = Iy = Iz, e.g. CH4, CCl4, SF6, P4, etc.
(c) Asymmetric Top Molecules of this type lack axes of symmetry of the order n ≥ 3. In such cases, Ix ≠ Iy ≠ Iz ,
e.g. H2O, CH2C12, etc.
(d) Symmetric Top In these type of molecules, two of the three moments of inertia are equal to each other, i.e.
I x = Iy ≠ Iz or Ix ≠ Iy = Iz. Such molecules have further been subdivided into two classes:
(i) Oblate top: Ix >Iy = Iz, e.g. BCl3, NH3, C6H6, etc.
(ii) Prolate top: Ix < Iy = Iz, e.g. CH3Cl, CH3F, etc.
Ix, ly and Iz are also denoted by IA, IB and IC, respectively.
2.7.1 Moments-of-inertia Defect

As will be discussed in Chapter 3, each vibrational level, say V, is associated with rotational levels. The effective
moments of inertia I A(V ) , I B(V ) and I C(V ) are obtained from the respective effective rotational constants A(V ) , B(V ) ,
C(V ) . The average values of rotational constants, A, B, and C but not average values of I , I and I matter in the
A B C
rotational energy states of the vibrating molecule. At equilibrium, the relation
I C(e) = I A(e) + I B(e)

holds good and to a fair degree of approximation, and
I C(0) I A(0) + I B(0) (2.76)
also holds good.
But this relation does not, in general, hold good for effective moments of inertia because in this case
h 1 h 1 h 1
= , = and =
8 π A c (1 I A )average
2
8π Bc (1 / I B )average
2
8π Cc (1 / I C )average
2
1
since I average ≠ (2.77)
(1/ I )average
Consequently, the following relation results in
Δ = I C(V ) − I A(V ) − I B(V ) (2.78)
where Δ is termed as moments-of-inertia defect. This defect is prevalent in nonlinear triatomic molecules such
as H2S, H2O, etc. This defect was first of all observed in H2O and exists even for the lowest vibrational state. The
moments-of-inertia defect enhances as V increases, e.g. in case of of cyanogen isocyanate, the values of IA, IB, and
IC are 6.7966633, 187.24865 and 194.54080 amu Å2 respectively in the ground state, while they are 6.1053922,
186.88268 and 194.39776 amu Å2 for the first excited state. The inertial defects in the ground and first excited
2
vibrational states are 0.49548 and 1.40969 amu Å respectively.
1
For nearly symmetrical top molecules, the B value obtained from (||) bands is B ( B + C ) . For plane molecules
2
such as N3H, C2H4, the relation IC = IA+ IB holds good for the equilibrium moments of inertia and also for ground–state
moments of inertia under very good approximations. Thus if A is known, B and C can be determined separately from
B. The following relations are easily found.
B ( A − B) A2 + B 2
(2.79)
C ( A + B) A2 + B 2 (2.80)
For nonplanar molecules, B and C cannot be obtained from B.
2.7.2 Rotational Spectra of Linear Polyatomic Molecules

The expression for the rotational energy and hence the spectrum of linear polyatomics is similar to diatomic molecules.
Further, the ground states of all linear polyatomic molecules are Σ states. For Λ ≠ 0, the rotational energy of these
types of molecules is expressed by Eqs (2.68) (i) and (ii). Further, if the linear polyatomic molecules have a nuclei
with quadrupole moment, i.e. HCN, the rotational levels will split up into sublevels as a result of interaction between
rotational motion and nuclear spin of the quadrupole nuclei of the molecule. The perturbation of energy levels results
in the splitting of rotational lines in the spectrum. The energy of a perturbed level is given by
⎡3
⎢ C (C + ) − I ( I + ) J ( J + ) ⎤⎥
EJ , FJ B ( J + ) + eQq ⎢ 4
BJ ⎥ (2.81)
⎢ 2I ( I + ) 2 J ( J + ) ⎥
⎣ ⎦
This equation is valid for J ≥ I and I and J ≠ 0. Here, C = FJ (FJ + 1) −I (I + 1) −J (J + 1) and FJ = J + I,

J + I − 1, 0, …, J I .
h
The resultant angular momentum, FJ ( FJ + 1) is quantised in gas phase only while in solids the nuclear
2π
spin angular momentum is quantised with respect to the electric field. The splitting of energy levels is governed by
the (2I + 1) rule if J >I and (2J + 1) if I >J. Thus, except for J = 0, higher levels will split up into three components
as shown in Fig. 2.17.
(a) (b) FJ
J 4
3 3
2
3
2 2
1
2
1 1
0
0 1
Fig. 2.17 Splitting of allowed rotational energy levels by nuclear quadrupole coupling
field: (a) the energy levels are without quadrupole field, and (b) the energy levels with qua-
drupole field, I = 1. The dark arrows are for ΔFj = +1 while the broken arrows are for ΔFj = 0.
In other words, we can say that for J ≠ 0, the splitting of energy levels is independent of the rotational motion
of the molecule, i.e. J. The transition between the energy levels occurs according to the selection rule
ΔJ = ±1, ΔFj = 0, ±1 (2.82)
From Eq. (2.81), it is evident that we can determine the nuclear quadrupole coupling constant eqQ from micro-
wave spectroscopy.
Let us now work out the quadrupole splitting for the J = 1 → 2 transition. The allowed transitions between the
sublevels according to rule ΔJ = ±1 and ΔFJ = 0, ±1 are also presented in Fig. 2.17. Applying Eq. (2.81) for two
sets of quantum numbers say (J, Fj′), and (J+ 1, Fj″), we get
ΔErot E J 1,FJF ′′ E J ,FJ ′

vrot =
hc
=
hc
= 4 B eqQ X J ( FJ X J FJ ) (2.83)
Here X is the coefficient of Q of Eq. (2.81).
The coefficients of eqQ for J = 1→ 2 transition as determined from Eq. (2.83) are recorded in the tabular form
as follows:
State Coefficients of eqQ

J = l, I = 1 F=2 +1/72
F=l −5/72
F=0 +5/36
J = 2, I = 1 F=3 +1/40
F=2 −7/80
F=1 +7/80
From Eq. (2.83), we write, for J = 1→ 2, I = 1
vrot ( F = → ) = 4B 0.101 eqQ ⎫

⎪
vrot ( F = → ) = 4B 0.156 eqQ ⎪
⎪
vrot ( F = 2 → 3) = 4 + 0.011 eqQ ⎬ (2.84)
vrot ( F = 1 → 2) = 4 − 0.018 eqQ ⎪
⎪
vrot ( = → )= 4 − 0.150 eqQ ⎪⎭
Thus, from Eq. (2.84), magnitude as well as the sign of eqQ can be determined.
Problem 2.15: (a) The moment of inertia of CO2 as determined from its rotational Raman spectrum is 43.217 au.
Determine the bond length of CO2.
(b) The average value of separation between the adjacent rotational lines as deduced from the ultraviolet emission
+ +
spectrum of ionised carbon dioxide CO 2 in the ground state is 0.7612 cm−1. Determine the bond distance of (CO 2) if
the ground state is 0.7612 cm−1. Determine the bond distance of CO +2 in the ground state.
Solution (a) The geometrical configuration of CO2 is O = C = O. From Eq. (2.3), we write
2
I ∑m r
i =1
i i
2
= mO rC2O + mO rCO
2
= 2m 2
mO rCCO × 16 rC2O = 32rC2O
Since I = 43.217 au
43.217
rCO = = 1.1621Å
32
(b) Since 2B0 = 0.7612 cm−1. Therefore, B0 = 0.3806 cm−1. With the aid of Eq. (2.34), we obtain
16.8575
I= = 44.292 au
0.3806
44.292
r= = 1.1765Å
32
Problem 2.16: Determine the moment of inertia of 13CS2 if the bond length of 12CS2 is 1.2151 Å.
Solution We know that isotopic substitution does not affect the bond length of molecule. The geometrical
configuration of 13CS2 is
S = 13C = S
Given rcs = 1.2151 Å, from Eq. (2.3), we obtain

2
I ∑m r
i =1
i i
2
= ms rcs2 + ms rcs2 = 2 × 32 × (1.2151) 2 = 94.49 au
The moments of inertia of 12CS2 and 13CS2 are the same since in this case isotopic substitution does not alter the
centre of mass of the molecule.
2.7.3 Structural Parameters of Linear Triatomics of the Type XYZ

As we already know that the rotational constant varies inversely with moment of inertia, consequently the spac-
ing between adjacent lines in the rotational spectra will decrease on isotopic substitution. Isotopic substitution
also does not affect the bond lengths of molecules. Therefore, in order to determine the bond lengths of XYZ type
molecules, the moments of inertia of some isotopes of XYZ should also be known, e.g. if we want to determine the
bond distances of HCN, the moments of inertia of its isotopes, viz. DCN or HCN15, should also be known. This
method is illustrated by taking the example of OCS in the next problem.
Problem 2.17: (a) The microwave absorption spectrum of OCS shows the following absorption lines:
Line No. 1 2 3 4
ν rot cm−1 0.81143 1.21714 1.26686 2.02886
Determine the bond lengths of OCS if the moment of inertia of 18OCS as determined from its microwave absorption
spectrum is 88.659 au.
(b) Determine the bond lengths of C2H2 if the rotational constants for C2H2 and C2HD as determined from the
rotational Raman spectra are 1.1838 and 0.9967 cm−1 respectively.
Solution (a) Since OCS is a linear molecule, there are only two rotational degrees of freedom. Hence Ix = Iy
and Iz = 0, The average of difference between the wave numbers of adjacent lines in the spectrum is
2B = 0.40581 cm−1, Hence B = 0.2029 cm−1
Using Eq. (2.34) we get
16.8575
I( ) . au
0.2029
I*(18OCS) = 88.65 au.
The bond lengths of OCS can now be determined with the aid of Eq. (2.14). For OCS and 18OCS, we may write
I( ) 2
mO rCO + mS rCS2 −
(mO rCO − mS rCS )2 (2.85)
Mm
(m * )
2
O r CO − m S rCS (2.86)
I * (OCS) m *O rC2O + m S rCS2 −
Mm + 2
( )
2
(mO rCO + mS rCS ) 2 m O*rCO + m S rCS

ΔI = I * −I = rCO
2
( m O* − m O ) + − (2.87)
Mm Mm + 2
Combine Eqs (2.87) and (2.85) in order to eliminate the term rco rcs. The resultant equation will then be in terms
of rco2 and rcs2 . Substitute the value of rcs2 in terms of rco2 into Eq. (2.85) to get rco. Once rco is obtained, rcs can easily
be determined. rco = 1.163 Å and rcs = 1.558 Å.
(b) From Eq. (2.34), we get
16.8575
I= = 14.240 au
1.1838
16.8575
I* = =16.913 au
0.9967
M m* = 2 × 12 + 2+1 = 27 amu
Mm = 24 + 1 +1 = 26 amu
With the aid of Eq. (2.29), we obtain,
1 27(16.913 − 14.240)
x= r +r = = 1.666 Å
2 CC CH 26( 2 − 1)
1 14.240 − 2 × 1 × (0.601) 2
y rCC = = 0.601Å
2 2 × 12
Therefore, rcc = 1.202 Å and rCH = 1.065 Å.
Problem 2.18: The spectroscopically measured values of moments of inertia of the molecules HCN and DCN
are 11.403 and 13.947 au respectively. Determine the bond lengths of HCN.
Solution We know that bond length is independent of isotopic substitution. Thus proceeding as in Problem
2.17 (a), it is found that rCH = 1.058 Å and rCN = 1.157 Å.
Problem 2.19: The wave numbers in the far infrared rotational spectrum of HCN are given below:
J 0 1 2 3 20 22
vrot,cm−1 2.96 5 .92 8.87 11.83 61.98 67.85
Determine the values of B and D by the method of least-squares fit.
Solution B and D can be determined with the aid of Eq. (2.67), i.e.
vrot
= 2 B 4 D( J ′ ) 2 , where J´= 0, 1, 2, 3,...
(J ′ )
vrot vrot
J J′ 1 ( ′ )2 cm−1 ( ′ )4 ( ′ )2 × , cm−1
( J ′ 1) ( ′ )
0 1 1 2.960 1 2.960
1 2 4 2.960 16 11.840
2 3 9 2.956 81 26.604
3 4 16 2.957 256 47.312
20 21 441 2.951 194481 1301.391
22 23 529 2.950 279841 1560.550

6
∑
i =1
1000 17.734 474676 2950.657
The data required for the method of least-squares fit is recorded in the tabular form as above:
Using normal Eq. (1.58) we get
12B0− 4000 D0 = 17.734
2000B0 − 1898704 D0 = 2950.657
Solving for B0 and D0, we obtain

B0 = 1.479 cm−1 and D0 = 4.06 × 10−6 cm−1
2.7.4 Rotational Spectra of Spherical-Top Molecules

The rotational energy levels of nonlinear polyatomic molecules in terms of angular momenta and moments of
inertia are expressed as
Px2 Py2 Pz2
EJ = + + (2.88)
2I x 2I y 2I z
where Px, Py, and Pz are the angular momenta about the three principal axes. According to quantum mechanics the
angular momentum, is quantised, i.e.
h h
P= PJ = J (J + 1) (2.89)
2π 2π
where PJ is a quantum number and can take up values −J to +J including 0. Thus, PJ can have (2J + 1) values. In
case of spherical tops, Ix = Iy = Iz, consequently Eq. (2.88) reduces to,
EJ =
1
2I
(
Px Py Pz = ) P2
2I
(2.90)
(
where P 2 = Px2 + Py2 )
Pz2 .
Combining Eqs (2.89) and (2.90), we obtain
h2
EJ = J (J + 1) = BhcJ
Bh J (J + 1) (2.91)
8π 2 I
Thus, the energy of the rotational level of rigid spherical tops is the same as for diatomics. For a nonrigid spheri-
cal top, we have to add the term − BhcDJ 2(J + 1)2 to Eq. (2.91). Consequently, just like diatomics, the rotational
spectrum of polyatomic spherical tops should consist of equally spaced lines with spacing between the adjacent
lines in the spectrum of the order of 2B. Further, since these types of molecules have three degrees of freedom, a
third quantum number, K, characterising the angular momentum about one of the symmetry axes rotating together
with the molecule needs to be introduced such that
⎡ h ⎤
PK = ⎢ ⎥ K (2.92)
⎣ 2π ⎦
Here K = J, J − 1, J − 2,..., 0,...,..., − J and can have (2J + 1) values. The degeneracy of the energy level in such
molecules is then given by
gJ = (2J + 1) · (2J + 1) = (2J + 1)2 (2.93)
The population of molecules among the energy states is then given by

−hh 2 J ( J + )
πT (2.94)
NJ ( J + ) N 0e
2 8π 2 I
−hh 2 J ( J + )
NJ ( J + ) πT
= e 8π 2 I
NA σ Z rot
It is to be remembered that the spherical top molecules are microwave inactive since their dipole moment is zero,
i.e. M = 0
2.7.5 Rotational Spectra of Symmetric-Top Molecules

We know that symmetric top molecules are of two types, i.e. prolate symmetric top and oblate symmetric-top. In
case of prolate symmetric-top molecules, Ix < Iy = Iz. Therefore, for such molecules, Eq. (2.88) transforms to
Px2
EJ = +
1
P
2I x 2I y y
( Pz )
Px2 ⎛ 1 1 ⎞
= ⎜ − ⎟+
1
2 ⎝ I x I y ⎠ 2I y
( + + ) (2.95)
Px2 ⎛ 1 1 ⎞ P 2
= − +
2 ⎜⎝ I x I y ⎟⎠ 2 I y
Just like spherical top molecules, symmetric tops have also three degrees of freedom. Consequently, in addition
to the quantum number J, the quantum number K, characterising the motion of the molecule about the symmetry
axis passing through the centre of mass of the molecule is introduced such that
⎛ h⎞
PK = ⎜ ⎟ K (Eq. 2.92)
⎝ 2π ⎠
Relation between J and K in case of symmetric tops is presented in Fig. 2.18. From Eqs (2.95), (2.92) and (2.89),
we obtain the energy of the rotational levels of the prolate symmetric-top molecules, i.e.
J
⎛ h2 h2 ⎞ 2 h2
EJ ,K =⎜ 2 − K + J (J ) (2.96)
⎝ 8π I x 8π 2I y ⎟⎠ 8π 2 I y
EJ ,K (A B )hcK 2 + Bhc J (J
(J ) (2.97)
K
h h
where A= ;B= 2 (2.98)
8π I x c
2
8π I y c Fig. 2.18 Relation between J and K for a symmetric-top
molecule.
Proceeding as above, we can work out the energy of the rotational levels of oblate symmetric top molecules.
The quantized energy states for such molecules are given by
EJ,K = Bhc J(J + 1) + (C − B)K2 (2.99)

where
h
C= (2.100)
8π 2 I z c
The effect of axial rotation on the moment of inertia of symmetric top molecules XY3Z is presented in Fig. 2.19.
Since in Eqs. (2.97) and (2.99), the K2 term is involved, the clockwise and anticlockwise rotations about
the symmetry axis do not affect the results. The degeneracy of each level will be equal to 2(2J +1) except
for the zeroth level where K = 0 and the degeneracy is (2J + 1) folds. Since K2 terms are involved, for K > 0,
each line will split up into (J + 1) components, i.e. same energy corresponds to their levels + K and −K and
will be doubly degenerate.
2.7.6 Transition Between Two Adjacent Energy States (Prolate/Oblate

Tops as Rigid Rotators)
Let J′, K′ and J′′, K′′ be the quantum numbers for the lower and upper levels respectively. From Eq (2.97), we may
write for the two levels,
EJ , K = BhcJ
c ′ (J
(J ′ + ) + hc ( A − B ) K ′ 2 (2.101)
EJ ,K = BhcJ
c ″ (J
(J ″ + ) + hc ( A − B )K ″ 2 (2.102)
Combining Eqs (2.101) and (2.102), we obtain Y
ΔE rot EJ ,K E J ′, K ′ (2.103)
(a)
v rot = = Y X Z K=0
hc hc
= B( J ′′ J ′ )(J
)( J ′′ J ′ ) ( A B )( K ′′ K ′ )(K
)( K ′′ K ′ )
Y
Applying selection rules, i.e. ΔJ = ±1, ΔK = 0, we obtain
vrot = 2B (J′ + 1) (2.104) Y
where J ′ = 0, 1, 2, 3. (b)
The selection rule of ΔK arises because there is no dipole moment

about the symmetry axis (rotation about symmetry axis is denoted Y X Z K<0
by K), hence radiation cannot interact with the rotations about this
axis. Consequently, it is expressed as ΔK = 0.
From Eq. (2.104), it is clear that the rotational spectrum of Y
prolate or oblate top molecules is similar to linear, di- and poly-
atomic molecules and is independent of K, i.e. the axial rotation
does not affect the dipole moment normal to the axis which is
always zero. Y
(c)
Y X Z K>0
2.7.7 Nonrigid Rotator Model of Prolate/
Oblate Symmetric-Top Molecules
Y
Let us now work out the microwave spectrum of symmetric top mol-
ecules taking into consideration the effect of centrifugal force. The Fig. 2.19 The effect of axial rotation on the moment
energy of the rotational levels as deduced quantum mechanically is of inertia of a symmetric top molecule: (a) there is no
expressed as axial rotation (K = 0), (b) there is clockwise (K >0), and
(c) anticlockwise axial rotation (K >0).
EJ,K = Bhc J (J+1) + (A−B) hc K2 − DJ hc J 2 (J +1)2 −DJK J (J+1)K2 − DK K4 (2.105)
where DJ , DJK and DK are the centrifugal distortion constants arising due to different types of motions.
Now for two energy states, we write
EJ′, K′ = Bhc J′ (J′ + 1) + (A − B)hc K′2 (2.106)
− DJhc J′2 (J′ + 1)2 − DJK J′(J′ + 1) K′2
EJ″, K″ = Bhc J″ (J″ + 1) + (A − B)hc K″2 (2.107)
− DJhc J″2 (J″ +1 )2 − DJK J″(J″ + 1) K″2
Combining Eqs. (2.106) and (2.107), we obtain
ΔE rot EJ EJ ′ ,K ′
v rot = =
,K
hc hc
= B(J″− J′)(J″+J′+ 1) + (A − B) (K″2−K′2) −DJ [J″(J″+1) −J′(J +1)[J″(J″ + 1)+J′(J′+1)]
−DJK [ ( ′′ ′′ ′ ′ )(( ′′ ′′ ′ ′ ) + ( ′′ ′′ 2 ′ ′ 2 )] (2.108)
= ( ′′ − ′ )( ′′′ + ′ + )[
)[ − ){{ ′′(J ′′ + 1) + ′(J ′ + 1)}]
+ (A − B) (K″2 - K′2)− DJK [(J″ K″ − J′K′)(J″K″ + J′K′)+( J″K″2 − J′K′2)]
Applying selection rules, i.e. ΔJ = J″ − J′ = + 1, ΔK = K" −K′ = 0, we get

ν rot = 2 J ′′ − 4 DJ J ′′ − 2DJK J ′′′K ″ cm −1 (2.109)
or vrot = 2B(J′+ 1) − 4D (J′+ 1)3 − 2D (J′+ 1)K′2, cm−1 (2.110)

J JK
Equations (2.109) or (2.110) reveal that rotational spectrum now depends upon K although the axial rotation
does not change the dipole moment perpendicular to the axis. The axial rotation actually increases the YXY angle
as well as the X–Y bond length as shown in Figs 2.19 (b) and (c). Consequently, the moment of inertia changes
and hence spectrum becomes dependent on K. The allowed energy levels of prolate symmetric top molecules have
been shown in Fig. 2.20. Some of the lines in the rotational spectrum as worked out from Eq. (2.110) are recorded
in the tabular form as follows and are sketched in Fig. 2.21.
State No. of Lines (J ¢+ 1) Wave Number, vJK ,cm−1

J′ = 0, K′ = 0 1 v0 ,0 = 2 B−4DJ
J′ = 1, K′ = 0 2 v1,0 = 4B−32DJ
K′ = ± 1 v1, 1 = 4B−32D J − 4DJK
J′ = 2, K′ = 0 3 v2,0 = 6B−108DJ
K′ = ± 1 v2, 1 = 6B−108DJ − 6DJK
K′ = ± 2 v2,, ±2 = 6B−108DJ − 24DJK (2.111)
J′ = 3 , K′ = 0 4 v3,0 = 8B−256DJ
K′ = ± 1 v3, 1 = 8B−256DJ − 8DJK
K′ = ± 2 v3, 2 = 8B−256DJ − 32DJK
K′ = ± 3 v3, 3 = 8B−256DJ − 72DJK
K=3
J=5
12B
K=3
K=2
J=5 J=4
32B + 4(A−B)
8B
J = 5, K = 1 12B K=3
32B + (A−B) J=3
J = 5, K = 0 K=2
30B 12B
J=4 20B + 4(A−B)
10B J = 4,
K=1 20B + (A−B) 8B
K=2
J = 4, K = 0 J=3 12B + 4(A−B)
20B J = 3, 8B
6B
K=1 K=2
8B 6B + 4(A−B)
12B + (A−B) J = 2
J = 3, K = 0 J = 2, 6B
12B K = 1
6B 6B + (A−B)
J = 2, K = 0 J = 1, 4B
2B + (A−B)
6B K=1
J = 1, K = 0 4B
2B
J=0 2B
0
K=0
Fig. 2.20 Rotational energy levels of prolate symmetric-top molecules.
It is to be noted that Eq. (2.104) permits determination of only one rotational constant. Hence the complete
geometry of a symmetric- top molecule cannot be determined by using this single value of B obtained from the
rotational spectra. In such a case, isotopic substitution is used in structure determination. The centrifugal distor-
tion leads to a slight separation of the components of J.
0 1 1 2 2 3 3 4
J′ J ′′
ν (cm−1)
2B 4B 6B 8B
1 1
0 0 0 0 2 2 1 1 0 0 3 3 2 2 1 1 0 0
K′ K′′
Fig. 2.21 Splitting of rotational lines for symmetric–top molecules of the type CH3F.
Problem 2.20: The v 10 , v 20 , and v 30 lines in the microwave spectrum of CH3F have been observed at 3.40475,
5.10701 and 6.80912 cm−1 respectively. Determine the moment of inertia of CH3F.
Solution The average difference of wave numbers between the adjacent lines is
2B = 0.702185 cm−1; ∴ B = 0.351092 cm−1

16.8575
I= = 48.01 au
0.351092
Problem 2.21: The wave numbers (cm−1) in the microwave spectrum of CH3F are given below:
v10 v11 v30
3.40475 3.40470 6.80912
Determine the values of B, DJ and DJK.
Solution From Eq. (2.111), we write

3.40475 = 4B − 32DJ
3.40470 = 4B − 32DJ− 4DJK
6.80912 = 8B − 256DJ
Solving for B, DJ and DJK , we get
B = 0.851 cm−1
DJ = 1.979 × 10−6
DJK = 1.25 × 10−5 cm−1
2.7.8 Rotational Spectra of Asymmetric-top Molecules

The three moments of inertia are different in such molecules. Consequently, the rotational spectrum of asymmetric
tops is complicated and hence difficult to analyse. Just like linear and symmetric-top molecules, there is no general
expression for the energy of the levels. The energy expressions are very complicated and are manipulated from
the energy expressions of oblate and prolate symmetric top molecules. The theoretical spectrum is then computed
from the approximated expression for the energy levels of asymmetric tops from the assumed values of structural
parameters. The structural parameters are varied by the iterative method till the theoretical spectrum matches with
the observed spectrum of the system under investigation.
Problem 2.22: Determine the equilibrium bond distances and bond angle of water if Ixx = 1.1443, Iyy = 0.6083,
Izz = 1.7526 au.
Solution The geometrical configuration of the water molecule

is shown in Fig. 2.22. H 1(−x1,y1) Y
Knowing the Cartesian coordinates (−x1, y1), (− x2, − y 2), we can

determine the bond distances and bond angle from the expressions
rCH
rOH = x12 + y12 (2.112)
y1
and tan θ = (2.113) θ
x1 −X X
0
Using Eq. (2.15), we proceed as under.
2 2
1 ⎛ 3 ⎞ 1 ⎛ 3 ⎞
∑m (y )
3
I xx
i =1
i i + zi − ⎜ ∑
M m ⎝ i =1
mi yi ⎟
⎠
− ⎜ ∑
M m ⎝ i =1
mi zi ⎟
⎠
Here, zi and the second term are zero. H 2(−x2,y2)

−Y
Therefore, Ixx = 2my 2 Fig. 2.22 Geometrical configuration of the water
molecule.
or I xx 1.1443
y2 = = = 0.5721
2m 2
Consequently, y = 0.7564.
2 2
1 ⎛ 3 ⎞ 1 ⎛ 3 ⎞
Further, I zz m x y (
∑ i i i M ⎜⎝ ∑
+ − )
m x
i i⎟
⎠
− ⎜ ∑
M m ⎝ i =1
mi yi ⎟
⎠
m i =1
Here, the third term is zero.
4m 2 x 2 ⎛ 4m 2 ⎞
Thus, I zz 2m ( x 2 + y 2 ) − = x 2 2m − ⎟ + 2my 2
Mm ⎝ M m ⎠
Rearranging
m 2 ) (1.7526 − 2 0.5721) 9
( I zz 2my
x2 = = = (1.7526 − 2 0.5721)
⎛ 4m2 ⎞ ⎛ 4⎞ 16
1
−
⎜⎝ 2m − M ⎟⎠ ⎜⎝ 2 ⎟⎠
m
18
= 0.3422
∴x=
0.3422 = 0.5850
From Eqs. (2.112) and (2.113), we get
o
rOH = 0.3422 + 0.5721 = 0.956 A
0.7564
tan θ = = 1.2929
0.5850
Hence q = 52.28°
Thus, the bond angle of water is 104.56°.

2.8 THERMAL DISTRIBUTION OF POPULATION AMONG

THE ROTATIONAL LEVELS
The Boltzmann distribution law for the population of every possible rotational energy level may be written as
NJ N 0 e − EJ / T
(2.114)
Some of the energy levels are degenerate, i.e, several levels have the same energy. Degeneracy can be lifted by
the Stark effect. Consequently,
N
N J = 0 g J e − EJ / T (2.115)
g0
where g0 is the degree of degeneracy of the zeroth energy level and gJ of the Jth energy level.
If degeneracy of all the energy levels for one mole of substance is taken into account then
N0
NA = ∑
g0 J
g J e − EJ / T
(2.116)
where ∑N
J
N A , i.e. the sum of molecules occupying all energy levels for one mole of the substance equals the
Avogadro’s constant, NA.
N0
NA = Z (2.117)
g0 rot
∑g e J
− EJ / T
(2.118)
Z rot = J
σ
Here, s is the symmetry number.
The physical significance and determination of Zrot, the rotational partition function will be discussed later on
in this chapter.
From Eqs (2.115) and (2.116),
N J 1 g J e − EJ / T
= (2.119)
NA σ Z rot
For linear molecules gJ = 2J +1, and the energy of the rotational levels of linear molecules is expressed by
Eq. (2.33), i.e.
EJ = BhcJ(J+ 1)
Combination of Eqs. (2.33) and (2.119) yields
N J 1 ( 2 J + 1) − B h c J ( J
= e 1) / T
(2.120)
N A σ Z rot
The effect of correction due to centrifugal distortion on the statistical mechanics of the rotator is negligibly small
and hence not taken into account.
Equation (2.120) shows that the population of different levels decreases exponentially as J grows. Further,
more degeneracy of a given state increases the population of that particular state.
Differentiating Eq. (2.120) with respect to J at constant temperature and applying the condition of
maxima, i.e.
⎛ dN J ⎞
⎜⎝ dJ ⎟⎠ = 0
T
1
⎛ κT ⎞ 2 1
we obtain J max =⎜ − (2.121)
⎝ 2 Bhc ⎟⎠ 2
Thus, the value of J at which population is maximum increases (i) with temperature, and (ii) as the molecule gets
heavier. Jmax can also be determined graphically. At a particular temperature, for different values of J, NJ/ NA is calculated
and then from the plot of J against NJ / NA, Jmax can be determined. J max
The behaviour of J with NJ / NA is presented in Fig. 2.23.
The maximum in the curve arises due to the degeneracy
factor, i.e. (2J + 1) otherwise the shape of the curve is
governed solely by the exponential term. Figure 2.23 NJ
indicates that the population and hence intensities of
NA
rotational lines at given temperature for low, and high
values of J will be low and is maximum near the value of
J = Jmax. To summarise, we conclude that the intensities of
rotational lines are not only proportional to the square of the
transition dipole moment Mt , but are also dependent on the
degeneracy of the state, moment of inertia of the molecule and 0 1 2 3 4 5 6 7 8 9 10 11 12
the temperature. The rotational lines in the spectrum of ortho- J
hydrogen are three times intense than lines of para-hydragen. Fig. 2.23 Hypothetical sketch showing the thermal distri-
This is because of the difference in statistical weights of the bution of the rotational levels at TK. The curve represents
symmetric and antisymmetric rotational states. The ratio of the function NJ/NA = (2 J + 1) e− BhcJ( J+1) /κT as a function of
line intensities in some of the systems exhibiting ortho-para rotational quantum number J. The broken line ordinates
give the relative populations of the corresponding rota-
modifications are given in the tabular form as under. tional levels.
From the comparison of the absorption intensities of two
lines corresponding to the same transition of different isotopic
species of the same molecule, we can determine the natural abundance of isotopes.
Molecule Intensity Ratio
Q:P
H2 3:1
D2 2:1
7
Li 2 1.67:1
31
P2 3:1
35
Cl2 1:4:1
127
I2 1.4:1
14
N2 2:1
He 2 ,12 C 2 , 16 O 2 , 32S2 , 80Se 2 ∞
Problem 2.23: (a) Determine which is the most intense line in the microwave spectrum of CO at (i) 27°C, and
(ii) 500°C. The bond length of CO is re = 1.131 Å. (b) Determine the fraction of CO molecules on the rotational
quantum level J = 5 relative to zeroth vibrational level at 500 K. Zrot = 179.9.
Solution (a) Knowing the value of moment of inertia of CO from its bond distance and atomic masses of its
atoms, we can determine the value of Jmax at different temperatures from Eq. (2.121). We know
−23
κ = 1.38 × 10−23 JK−1 = 1 38 × 10 × 10 amu Å2 s−2 K−1 = 8.31 × 1023 amu Å2 s−2 K−1
47
1.6604
h = 3.9890 × 1013 amu Å2 s−1
From Eqs (2.9) and (2.10), we get
12 × 16
I= × (1.131) 2 = 8.7737 amu Å2
12 + 16
From Eq. (2.34)
16.8575
B= = 1.921 cm −1
8.7737
Substituting the values of B, h, κ, c and T in Eq. (2.121), we obtain

1
⎡ 2
⎤2
1 ⎢ 8.31 × 10 +23 (amu A s −2 K −1 ) × 300(K) ⎥
Jmax(300 K) = − +
2 ⎢ 2 × 1.921(cm −1 ) 3 × 1010 (cm
c s ) × 3.9890 × 10 (amu
1 13
2
1 ⎥
⎣ (
(am A s )⎦
= (54.16)1/2 −0.5 7
1
Jmax (773 K) = ⎡54.16 × 773 ⎤ 2 − 0.5 = 11

⎢ 300 ⎥⎦
⎣
Thus, the rotational lines corresponding to the transition = 6→7 at 27°C and J = 10 → 11 at 500°C will be more
intense as compared to the other lines in the respective spectrum.
(b) The value of Jmax at any temperature can also be determined graphically as follows. In order to determine Jmax
by this method, the factor e−Bhc/κT needs to be defined. Given B = 1.9319 cm−1
− Bhc −1.9319(cm −1 ) × 6.626 × 10 −34 (J s) × 3 × 101100 ( )
=
κT 1.380 × 10 −23
2
( 1
) × 300 K
− Bhc
or
e κT
e −0.0092759
J ( 2J +1) J(J+ 1) e−0.0092759J (J + 1) = j (2J+1) j
0 1 0 1 1
1 3 2 0.981 2.944
2 5 6 0.945 5.675
3 7 12 0.894 6.262
4 9 20 0.830 7.476
5 11 30 0.757 8.327
6 13 42 0.677 8.805
7 15 56 0.594 8.922
8 17 72 0.512 8.717
9 19 90 0.433 8.245
10 21 110 0.360 7.569
11 23 132 0.293 6.760
12 25 156 0.235 5.881
13 27 182 0.184 4.990
14 29 210 0.142 4.134
15 31 240 0.107 3.346
16 33 272 0.080 2.647
17 35 306 0.058 2.048
18 37 342 0.041 1.550
19 39 380 0.029 1.148
20 41 420 0.020 0.833
21 43 462 0.013 0.591
22 45 506 0.009 0.411
23 47 552 0.005 0.280
24 49 600 0.003 0.187
25 51 650 0.002 0.122
Using Eq. (2.94), the data required for the plot of NJ /NA against J is generated and recorded as under:
The curve of NJZrot/NA as a function of J is shown in Fig. 2.24. From the graph, the value of Jmax comes out to be
7 and is in agreement with the value of J determined from Eq. (2.121).
(b) Using Eq. (2.120) we obtain
N 5 ( 2 × 5 + 1) −1.9218 × 5 (5 + 1) × 3 × 10 × 6.6252 × 10
10 34
= e
NA 179.9 1 3805 × 10 − 23 × 500
11
= e −0.166 = 0.0518
179.9
10.0 1.00
(b)
(a)
8.0 0.80
6.0 0.60
N J Z rot NJ Z rot
NA NA
4.0 0.40
2.0 0.20
0.0 0.00
0 2 4 6 8 10 12 14 16 18 20 22 24 26
J
Fig. 2.24 The plot between (NJ /NA)× Zrot and rotational quantum number
J, at 300 K (a) with degeneracy, and (b) without degeneracy.
Problem 2.24: Determine the natural abundance of H35Cl and H37Cl in HCl if the relative intensity of the two
lines for the same transition in the infrared rotational spectrum of HCl is 3:1.
Solution We know that intensity is proportional to the relative abundance of isotopic species of the same mol-
ecule and for the same transition.
I 3
Thus, natural abundance of H35C1 = × 100 = × 100 = 75%
I I* 3 +1
where I* is the intensity due to H37C1.
Therefore, natural abundance of H37C1 = 100 − 75 = 25%.
Problem 2.25: Determine the relative intensity of the two rotational lines corresponding to the same transition
in case of CO if the natural abundance of 13C16O in CO is 0.8%.
Solution The natural abundance of 13C16O = 0.8%.

Hence the natural abundance of 12C16O = 100 −0.8 = 99.2%.
We know that natural abundance of I2CI6O
I
= × 100 (2.122)
I + I*
where I* is the intensity due to 13C16O.
Therefore, natural abundance of 13C16O
I*
= × 100 (2.123)
I + I*
Comparing Eqs (2.122) and (2.123), we get
I 99.2
= = 124 : 1
I* 08
i.e. for the same transition, the rotational line of 12C16O is 124 times intense than that of 13C16O.
2.9 STARK EFFECT IN RELATION TO DIPOLE MOMENT

DETERMINATION FROM ROTATIONAL SPECTRUM
The splitting and shifting (with respect to unperturbed line) of spectral
lines under the influence of an external electric field E is known as Stark
effect. Thus, when a molecule is simultaneously subjected to microwave +4 1
radiation and an electric field, the splitting of rotational lines occurs. The
electric splitting and displacement of spectral lines is due to the interaction +3 2
of dipole moment of the molecule with the external electric field. The
electric field splits the levels specified by J into (J + 1) levels specified by +2 3
quantum number, MJ , MJ actually is the component of J in the direction of
external field, i.e. electric or magnetic and determines which of the pos- +1 4
sible (2J + 1) orientations, the molecule can take up. The nine possible
Stark rotational orientations of molecule for J = 4 are shown in Fig. 2.25. 0 5
MJ can take up any integral value including zero, i.e. MJ = 0, 1, 2, 3,...,
i.e. J, J − 1,..., − J,... The magnitude of splitting depends not only on J but −1 6
also on M J2 . Thus depending upon M J , some levels are moved upward
and others downward by the field. −2 7
The energy of perturbed rotational levels as deduced from the solution

−3 8
of Schrodinger wave equation is expressed as
1 −4 9
EJ ,K ,M J EJ + α J K M J E βJ ,K ,M J E 2 + ... (2.124)
2
Fig. 2.25 The nine degenerate rotational
where EJ is the energy of the rotator in the absence of field E; α J , K , M J and
orientations of a molecule with J = 4 down-
βJ , K ,M J are the first-and second-order Stark effects, respectively. When dipole ward by the field.
moment M has a component in the direction of J, we have the first-order Stark
effect and second-order Stark effect when it is perpendicular to the direction
of J as shown in Fig. 2.26.
The Stark frequency of transition is determined from the solution of the E
integrals,
Δ J =∫ J′ ME s θ ψ J ′ dτ (2.125)
= ∫ψ J ″ M E
J
Δ J s θ ψ J ″ dτ (2.126)
MJ
It is found that the Stark shift
Δvrot ∝ ( ME
M )2 (2.127) M
for linear molecules
and Δvrot ∝ ( ME
M ) (2.128)
for symmetric top molecules.

Thus, from the slope of the straight-line plot between Δvrot and E2/E
as the case may be, the value of dipole moment M can be determined.
Fig. 2.26 A diatomic molecule under the
For linear molecules, M = slope while for symmetric-top molecules, influence of an external electric field.
M = slope.
2.9.1 Stark Transition Between Adjacent Levels in Symmetric-Top Molecules

Depending upon convenience, the first- or second-order Stark effect may be exploited to determine M of
symmetric–top molecules. For symmetric-top molecules, the energy correction due to first-order Stark effect
is given by,
−M K MJ
α J ,K ,MJ = (2.129)
J (J )
Thus, for symmetric-top molecules, Eq. (2.124) becomes
M K MJ
E J , K , M J = BhcJ
c (J )− hcE
c (2.130)
J (J )
The selection rule is ΔMJ = 0 when an electric field is applied in the direction of the incident radiation while
ΔMJ = ±1 when the direction is perpendicular to the incident radiation.
Let us now consider the transition between two adjacent levels characterised by (J, K′, MJ) and (J + 1, K″, MJ″).
From Eq. (2.130) we write
M K′ MJ′
E J , K ′ M J ′ = BhcJ
c (J )− hcE
c (2.131)
J (J )
M K ″ MJ ″
EJ , K ″ MJ ″ = Bhc( J 1)( J + 2) − hcE
c (2.132)
( J 1)( J + 2)
ΔErot E J , K ″, MJ EJ K ′ MJ ′
vrot = =
hc hc
⎡ K ″MJ ″ K′MJ′ ⎤
= 2B( J + 1) − M ⎢ − E (2.133)
⎣ ( J + 1)( J + 2) J ( J + 1) ⎥⎦
Applying selection rules, ΔK = 0, ΔMJ = 0 we get K′ = K″ = K; M J M J ′ = M J (say).

Consequently, Eq. (2.133) becomes
2M K M J E
vrot B( J + 1) +
2B (2.134)
J ( J + 1)( J + 2)
The first-order Stark rotational lines for the J = 1→2 transition as worked out from Eq. (2.134) are recorded in
the tabular form as follows. We know that for J = 1, K = 0, 1 and MJ = 0, ±1.
State No. of Lines Wave Number

vJ,K,M J ,cm −1
J = 1, K = 0, MJ = −1 v1,0, −1 = 4B + 0
=0 3 v1,0,0 = 4B + 0
= +1 v1, 0, +1 = 4B + 0
J = 1, K = 1 ,MJ = −1 v1,1, −1 = 4B − ME
3
=0
3 v1,1,0 = 4B + 0
= +1 ME
v1,1, +1 = 4B +
3
Thus, the Stark lines for J = 1→2 transition appear symmetrically on both sides of the original line, i.e. the line in the
ME
absence of external electric field. The Stark shift of symmetric tops for J = 1→2 transition is . For M = 1 Debye
ME 3
unit (10−18 esu), E = 1 esu (300 volts cm−1); comes out to be of the order 0.0016 cm−1 and is easily observable.
3
2.9.2 Stark Transition Between Adjacent Levels in Linear Molecules

In case of linear molecules, there is no component of dipole moment M in the direction of J, consequently the
energy of the rotational levels may be expressed as
1
E J , M J = Bhc J(J + 1) + hc β J , M J E2 (2.135)
2
The values of b are defined by
−2 M 2
b 0,0 = cm−1 for J = 0 (2.136)
B
β J , M J = −2M ⋅
2
3 J2 − J (J + 1)
(2.137)
B J (J + 1)( 2J − 1)( 2J + 3)
Let (J′, MJ′) and (J ″, MJ″ ) be the quantum numbers for ground and excited states respectively. With the aid of
Eq. (2.135) and selection rule, ΔJ = 1, we arrive at the result
ΔErot EJ EJ ′ M J ′
(2.138)
vrot = =
,M
M J ′′
hc hc
1
= B( J ″ − J ′ )( J ″ + J ′ + ) + [β J MJ − βJ MJ ′ ]E 2
2
Since ΔJ = +1
1
vrot B ( J ′ 1)
2B [β J ,M
MJ βJ MJ ′ ]E 2 (2.139)
2
The factor [β J , J β J ,,MM J ′ ] can be deduced from Eqs. (2.136) and (2.137). The Stark shifts for some typical
transitions have been worked out with the aid of Eq. (2.139).
(a) Stark Shift for J = 1 → 2 transition

For the transitions, J = 1 → 2 and J = 0 → 1, the Stark splitting is shown in Fig. 2.27.
From Eq. (2.139), we write
For J = 1 → 2, MJ = 0 → 0, ΔMJ = 0, we get MJ J
±2
−32 M 2
β20 β10 = J
105 B 2 2
For J = 1 → 2, MJ = 1 → 1, ΔMJ = 0, we get ±1

0
26 M 2
β21 β11 =
105 B ±1
Thus, for J = 1 → 2 transition, from Eq. (2.139) we get 1

1
16 M 2 E 2
vrot ( M J ) = 4B − (2.140) 0
105 B
0 0 0
13 M 2 E 2 Fig. 2.27 Stark splitting for J = 0→1 and J = 1→2

vrot ( M J ) = 4B + transitions .
105 B
From Eq. (2.140), it is evident that for J = 1 → 2 transition in case of linear molecules, the two Stark lines are
not exactly symmetrical on both sides of the original line. The Stark shifts are −16M2E2/105B and 13M2E2/l05B.
(b) Stark Shift for J = 0 → 1 Transition For J = 0 → 1 transition, Eq. (2.139) is written as
1
vrot B( J ′ 1)
2B
2
( 10 )E2 (2.141)
16 M 2 (2.142)
β10 β00 =
15 B
Thus, Eq. (2.141) becomes
8M 2 E 2 (2.143)
vrot 2B +
15 B
Hence, the Stark shift in case of J = 0 → 1 transition is 8M2E2/l5B.
To conclude, we can say that unlike symmetric–top molecules, the Stark shifts in linear molecules not only
depend on M and E but also on B and hence IB.
The observed splitting for J = 1 → 2 transition in case of OCS is shown in Fig. 2.28 while the plot between Δvrot
and E2 is presented in Fig. 2.29.
No Field
1
=+ ±1
ΔJ →
=±
1
MJ
Δν–rotIn MHz
750 V cm−1
M J = 0→0
–
Δν rot
–
Δν rot
ΔJ = +1
1050 V cm−1
E 2In (e.s.u. cm−1)2

– –
Δν rot Δν rot Fig. 2.29 A curve between Δ ν rot and E2 for the J = 1 → 2
Fig. 2.28 Stark splitting of the J = 1 → 2 lines of OCS. transition of the linear OCS.
The second–order Stark energy correlation for linear molecules under a weak field is expressed as
M 2 [3M J2 FJ (F
( FJ )][ C
C(C
(C ) FJ (F
( FJ )J(J
J (J )]
βJ , FJ = − (2.144)
B 2JJ ((JJ )( 2J 1)( 2J 3)2 J ( J 1)( 2FJ 1)( 2 J 3)
Here, C = FJ(FJ+ 1) − J(J+1)−I(I +1) and
FJ = J + I, J+ I − 1,...,0,...,(J −I)
When FJ is an integer, MJ = FJ+ 1;
MJ = FJ, FJ − 1 FJ − 2, … 0
When FJ is an integral multiple of 1/2,
1 1
MJ FJ + ; M J FJ , FJ 1 FJ − 2, ...
2 2
The selection rules are ΔJ = ±1, ΔF = 0, ±1.
−M 2
Therefore, for J = 0, β0 , FJ = cm −1, which does not depend on FJ nor MJ .
6B
The additional effect due to polarisability of the molecules, i.e. the polarisabilities along the molecular axis (a||)
and in a direction perpendicular to the molecular axis (a ⊥) is not considered. The coefficients of the main term M2/
B for .J = 1 states are recorded in the tabular form as follows.
Spectrum I FJ MJ β J , FJ
HCN 1 0 0 0
−0.050
0⎫
⎬
1 1⎭ +0.025
0⎫ +0.050
2 ⎪ +0.025
1⎬
2⎪
⎭ −0.050
HBr 3/2 1/2 1/2 0

or HCl 3/2 −0.04
/ ⎫
⎬
3 / 2⎭ +0.04
1/ 2 ⎫ +0.04
5/2 ⎪ + 0.01
3 / 2⎬
5 / 2⎪
⎭ −0.05
CO 0 1 −0.20
0⎫
⎬
1⎭ +0.10
HF 1/2 1/2 1/2 0

−0.10
1/ 2 ⎫
3/2
⎬ +0.10
3 / 2⎭
In case of HI, I = 5/2, the spin of H is neglected. In case of DF, DC1, etc., the spin of D is also neglected.
The coefficients for J = 2 state can also be obtained from Eq. (2.144). After applying the selection rules,
the position of the Stark lines and hence Stark shift for J = 1→2 transition relative to the original line can be
obtained.
Further, we know that the degeneracy of any rotational level is given by (2J + 1) and the energy of each degen-
erate level is the same. Thus, in the absence of an external field, the transitions between the states will have the
same energy and the observed rotational spectrum will be simple. Since the degeneracy gets lifted under the influ-
ence of an external electric field, the Stark rotational spectrum will have multiple structures for all the possible
transitions between the Stark rotational levels. Thus, with the aid of Stark effect, we can assign a particular value
of J to an observed spectral line.
2.10 INVERSION IN AMMONIA AND ITS ANALOGUES

The phenomenon of inter-transformation of an atom X in Y
Y
pyramidal molecules XY3, namely NH3, ND3, PH3, AsCl3,
AsH3, etc., from one side of Y3 plane to the other side of the
plane gives rise to an inverted configuration with respect X X
Y Y
to the original configuration and is termed as inversion or
mirror reflection. The inversion of XY3 type molecules is
presented in Fig. 2.30.
These two configurations necessarily have the same Y Y
energy. The inverted configuration cannot be obtained with Fig. 2.30 The inversion of a pyramidal molecule XY3.
the rotation of the molecule. Basically, inversion is a vibrational problem n(cm−1)

and is associated with the vibrations of the atom X relative to Y3.
V=9
The potential energy of a pyramidal molecule as a function of distance
of atom X from the Y3 plane is shown in Fig. 2.31. V =8
The two minima, separated by a potential hill (which actually represents V=7
the hindrance to inversion), correspond to the two configurations. The V=6
vibrational levels associated with the vibrational motion of the atom X V=5
relative to Y3 is also presented in Fig. 2.31. The splitting of levels into two V=4
sublevels is due to the resonance interaction between the two identical con- V=3
figurations and is known as inversion doubling. The magnitude of splitting
V=2
grows with the growth of the vibrational quantum number. The fundamen-
V=1
tal vibrational bands of NH3 appear as doublets, i.e. (3325.9, 3337.5 cm−1)
−1
and (931.58, 968.08 cm ) in its gaseous infrared spectrum. It is to be noted V=0
X
that the interaction involves energy difference in the microwave region. x0
According to classical mechanics, the inversion is not possible since X
necessary activation energy is not available to cross over the inversion bar- Fig. 2.31 Potential energy of pyramidal
rier. In reality, the quantum mechanical particle, i.e. the atom X here, leaks molecules XY3 as a function of distance of the
or tunnels through the classically forbidden region (potential hill) from X atom from the Y3 plane.
one classically allowed region, i.e. potential minimum to the other allowed
region, i.e. the second minimum. The time taken by the atom X to tunnel through the barrier depends upon its height
and shape. Consequently, the inversion is possible due to the finite height of the potential barrier. The average time
of inversion is inversely proportional to the energy difference of the two sublevels for a given V and is given by,
1 5.307 × 10 −11
ti = = (2.145)
2 ∏ ΔE
Ei c ΔEEi
where ΔEi is in cm−1.
The phenomenon of inversion has been thoroughly studied theoretically as well as experimentally in case
of ammonia. The inversion doubling splits the rotational lines of NH3 into doublets as shown in Fig. 2.31. The
doublet splitting of the lines is twice the separation of the inversion doublet levels. The height of NH3 pyramid is
one half of the separation between the two maxima. PH3, AsH3, AsCl3, AsD3, BiCl3, etc., do not show inversion
doubling since the inversion frequencies in these systems are slower because of the increasing height of potential
barrier and increasing of the mass of the atom concerned. Further, pyramidal–type molecules do not exhibit optical
isomerism while tetrahedral types exhibit optical isomerism. In these systems, if the substituents are bulky, the time
of inversion by the mechanism of tunnelling may be of the order of 109 years. This led to infer why bacteria consume
only one of the two forms, i.e. right or left of a certain carbon compound. The asymmetric molecules in bacteria have
not undergone inversion since the particular genus came into being.
Problem 2.26: Determine the time of inversion and barrier to inversion in case of NH3 if the separation of
inversion doublet for the ground state is 0.66 cm−1. [The splitting of the upper levels, i.e. ΔEi = 1 (V1 = 1), 35.7
(V2 = 1), 312.5 (V2 = 2) cm−1]
Solution The time of inversion can be determined from Eq. (2.145), i.e.
5.307 × 10 −11
ti = = 8.04 × 10 −11 s .
0 66
Thus, the time of inversion in case of ammonia is of the order of 10−11 s. However, the time decreases as the vibra-
tional quantum number increases. The average time of inversion is of the order of 2.5 × 10−11 s. In case of PH3 this
time is 1.1 × 10−7 s. The time of inversion may be in hours or even in years also. The barrier to inversion may be
determined from the expression,
k Ze − Ea / RT (2.146)
where Z is the frequency factor; R = 1.985 cal mole−1 K−1; and T = 298 K.
Substituting the values of k, R and T into Eq. (2.146), we obtain
( )
−1
Ea 1.985 l Z − 1.985 × 298 ln 8 4 × 1
98 298 ln
= 591.53 lnZ −13749.527
Assuming Z to be of the order 1014, we get

Ea = 5.319×103 cal mole−1 = 5.319 kcal mole−1
The actual value of Ea is of the order of 6 kcal mole−1. The estimated value of Z for ammonia is 3.161 × 1014.
It gives Ea = 6 kcal mole−1.
Problem 2.27: The average values of rotational constants as deduced from the pure rotational spectra of NH3
and ND3 are 9.945 cm−1 and 5.138 cm−1 respectively. Determine the geometrical parameters of NH3.
Solution In order to determine the bond distance of NH3, we write from Eq. (2.23)
3mH rN2H ⎡ ⎛ 3mH ⎞ 2 ⎤

I= ⎢2 − 1 − sin β ⎥
⎡ 3m ⎤ ⎣ ⎜⎝ mN ⎟⎠ ⎦
(2.147)
2 ⎢1 + H ⎥
⎣ mN ⎦
3m D rN2H ⎡ ⎛ 3m D ⎞ 2 ⎤
I* ⎢ sin β ⎥ (2.148)
⎡ 3m D ⎤ ⎣ ⎝ mN ⎠ ⎦
2 ⎢1 +
⎣ m N ⎥⎦
Dividing Eq. (2.147) by Eq. (2.148), we obtain
⎡ 3m D ⎤ ⎡ ⎛ 3m H ⎞ 2 ⎤
1+ 2 − 1− sin β ⎥
I m H ⎢⎣ m N ⎥⎦ ⎢⎣ ⎜⎝ m N ⎟⎠ ⎦ (2.149)
=
I* m D ⎡ 3m H ⎤ ⎡ ⎛ 3m D ⎞ 2 ⎤
⎢1 + m ⎥ ⎢ 2 − ⎜⎝1 − m ⎟⎠ sin β ⎥
⎣ N ⎦⎣ N ⎦
We know that B = 9.945 cm−1 and B* = 5.138 cm−1;

I B* .138
= = = 0.5166, I 1. , I * = 3.280 au (2.150)
I* B 9.945
I
Substituting the values of from Eq. (2.150), mN, mD and mH into Eq. (2.149) and solving the resulting equa-
I*
tion, we obtain
sin b = 0.927
Therefore, b = 67.971°.
Substituting the values of I, and the masses of the respective atoms into Eq. (2.147), we get
2
rNH = 1.036 Å
Hence, rNH = 1.018 Å

From Eq. (2.24), the height of the pyramid is hp = 1.018 × cos 67.971° = 0.381 Å.
rHH = 3(1.018 × 1.081 − 0.381 × 0.381 = 1.634 Å
3(1.018 × 1.018 − 0.381 × 0.381)
cosa = 1 −
2 × 1.018 × 1.018
= 1−1.289 = − 0.289
Hence, a = 106.79°.
2.11 BARRIER TO INTERNAL ROTATION

The rotation of one part of the molecule relative to another part of the molecule about a single or double bond is
termed as internal rotation, e.g. two groups, viz. CH3 and OH of methanol rotate about the C—O bond, two chemi-
cally equivalent CH3/CH2 groups of C2H6/C2H4 rotate about the C—C/C = C bond. Absolute free internal rotation
is possible only when the hindrance, i.e. potential-energy barrier to internal rotation, is zero which of course is not
possible theoretically and experimentally in real molecular systems. Therefore, for internal rotation to occur, some
energy is required to overcome the potential-energy hill. When the potential–energy barrier is ≤ κT(2 kcal), the
internal rotation is said to be free internal rotation, e.g. CH3NO2 (C—N), CH3OH (C—O), C2H6 (C—C), CH3—
CC13 (C—C), (CH3)2 S(C—S), (CH3)2 SiH2, CH3SiH3 (C—Si), CH3COOH (C—C), N2H4 (N-N), (CH3)2O(C—O),
etc. On the other hand, when the barrier height exceeds KT, the internal rotation is known as torsional oscillation,
e.g. in C2H4. Thus, free rotation and torsional oscillation are the limiting cases of internal rotation. Internal rotation
appears when the atoms in a molecule form configurations separated by a potential-energy minimia and just like
inversion, the internal rotation is possible by the quantum mechanical tunneling effect.
As a result of internal rotation of methyl groups in simple molecules, the rotational lines in the pure rotational
spectra of molecules split up into two components. The spacing between the doublets is of the order of a few
megahertz. From the splitting, the barrier to internal rotation is determined. The error is of the order of 5 percent.
The barrier to internal rotation about single bonds in some typical molecules is
C—C in (CH3)2O:2.58, C—S in (CH3)2S:2.10, C—Si in (CH3)2SiH2: 1.83,C—O in CH3OH:1.07,
C—C in CH3—CCI3:2.70 and C—C in CH3CH2Cl:3.68 kcal mole−1.
The potential energy for C2H6, CH3OH type molecules is expressed as
U (q) = U ⎛⎜ θ
2 ⎞
π⎟ (2.151)
⎝ 3 ⎠
while for C2H4 type molecules, it is
U (q) = U (q + π) (2.152)
where q is the angle of torsion.
The plot of potential energy as a function of angle of torsion in C2H4 or in CH3OH and C2H6 type molecules is
presented in Fig. 2.32. The two identical minima in C2H4 correspond to the two energetically identical configura-
tions of C2H4 while the three minima in C2H6 type molecules correspond to three energetically equivalent configura-
tions of C2H6. Thus, as a result of interaction among the energetically identical configurations, the torsional levels
will split into sublevels depending upon the number of configurations. The separation between the torsional levels
increases as the torsional quantum number increases. In case of C2H4, each level of torsion oscillation splits into two
sublevels while in C2H6 type molecules, each torsion oscillation level also splits into two sublevels but the sublevels
are doubly degenerate as shown in Fig. 2.32 (b). The degenerate sublevels are alternately the upper and lower one.
The energy level of torsion oscillation lies significantly below the potential energy barrier and is expressed as
(a)
Vt
U 3
2
1
0
−90° 0° 180° 210°
(b)
Vt
2
U
1
−60° 0° 120° 240° 300°

Fig. 2.32 Potential energy as a function of torsion angle in molecules of
the (a) ethylene type C2H4, and (b) ethane type C2H6 or CH3OH, etc.
⎛ 1⎞
Et Vt + ⎟ (2.153)
t
⎝ 2⎠
where Et is the energy of the torsion oscillation level, wt is the frequency of torsion oscillation and Vt is torsional
quantum number and can take on all integral values including zero, i.e. Vt = 0, 1, 2….. . The potential energy of
torsion oscillation is expressed as
1
U= U0(1 − cos nq) (2.154)
2
where n is the number of identical minima, and U0 is the height of the potential hill separating the minima. For
small oscillations, q is small and hence,
cos nq = 1 − n θ
2 2
(2.155)
2
Consequently, Eq. (2.154) reduces to
2
U = U0 ⎛ nθ ⎞ (2.156)
⎜⎝ ⎟⎠
2
Thus, the frequency of torsion oscillation is given by
wt = n U 0 A1 A2 (2.157)
A
where A1 and A2 are the rotational constants of the two parts of the molecule that exhibit torsional motion rela-
tive to each other about the same axis. The respective moments of inertia are represented by I 1A and I A2 . A is the
rotational constant of the whole molecule about the axis of torsion such that
A A1 + A2 or ⎫
⎬ (2.158)
IA I A(1) + I A( 2 ) ⎭
h
Here A1 = (2.159)
8π I c2 (1)
A
and
h
A2 = (2.160)
8π I c
2 ( 2)
A
Combining Eqs (2.158), (2.159) and (2.160), we get

h
A A1 + A2 = (2.161)
8π 2 I r c
I A( ) I A( )
where Ir = (2.162)
I A( ) I A( )
and is called reduced moment of inertia of the molecule. Here A, A1, A2 and U0 are in cm−1. Consequently, wt, will
also be in cm−1.
For symmetric molecules of the type C2H4, C2H6, C2(CH3)2, etc., A1 = A2 = 2 A. Hence Eq. (2.157) for such type
of molecules reduces to
wt = 2n U 0 A (2.163)
Thus, from Eqs (2.157) and (2.163), we can determine the barrier to internal rotation from the observed value of
torsional oscillation and the rotational constants of the molecule.
In case of symmetric molecules, the internal rotation and torsional oscillation will be microwave and infrared
inactive respectively, since there will be no oscillating dipole moment. Hence, just like ordinary rotation spec-
trum, there will be no pure rotational spectrum of such molecules due to internal rotation.
For nearly asymmetric molecules, such as CH3OH, the free internal rotation and torsional oscillation will be
infrared active. The energy of internal rotational level (on the assumption of free rotation) as deduced quantum
mechanically is given by the formula
2
AA ⎡ A⎤
Eint rot Ek1 , k =h 1 2 ⎢ k1 k ⎥ (2.164)
A ⎣ A1 ⎦
where k = ±K, is the component of total angular momentum J about the top axis, also called unique axis, and k1,
is quantum number of angular momentum of part ‘1’ of the molecule.
k1 = 0, ±1, ±2... (2.165)
Equation (2.164) may be expressed as
A2 ( k k1 ) − Ak 2
2
Ek1 , k A1k12 (2.166)
where, k – k1 = k2 is the quantum number of angular momentum of part ‘2’ of the molecule about the top axis.
Theoretical studies reveal that as the potential barrier grows from zero (free rotation) to large values in a molecule,
the levels of free rotation expressed by Eq. (2.164) pass gradually into levels of torsional oscillations.
As stated earlier that for symmetric molecules, viz. C2H4, C2H6, etc., A1 = A2 = 2A. Consequently, the expression
(2.166) reduces to
Ek1 , k A( 2k1 k ) 2 = A [k1 ( k1 k )]2 (2.167)
= A( k1 − k2 ) 2 = AK i2
AK
Here Ki = | k1 − k2 | is the quantum number for internal rotation.

Since K = |k1 + k2|, therefore, k1 − k2 = 0, ±2, ±4,.. for even values of K and k1 − k2 = ± l, ±2, ±3,... for odd values
of K. The selection rules for internal rotation are ΔKi = 0 for ΔK = 0 in case of || bands, i.e. Ki, does not change for
|| bands, and ΔKi = ± 1 for ΔK = ±1 in case of ⊥ bands, i.e. Ki changes by ±1 for ⊥ bands.
Combining (i) Eq. (2.167) with Eq. (2.97), and (ii) Eq. (2.97) with Eq. (2.166), we obtain the expressions for the total
energy of the asymmetric and symmetric molecules, respectively. For symmetric molecules, viz. C2H4, C2H6, etc.,
c ( J + 1) + hc(
E J, K , k1 , k2 = BhcJ hc( A B ) K 2 + Ahc (k
( k1 k2 ) 2 (2.168)
and for nearly symmetric molecules, viz. CH3OH,

c ( J + 1) − BhcK 2 + A1h
E J, K , k1, k2 = BhcJ ckk12 + A2 h
hck ckk22
hck (2.169)
2.11.1 Selection Rules for Transition Between Two Adjacent

Rotational Energy Levels in Slightly Asymmetric Molecules
of the Type CH3OH
In CH3OH, the rotation of the OH group about the top axis is responsible for the transition dipole moment. The
selection rules for the pure internal rotation spectrum are
ΔJ = 0, ±l, ΔK = ±l, Δk1 = ±l, Δk2 = 0 (2.170)
where K1 = |k1| and K2 = |k2|. The subscripts ‘1’ and ‘2’ correspond to OH and CH3 groups respectively.
Now if two sets of quantum numbers, i.e. (J′, K', k1' k2' ) and (J ″, K", k1" k2" ), correspond respectively to the
lower and upper rotational energy levels then we can write from Eq. (2.169)
E rot EJ ″ , K ″ , k 1 , k 2 EJ ′ , K ′ , k ′1 , k 2′
ΔE
v rot = =
hc hc
= 2B (J ″ − J ′ )(J ″ + J ′ + 1) − B ( K ″ 2 + K 2 ) (2.171)
+A 1 ( k 1″ − k 1′ )( k 1″ + k 1′ ) + A 2 ( k 2″ − k 2′ )( k 2″ + k 2′ )
Applying the selection rules expressed by Eq. (2.170), we obtain

v rot A1 − B K′
BK
2BK A 1k 1′ for ΔJ = 0 (2.172)
The upper signs correspond to positive ΔK and Δk1, while the lower signs correspond to negative ΔK and Δk1. The
lines in the spectrum as deduced from Eq. (2.172) are recorded in the tabular from as follows:
State Wave Number, v K , k 1 ,cm −1
K′ = 0, k1′ , = 0 v0 ,0 = A1 − B
±1 v0 , 1= A1 − B +2A1
±2 v0,, ± 2 = A1 − B +4A1
±3 v0,, ± 3 = A1 − B + 6A1
K′ = 1, k1′ = 0 v1,0 = A1 − B − 2B
±1 v1, 1= A1 − B− 2B+2A1
±2 v1,, ± 2 = A1 − B− 2B+4A1
±3 v1,, ± 3 = A1 − B− 2B+ 6A1
The table reveals that the Q lines of the free internal rotation spectrum consist of a series of bands of spacing
2A1, each of which consists of sub-bands whose zero lines have a spacing 2B. It is to be noted that A1 >> B. Such
a series of bands for large values of k1 have been observed in the infrared spectrum of CH3OH in the region 600−
860 cm−1.
The selection rules for the rotational–vibration spectrum are as follows:
(a) For || Bands (ΔK = 0) Δk1 = 0, Δk2 = 0, i.e. the internal rotation does not influence the structure of a ||
band just as for molecules of the type C2H6, provided there is no strong interaction between vibration and internal
rotation.
(b) For ⊥ Bands Δk1 = ± 1, Δk2 = 0, or Δk1 = 0, Δk2 = ± 1 depending on whether the dipole moment of the vibra-
tional transition is in part ‘1’, i.e. OH or part ‘2’, i.e. CH3 of the molecule in case of CH3OH. With these selection
rules for ⊥ bands, it can be shown from the formula (2.171) that the ⊥ bands have a double rotational structure as
has been observed in case of molecules of the C2H6 type. Each sub-band with a given K and ΔK (= ±1) is made
up of sub-sub-bands corresponding to different k1 values and Δk1 = ±1 (OH group) or to the different k2 values
and Δk2 = ±1 (CH3 group).
The spacing of the sub-sub-bands will be ∼ 2A1 or 2A2 while the spacing of the subbands will be 2B. Since A1
and A2 >> B, a number of sub-bands of spacing 2A1 or 2A2, each of which consists of lines with spacing 2B are
possible.
Problem 2.28: From the Raman spectrum of C2H4 in the gaseous phase, the band at 825 cm−1 has been
assigned to torsional oscillation about the C = C bond. The infrared absorption spectrum of C2H4, also in the gas-
eous phase, shows a fine structure of fundamental at 3.22m. The spacing between the equidistant lines in the fine
structure is 9.734 cm−1. Determine the barrier to internal rotation about the C = C bond.
Solution Given 2A = 9.734 cm−1, wt = 825 cm−1, n = 2

Therefore, A = 4.867 cm−1
From Eq. (2.163), we obtain the barrier to internal rotation,
ω t2 825(cm −11 ) × 825(cm 1 )

U0 = =
4n2 A 4 4 × 4.867(cm −1 )
= 8740 cm−1 = 25 kcal mole−1

Thus, barrier to torsional oscillation is about 25 kcal mole−1.
Problem 2.29: The infrared spectrum of CH3OH in the gaseous phase has been recorded in the 600–800 cm−1
region. The spacing between the successive lines has been found to be about 40 cm−1, The strong band at 270 cm−1 in
the infrared spectrum of gaseous CH3OH is assigned to torsional oscillation about the C—O bond. Determine barrier
to torsional or twisting oscillation.
Solution Here n = 3, 2 A1 = 40 cm−1, wt = 270 cm−1
From Problem (2.4), the moment of inertia due to CH3 is
I CH3 I A( ) = 3.286 au
With the help of Eqs (2.34) and (2.160), we get
16.8575
A2 = 130 cm −1
= 5.130
3.286
Thus, A = A1 + A2 = 20 + 5.130 = 25.130 cm−1
The Eq. (2.157) may be written as
ω t2 A
U0 =
n2 A1 A2
Substituting the values of various parameters, we obtain
270 × 270 × 25.130
U0 = =1984 cm −11 = 5
5.67
67 k
kcall mole
l
9 20 × 5.130
2.12 ROTATIONAL SUM OVER THE STATES

AND ROTATIONAL CONSTANTS
In statistical mechanics, the quantity ∑ge
i
i
− Ei / T
was originally known as Zustandsumme (in German), i.e. state
sum or sum over the states, and is denoted by the symbol Zrot as
Z rot ∑ge i
i
− Ei / T
(2.173)
However, with time the term state sum became known as partition function which of course is a misnomer.
Partition function indicates how at equilibrium, the systems are partitioned among the available energy states,
although it is the individual term in the summation which gives us the information. It is denoted by Q. This func-
tion plays an important role in statistical mechanics. The formulae for rotational partition functions are different
for linear, symmetric and asymmetric-top molecules.
2.12.1 Sum Over States for Diatomic and Linear Polyatomic Molecules
For diatomic and linear polyatomic rigid rotator molecules, the rotational partition function is given by
( J )e − BhcJ ( J )/ T
Z rot = ∑ (2.174)
J σ
since for linear rigid rotators
EJ = BhcJ (J+ 1) (Eq. 2.33)
Here, s is called the symmetry number (spin of identical nuclei is zero). It is equal to the number of indistinguish-
able positions into which a molecule can be turned by rigid rotations.
s = 2 for homonuclear diatomics and s = 1 in the case of heteronuclear diatomic molecules. The nuclear spin
factor can be neglected in statistical calculations and hence is not incorporated in the formula.
At ordinary temperature, the number of rotational levels involved are very large, hence the asymptotic expansion
is employed, i.e.
1 ⎡ κT 1 1 hcB
c 4 ⎛ hcB
c ⎞
2
1 ⎛ hcB
c ⎞
3
⎤
Z rot = ⎢ + + ⋅ + + + ...⎥ (2.175)
σ ⎣ hcB
c 3 315 κ T 315 ⎝ κ T ⎠ 315 ⎝ κ T ⎠ ⎦
When B is small and T is large, the higher terms may be omitted. Consequently, Eq. (2.175) reduces to
κT T
Z rot = = 0.6951 (2.176)
σ hcB
hcB σB
Equation (2.176) may also be deduced from Eq. (2.144) by replacing the summation by integration. By making
use of the above formulae, we can determine the value of Zrot from the observed value of rotational constant, B. It
is to be noted that the influence of nuclear spin is omitted in the formula.
Problem 2.30: Determine the rotational partition function for CO and HCN at 500 K. The rotational constants
for CO and HCN are 1.9319 and 1.4878 cm−1 respectively.
Solution The CO and HCN molecules are heteronuclear, hence s = 1

Using Eq. (2.176)
500
Z rot (CO) = 0.6951 × = 179.9
1.9319
500
Z rot ( HCN ) = 0.6951 × = 233.6
1.4878
Problem 2.31: Determine the rotational component of the sum over states for CO2 and CO +2 at 500 K. The
rotational constants (in the ground state) for CO2, and CO +2 are 0.3895 and 0.3806 cm-1, respectively.
Solution The symmetry number in both the cases is 2.

+
B0 (CO2) = 0.3895 cm−1 and B0 (CO 2 ) = 0.3806 cm−1
With the aid of Eq. (2.176), we get
500
Z rot (CO 2 ) = 0.6951 × = 446.1
2 × 0.3895
500
Z rot (CO 2+ ) = 0.6951 × = 456.5
2 × 0.3806
2.12.2 Sum Over States for Symmetric-Top and Asymmetric-Top Molecules

The rotational partition function for symmetric tops is given by
( J + ) −[ BJJ ( J ) ( A − B ) K 2 ]κhcT
∞ +J
∑ ∑J σ e
Z rot =
J = 0 K
(2.177)
since the energy of the rotational levels of symmetric-top molecules is expressed as

E J , K = BhcJ
c (J ) + ( A B )hcK
c 2
where K = J, J −1, …, −J.
For all cases except at very low temperatures, the following expansion of Eq. (2.177) is used.
e Bhc /( 4 T)
π ⎛ κT ⎞
3
⎡ 1⎛ B ⎞ Bhc 7 ⎛
2
B ⎞ ⎛ Bhc ⎞
3
⎤
Z rot = × × ⎢1 + 1 − + × 1 − + ...⎥ (2.178)
σ B A ⎝ hc ⎠
2
⎣ 12 ⎝ A ⎠ κ T 480 ⎝ A⎠ ⎝ κT ⎠ ⎦
For small values of B/T, the higher terms in Eq. (2.178) are ignored. Consequently, Eq. (2.178) becomes
π ⎛ κT ⎞
3
1
Z rot ( Z rrotot ) B =
T
→0 σ B 2 A ⎝ hc ⎠
1.02718
= T 3 / B2 A (2.179)
σ
Thus, knowing B and A from the rotational spectrum of symmetric tops, we can determine the value of Zrot for
these type of molecules.
For spherical-top molecules, the formula for rotational partition function can be obtained by substituting A = B
in the expression for symmetric-top molecules.
In case of asymmetric-top molecules if B and C are not too different, the constants A and B in Eq. (2.178) are
replaced by A and BC respectively in order to obtain the formula for rotational partition function of asymmetric
tops. Consequently, for asymmetric tops
BChc /( 4κ T )
π ⎛ κT ⎞ × ⎡ 1⎛ BC ⎞ ⎤
3
e BC hc
Z rot = × ⎢1 + ⎜1 − ⎟ + ...⎥ (2.180)
σ ABC ⎝ hc ⎠ ⎣ 12 ⎝ A ⎠ κT ⎦
For sufficiently high temperature or small rotational constants, Eq. (2.180) reduces to
π ⎛ κT ⎞
3
1 1.02718 T 3
Z rot = =
σ ABC ⎝ hc ⎠ σ ABC
0.0148386
= T 3 I A AB I C (2.181)
σ
Problem 2.32: The rotational constants for halomethanes are given in the tabular form as follows:
Determine their rotational partition functions at 500 K.
Variation Number System B0 (cm-1) A0, (cm-1)
1 CH3Cl 0.49 5.09

2 CH3Br 0.31 5.08
3 CH3I 0.28 5.07
Solution Using Eq. (2.179), we can determine the value of Zrot. The calculated values of Zrot at 500 K are
recorded in the tabular form as follows.
Variation Number System s Zrot Zrot /s
1 CH3Cl 3 10388.35 3462.78

2 CH3Br 3 16436.45 5478.81
3 CH3I 3 18215.44 6071.81
Problem 2.33: Determine the rotational partition functions for asymmetric–top molecules such as C2H4 and
C2D4 at 500 K. The moments of inertia of the two systems are
System IA(au) IB(au) IC (au)

C2H4 3.464 16.911 20.386
C2D4 6.913 22.843 29.763
Solution s = 4 in both the cases. Using Eq. (2.181), we get
0.0148386
Z rot (C2 H 4 ) = 500 × 500 × 500 × 3.464 × 16.911 × 20.386
4
5733.073
= = 1433.268
4
0.0148386
Z rot (C2 D 4 ) = 500 × 500 × 500 × 6.913 × 22.843 × 29.763
4
11373.552
= = 2843.388
4
2.12.3 Partition Function for Spherical-Top Rigid Molecules

In case of spherical-top molecules, A = B. Consequently, Eq. (2.179) becomes
3
π ⎛ κT ⎞
3
1 1.02718 ⎛ T ⎞
Z rot ( Zrot ) B
→3
=
σ B 3 ⎝ hc ⎠
=
σ ⎜⎝ ⎟⎠
B
(2.182)
T
Problem 2.34: Determine the partition functions for CH4, CD4, SiH4 and SiD4 at 500 K if the values of the
respective rotational constants are 5.252, 2.64, 2.96 and 2.87 cm−1.
Solution Here, s = 12 in all the cases. The partition functions as determined from the formula (2.182) are
recorded in the tabular form as follows.
Variation Number System Zrot Zrot/s
1 CH4 954.145 79.512
2 CD4 2677.291 223.107
3 SiH4 2255.090 187.924

4 SiD4 2361.993 196.832
2.12.4 Sum Over States for Molecules with Internal Rotation

In case of polyatomic molecules having groups of atoms that are capable of rotating one with respect to another,
the partition function is computed by employing the formula
1
1 ⎛ 8π 3κ T I red ⎞ 2 0.359965 (2.183)
Zintrot = ⎜ ⎟⎠ = TI red
σ⎝ h2 σ
Here, s is the degree of symmetry of a group of atoms, and Ired is the reduced moment of inertia (in au) of
internal rotation. It is to be noted that for the rigid C2H6 molecule, the value of s is 6. However, when there is free
rotation, there are three indistinguishable positions for the two CH3 groups relative to each other. For each of these
three indistinguishable positions, we have the six positions as for rigid C2H6.
Consequently, the total number of indistinguishable positions are = 3 × 6 = 18.
Thus, s = 18 in this case.
I1 I 2
I red = (2.184)
I1 I 2
I2 is the moment of inertia of one half of the molecule about the top axis and I1 is the moment of inertia of the
whole molecule about an axis perpendicular to the top axis.
Problem 2.35: Determine the value of the free internal rotational component of the sum over states at 500 K
in case of CH3OH if I CH3 = 3.284 au and IOH = 0.8136 au.
Solution Here the symmetry number sCH3 × sOH = 3 × 1 = 3,
From Eq. (2.184) we get
I CH3 I OH 0.8136 × 3.284
I red = = = 0.652 au
I CH3 I OH 0.8136 + 3.284
Using the formula (2.183), we obtain
0.359965 1
Zintrot = × (500 × 0.652) 2 = 2.166.
3
2.13 APPLICATIONS OF MICROWAVE SPECTROSCOPY

Rotational spectroscopy is employed to estimate the relative abundance of isotopes and for the detection of
conformational isomers since they have different moments of inertia and hence rotational spectra. The symmetry
and geometry of molecules are also determined from microwave spectroscopy. It also serves as a potential tool to
determine the dipole moment of linear and symmetric-top molecules. The advantages of the microwave method over
other methods of measuring dipole moments are its accuracy and sensitivity, which makes measurements possible
even with small quantities of a sample. The sensitivity of the microwave method is especially significant in case of
low-volatilcity substances whose dipole moments cannot be determined by methods based on dielectric constant and
index of refraction measurements of these substances due to their low vapour pressure. The dipole moments when
measured in solution by these methods are likely to be distorted due to the interaction of the sample under study
with the solvent. The microwave method like other methods based on dielectric constant and index of refraction
gives dipole moment of the molecule as a whole. The experimental method for determining the dipole moments of
intramolecular bonds from intensities of the infrared vibrational spectra will be discussed in Chapter 3 on infrared
vibrational spectroscopy. We know that as a result of interaction between the rotation of the molecule and reorienta-
tion of the nuclear spin, the rotational energy levels get perturbed. This effect appears in the rotational spectrum of
the molecules. Thus, from the spectral features one can get information on the nuclear spin and about the symmetry
of the electronic environment in the region of the nuclei. Microwave spectroscopy has also played a major role in the
chemical study of interstellar space. A number of chemical species such as OH, CS, CH, CH+, CN radicals and mol-
ecules CO, SiO, CH3CN, CH3OH, HCOOH, HCN, HNCO, COS, NH3, H2O, etc., have been detected. In the light of
these species, we can also look for other possible candidates in interstellar space, e.g. the existence of CN, HCN and
HNCO led to infer that NC, NCO may also be searched. The presence of NH3 and H2O may throw light on the origin
of biological molecules and hence the origin of life itself. Since the intensities of rotational lines are temperature
dependent, the temperature of interstellar material can also be estimated. The temperature is generally determined
from the temperature effect on the rotational spectrum of NH3. It is to be remembered that the microwave spectrum
of the molecule is observed as a whole. So just like vibrational and nuclear magnetic resonance spectra, it cannot be
used for the study of functional groups, viz., CH3, OH, SH, NH, CO, etc., present in the molecule.
PROBLEMS
1. Determine the atomic mass of the unknown atom say 7. The pure rotational spectrum of the CO molecule shows
in the following diatomics from their rotational constants. the following lines: 7.682, 11.508, 15.316, 19.100 cm−1
Write the molecular formulae of the diatomics and calcu- Compute r0 and D0 for the molecule.
late their bond lengths also. 8. The number of lines in the pure rotational band of some
molecule is given by
Molecule 2B(cm−1)
vrot 1000 (2n − 1)cm−1 and vrot 1000 (2n + 1)
X Br79
0.195
for positive and negative values of n respectively. Calcu-
X79Br 0.188 late the moment of inertia of the molecule.
X37C1 21.32 • Show B = 1000 cm−1
3
HX 7.48 9. Using the moments of inertia data from the book for
F2, H2, HI, NO and OH diatomics, calculate (a) their
35
C1X 1.03 bond lengths, (b) for J = 5, their angular momenta and
X F 19
1.01 velocities, and (c) determine the energy absorbed in
the J = 5→6 transition.
2. Find the product of the moments of inertia, IA, IB and IC of • Angular momentum =
the water molecule if the equilibrium bond distance and
bond angle are re = 0.957 Å and q = 104°31' respectively. [Ans: For OH, (a) 0.96 Å
[Ans: 55.84 × 10−142 kg3m6] (b) 5.77 × 10−27 erg s, 3.90 × 1013 rad s−1 (c)227cm−1]
3. Using the values of reduced moments of inertia from the 10. Determine the energy of rotation and the rate of rota-
book for SO2 and H2S, calculate their bond lengths and tion for J = 10 for the following molecules from their
rotational constants. microwave spectra.
4. Determine the relative population of the first five rota-
tional levels of Br2, BrCl, DF, and DI if their respective Molecule 2
H37Cl 2
H79Br H19F
2
H19F
3 35
Cl19F 37
Cl81Br
values of Be are 0.0821, 0.152, 11.00 and 32.84 cm−1.
2B(cm−1) 10.94 8.64 42.17 15.48 1.03 0.187
5. The bond lengths of IO and FO molecules are 1.868 and
1.32 Å respectively. Estimate which is the most intense
line in their pure rotational spectra at (a) 100 K, and 11. Calculate the rotational partition functions for the mol-
(b) 1000 K. Be = 0.340 cm−1 for IO and 1.104 cm−1 for FO. ecules listed in Problem 6 at 25°C and 100°C.
6. Given the equilibrium bond lengths of the following rigid 12. Calculate the energy in calories per mole needed to
rotator diatomics; calculate the wave numbers of the first raise NO from its lowest rotational level to the level with
five lines in their rotational absorption spectra. J = 4. Given r0 = 1.151 Å. [Ans: 97.42 cal mole−1]
13. Estimate the values of centrifugal distortion constants
Molecule 12
C19F I2C35Cl l2C31P Si1H
28 28
Si35Cl 14N2H 16
O2H from vibrational frequency and rotational constant data
for the following nonrigid diatomic rotators. Also com-
r
e
(Å) 1.267 1.645 1.558 1.521 2.058 1.038 0.971 ment upon the values so obtained.
19. The following spectroscopic constants are reported for

Molecule H2 I2 IO BrCl pure samples of 72Ge32S and 74Ge32S:
B(cm−1) 60.86 0.037 0.340 0.152
−1
Parameters 74
Ge32S 72
Ge32S
w (cm ) 4400.4 214.55 681.47 444.27
Be (MHz) 5593.08 5640.06
14. Frame different types of problems from the rotational αe (MHz) 22.44 22.74
constant data for the molecules given below:
D (kHz) 2.349 2.388
Molecule 2B (cm−1)
re(V = 0)(pm) 0.20120 0.20120
H Cl
1 35
21.35
H35C1
2
10.97 Determine the frequency of the J = 0 to J = 1 transition
H Cl
3 35
7.51 for both the isotopic species in their ground vibrational
state. The width of a microwave absorption line is of the
H19F
1
42.22 order of one kHz. Could you distinguish a pure sample
H F
2 19
42.17 of 74Ge32S from a 50/50 mixture of the two isotopic spe-
cies using microwave spectroscopy?
H19F
3
15.48
H Br
1 81
17.06
⎡ ν,J 0 ν,J 1 ⎤ Δν
H Br
2 81
8.64 ⎢ Molecule 1 1 ⎥
⎢ (cm ) ((cm (cm 1) ⎥
cm )
H81Br
3
5.83 ⎢ 74
Ge 32S 0.372372 0..744742 0.372370⎥
⎢ ⎥
35
Cl19F 1.03 ⎢ 72 ⎥
⎢ Yes Ge 32S 0.375496 0.750990 0.375494⎥
37
Cl F 19
1.01 ⎣ ⎦
15. Determine the H⎯C and C N bond lengths for HCN

given 20. The spectroscopic parameters of some diatomic mol-
[Ans: 1.07, 1.15 Å] ecules in the ground electronic state are indicated
below. Determine the values of B0, B1, B2, and B3 of the
H12C14N
1
B0 = 44316 MHz diatomics.
D12C14N
2
B0 = 36208 MHz
Molecule Be(cm−1) ae(cm−1)
16. The vibration–rotation constant for 69Ga35Cl is +0.004
cm−1 and 2B0 = 0.2972 cm−1. Determine the value of re. H2 60.8530 3.0622
[Ans: 2.213 Å]
17. The atoms of CSl are separated by 3.31 Å. Calculate the O16O
16
1.4456 0.0159
reduced mass and the rotational inertia of the molecule. C16O
12
1.9313 0.0175
[Ans: 105.6 amu; 7.7 ×10−34 kg-m2]
18. The rotational spectrum of 12C32S gives the following N16O
14
1.67195 0.0171
data.
N14N
14
1.9982 0.01732
Determine the value of Be and ae from these data.
[Ans: ae = 0.00592 cm−1, Be = 0.82004 cm−1] H19F 20.9557 0.798
−1
V BV (cm ) H35Cl 10.5934 0.3072
0 0.81708
H79Br 8.4649 0.2333
1 0.81116
H127I 6.5122 0.1689
2 0.80524
3 0.79932 K39K
39
0.05674 0.000165
APPENDIX 2.1A: ENERGY LEVELS OF A DIATOMIC RIGID ROTATOR

The two atoms, say 1 and 2, in a diatomic molecule form H = Kinetic energy due to both the masses + Potential
a rigid rotator, i.e. when the diatomic molecule rotates, energy. (2.1A-1)
the interatomic distance between the two atoms does not Since in the case of rigid rotator, the interatomic distance
change. If m1 and (x1, y1, z1) and m2 and (x2, y2, z2) are the remains fixed, the potential energy will be constant and can
masses and coordinates of the atoms 1 and 2 respectively be considered as zero. Thus Eq. (2.1A-1) reduces to
then the classical Hamiltonian in the Cartesian coordinates H = Kinetic energy due to both the masses.
system will be
Substituting the values of the respective differentials into

(2.1A-2) Eq. (2.1A-8) and rearranging the resultant equation, we get
sin ∂ ∂Θ 1∂Φ 8 2IE

sin sin (2.1A-9)
Θ ∂ ∂ h
Also we know that Separating the variables, we obtain

x = r sinq cosj sin ∂ ∂Θ ∂Φ
y = r sinq sinj (2.1A-3) sin s (2.1A-10)
Θ ∂ ∂ Φ∂ 2
z = r cosq
So Eq. (2.1A-2) in terms of spherical coordinates can be where
expressed as
Now each side of Eq. (2.1A-10) contains one variable and
⎡ dr d
2
d
2
H m1 + + they must be equal to the same constant. Let this constant

2 dt dt dt be equal to m2 so that
1 dr
2
d d
2
⎤
+ m2 r2 +r n (2.1A-4) (2.1A-11)
2 dt dt dt
If the distance r of the masses from the origin is fixed then

the derivatives and (2.1A-12)
dr The solution of Eq (2.1A-12) is

dt dt
Thus Eq. (2.1A-4) reduces to Φm(j) must be of the well-behaved class, i.e. it must be a
single-valued function. Therefore Φm(φ) must have the same
(2.1A-5) values for j = 0 and 2π, i.e.
The first term within brackets of Eq. (2.1A-5) represents the

moment of inertia (say I) of a rigid rotator. So
and
should be the same. Therefore,
(2.1A-6)
Thus the solution of the Schrödinger wave equation of type

or
(2.1A-7) or
will give the energy levels of the rigid rotator. We know that the
which can be true only if m has integral values including
rigid rotator behaves like a single particle of mass ‘I ’ placed
zero, i.e. m = 0, 1, 2, 3…….. Such characteristics are those
at a fixed distance equal to unity (since r = 1) from the origin
that would be expected of a quantum number.
(which is the centre of mass). Consequently, Schrödinger
wave equation for a rigid rotator is given by Next, the solution of Eq. (2.1A-11) is obtained as follows:
Multiplying Eq. (2.1A-11) by we get
∂2 2
+ ro
∂x y2 ∂ 2
h 1 ∂ ∂Θ mΘ
si + =
sin ∂ si 2
and in spherical polar coordinates
(2.1A-13)
Ψrot 8 2IE
sn + ro =0
sin ∂ si ∂ 2 h
Let x = cosq
(2.1A-8)
∂x
Equation (2.1A-8) indicates that Ψrot is a function of q and ∴ si
∂
j, i.e. ∂Θ
So we can write as
Ψrot = Ψrot (q, j) ∂
If Ψrot is a product of two functions and each function has ∂Θ ∂Θ ∂x ∂Θ

= − si
one variable only then ∂ ∂ ∂x
Consequently, Eq. (2.1A-13) becomes
Ψ (q, j) = Θ(q) Φ(j)
∂Ψ ∂Θ
Φ
∂
or (2.1A-14)
which is the Legendre’s differential equation. x can take Taking the first and second derivatives and proceeding as in
on values from cos0 or (+1) to cosπ or (−1). The solution of the case of a harmonic oscillator, we arrive at the following
Eq. (2.1A-14) can be obtained by expressing Θ as: recursion formula.
aν + 2 (ν + m)(ν + m + 1) − λ
(1 − x 2 )|m|2| L = (2.1A-16)
aν (ν + 1)(ν + 2)
Thus
for the coefficients of a power series of L(x). Here, ν is a zero
or an integer. This formula is analogous to Eq. (2.1A-15). For
∂Θ
m m
−1 ∂L L(x) to be a polynomial, the series L(x) must break off after a
= −mx (1 − 2 2
) L + (1 − x ) 2 2
∂x ∂x finite number of terms which is possible only if

or (n + m) (n + m + 1) − l = 0
∂Θ
m m
+1 ∂L or l = (n + m) (n + m + 1) = l (l + 1)
(1 − x 2 ) (1 2
) 2 L + (1 − x 2 ) 2
∂x ∂x Since ν and m qualify the condition of a quantum number,
Hence the first term of Eq. (2.1A – 14) is the sum (ν + m) = l should also qualify the same condition,
i.e.
∂ 2 ∂Θ ⎤ ∂ ⎡ +1 ⎤
m m
⎡ l = 0, 1, 2, ……..
⎢(1 − x ) ∂x ⎥ = ∂x ⎢ −mx (1 − x ) L + (1 − x ) L′ ⎥
2 2 2 2
∂x ⎣ ⎦ ⎢⎣ ⎥⎦
8π 2IErot
We know that λ =
∂L h2
where L′ =
∂x
or 8π 2IErot
So = l(
l l + 1)
⎡ m m
−1 ⎤
h2
∂ ⎡ 2 ∂Θ ⎤
⎢( − x ) ∂x ⎥ = ⎢ −m( − x ) + m x ( − x ) ⎥ L
2 2 2 2 2 2
∂x ⎣ ⎦ ⎣⎢ ⎥⎦
h2
⎡ m
⎤ or Erot = l (l + 1)
− ⎢2x (m )( )( x 2 ) 2 ⎥ L′ 8π 2I
⎢⎣ ⎥⎦
Replacing I by J, we get
⎡ m
+1 ⎤
+ ⎢( x 2 ) 2 ⎥ L″
⎢⎣ ⎥⎦ h2
EJ / Erot J(J + 1) (2.1A-17)
8π 2I
∂L 2
Here L′′ =
∂x 2
where J is known as the rotational quantum number. Or
Consequently, Eq. (2.1A-14) in terms of x can be written as
( 2
) ′′ − 2( ) ′ [λ ( + 1)] L = 0 EJ / Erot Bhc J(J + 1) ; J = 0, 1, 2, 3,……. (2.1A-17a)
or (1 − x 2 ) ′′ 2α xL′ + β L = 0 (2.1A-15) EJ
or the rotational term FJ = = BJ(J + )
hc
where a = m + 1, and b = λ − m (m + 1)
The solution of Eq. (2.1A-15) can be obtained by the power h
series technique, i.e. where B = cm−1 is called the rotational constant.
8π 2Ic
∞
Equation (2.1A-17a) is the expression for the energy levels
L(x) = ∑
v = 0,1,2..
aν x ν = a 0 + a1x + a2 x 2 + ...
of a rigid rotator.
APPENDIX 2.2A: ENERGY LEVELS OF A RESTRICTED ROTATOR

The internal motions of complex molecules in crystals are where q is the angle of rotation. For a molecule made of light
the examples of a rigid rotator moving under the influence atoms, such as H2 or O2, n = 2, U = 0 at q = 0, π, 2π….. etc.
of a potential field. Accordingly, the wave equation for this system is expressed as
Let us now consider a rigid rotator whose moment of iner-
d 2 Ψ( ) 8π 2I U0
tia I, moving in a sinusoidal potential field of type + 2 [E [ Er (1 nθ )] ( ) 0 (2.2A-2)
dθ 2 h 2
U0
U(q) = (1 − cos nθ ) (2.2A-1)
2
Making substitutions: h
where A = is the rotational constant for the whole mol-
θ 8π 2Ic
x n ; nθ 2x
2 ecule. For the various values of n; we obtain,
n
dx = dθ U0
2 Em −
= 1, 2 = Am 2, m = 0, 1, 2, 3, 4……..
n2 n
dx 2 = dθ 2 hc
4
U0
Em −
into Eq. (2.2A-2), we obtain 2 = Am 2, m = 0, 2, 4, 6, …….
n = 2,
hc
d 2 ( x ) 32π 2I U π 2IU0
+ [ 2 2 {Er − 0 } + cos 2x
2 ] (x ) 0 (2.2A-3)
dx 2 nh 2 n 2h 2 U0
Em −
d 2
(x ) n = 3, 2 = Am 2, m = 0, 3, 6, 9……….
or + (br x )Ψ( x ) = 0 (2.2A-4) hc
dx 2
n 2r 2
where m 2 =
4
in which Φ = 8π IU0
2
(2.2A-5)
2 2
nh
1
Thus for n >1, only th of the energy levels for the free rota-
n
32π I
2
U
and br = (Er − 0 ) (2.2A-6) tor are allowed. The connection between this fact and the
n 2h 2 2
origin of symmetry number in rotation partition function is
obvious. These energy levels are doubly degenerate.
Equation (2.2A-4) is known as Mathieu’s differential equation,
the solution of which is hard and cumbersome. However the
limiting cases of this equation correspond to the motion of a Case II: For br << 2 Φ, the wave function will have
free rotator or that of harmonic oscillator:
appreciable values only for x ≅ 0. so Eq (2.2A-4) can be
Case I: For br >>2Φ, Eq: (2.2A-4) reduces to written as
d 2Ψ
d 2 (x ) + [br ( 2x 2 )]
) 0 (2.2A-11)
+ br ( x ) = 0 (2.2A-7) dx 2
dx 2
This is the equation for the harmonic oscillator and can be
The solutions of this equation are rewritten as
± ir nθ d 2Ψ
Ψ = e ±irx
±r
=e 2 (2.2A-8) + [(br 2 ) 4 x 2 ]Ψ = 0
dx 2
where r is apositive integer including zero.

or d Ψ
2
+ [λ − 2
x 2 ]Ψ = 0 (2.2A-11a)
dx 2
d Ψ 2
d Ψ 22
= i 2r 2e ± irxr −r Ψ o
2
+r Ψ = 0
dx 2 dx 2
in which λ = br + 2Φ and a = 2 Φ . Equation (2.2A-11a) is
So br = r2 analogue to Equation (2.2A-4) of the harmonic vibrator. Thus
By the symmetry of the potential field, the energy levels are given by the relation
λ = (2r + 1) a , i.e.
2π
Ψ( ) = Ψ
Ψ(( ) (2.2A-9)
n br = 2 (2r + 1) √Φ − 2Φ so that
According to Eq. (2.2A-2), only even integral values of r Er = (r + 1/2) h ν0 r = 0, 1, 2, 3…….. (2.2A-12)
including zero are allowed, i.e. r = 0, 2, 4, 6,…. Since br = r2,
so the energy levels are deduced from Eq. (2.2A-6). These ν
or Er = (r + 1 ) 0 (2.2A-12a)
are. hc 2 c
U0 n 2h 2r 2
Er − = , r = 0, 2, 4, 6,…….. (2.2A-10) where ν 0 =
n U0
(2.2A-13)
2 32π 2I 2π 2I
U0
Er − 2 2 2 2
or 2 = n hr = n r A (2.2A-10a) For the intermediate case br ~ Φ, the energy levels cannot be
hc 32π 2Ic 4 well approximated by either of the above relations.
CHAPTER 3 INFRARED SPECTROSCOPY
No hypothesis can lay claim to any value until it assembles many phenomena under one concept.
—Goethe
3.1 INTRODUCTION
Just like rotational motion, the vibrational motion of molecules is also quantised and the vibration of a molecule
manifests itself in the heat capacity of gases and in the molecular vibrational spectra. The discrete nature of the
vibrational spectra is an experimental evidence to the quantum nature of the vibrational motion of molecules.
Since the energy of vibrational transitions lies in the infrared region of the electromagnetic radiation, the vibra-
tional spectrum is also known as infrared spectrum (IR). Thus, IR absorption spectroscopy deals with the absorp-
tion of photons in the 10−12,000 cm−1 frequency region of the spectrum. As a result of the absorption of photons,
the vibrational, rotational or vibrational–rotational configuration of the molecules gets perturbed in contrast to
visible/ultraviolet spectroscopy where, in addition, electronic configuration of the molecules also gets affected
due to the absorption of visible/ultraviolet photons.
Since the energies involved in IR spectroscopy are of the order of tenths of electron volts, IR radiation
affects only vibrational and rotational transitions within molecules. Indeed at normal temperatures, IR absorp-
tion usually arises primarily from vibrational rather than rotational transitions. Thus, the rotational–vibrational
motions are usually found in gases. The change (Δ J) in the total rotational quantum number gives rise to P, Q
and R branches in the spectra of gases (Δ J = −1 for P, ΔJ = 0 for Q and Δ J = +1 for R). Since the rotational
frequencies depend upon the moments of inertia of the molecules, the rotational–vibrational spectra are of help
in the study of molecular structure. In condensed gases, liquids and solids, only the vibrational motions have
been generally observed. The vibrational frequencies of polyatomic molecules depend upon (a) the masses of
linked atoms, (b) their geometric interrelationships, and (c) the forces due to deviation from equilibrium bond
configurations.
IR spectroscopy is divided into three major regions:
(a) Far IR (10–200 cm–1) This region deals with pure rotational motion of the molecules. It is seldom used in
chemical spectroscopy since it is difficult to generate and detect such low frequencies. The region from 200–670
cm−1 is also sometimes included in the far IR. The primary use of this region has been for the determination of the
structures of inorganic and metal organic species based on absorption measurements. It is to be noted that heavy
metal iodides generally absorb in the region below 100 cm−1, while the bromides and chlorides have bands at
higher wave numbers. Far IR studies of inorganic solids have also provided information about the lattice energies
of crystals and transition energies of semiconducting materials.
(b) Mid IR (200–4,000 cm−1) This is the most suitable and important region for qualitative as well as
quantitative IR spectral investigations since most of the frequencies due to functional groups in molecules fall
in this region. The studies in this region are based upon absorption, reflection and emission measurements of
samples.
(c) Near IR (4,000–12,000 cm−1) Almost all the bands in this region are due to either overtones or combina-
tion tones of A—H stretching vibrations where A = C, N, O, S, etc. The importance of this region is mainly from
the point of view of quantitative determination of functional groups specially containing hydrogen atoms. This
region also finds considerable use for the routine quantitative determination of certain species, such as water,
carbon dioxide, sulphur dioxide, low-molecular-weight hydrocarbons, amine nitrogen and many other simple
compounds that are of interest in agriculture and industry. These determinations are usually based upon diffuse
reflectance measurements of untreated solid or liquid samples or transmission studies of gases.
3.2 MECHANISM OF IR ABSORPTION AND IR ACTIVITY

The inter-atomic distances within molecules fluctuate cyclically about average values through one or more
simultaneous vibrational motions, viz. stretching, bending, torsion, etc. When such motions change the dipole
moment of a given bond, its electric field oscillates at the same frequency as the bond vibration (1010 −1011 Hz,
corresponding to wavelengths 2−20 m). Thus, when molecules are exposed to infrared radiations, they will expe-
rience its electric field. If the field is reversed periodically as it is in the environment of radiation then the dipole
charges will experience forces alternately. As a result of this interaction, the dipole spacing will increase and
decrease. Hence the net effect of the oscillating electric field of the IR photons will make the dipole moment of
any molecule vibrate at the frequency of the photon. The pictorial view of the effect of electric field of radiations
on the dipole is shown in Fig. 3.1.
The molecules have certain nuclear vibrational frequencies Electric Field
where the vibration causes an oscillation of the dipole moment. − + − +
The photon will excite only those vibrations which cause dipole
moment to oscillate, i.e. the absorption of IR radiation will occur + + + +
if the frequency of photon matches or resonates with any one S C S C
of the frequencies of the molecule. Since dipole moment (M) is − − − −
defined in the case of a simple dipole as the product of magni- Dipole
tude of either charge (e) in the dipole and the charge spacing (r), Force
+ − + −
i.e. M = er, therefore, if a nonvibrating molecule has a centre of Time
symmetry and none of the vibrations alter the centre of symmetry Fig. 3.1 The effect of electric field (reversed periodi-
at any moment will necessarily be infrared inactive. The dipole cally) on the dipole increase (S)/decrease (C) of dipole
moment does not change since it is always zero. Thus, the gross spacing.
selection rule for IR vibrational spectrum is that the molecule
must have permanent dipole moment, i.e. M ≠ 0, and must
± dM
possess oscillating (also called transition) dipole moment Mt. dr obs
Bond Moment, M
Variation in dipole moment of a bond A—H with the inter-

nuclear distance is shown in Fig. 3.2. The two cases shown in Mobs
the figure illustrate the possible differences in the sign of the
slope (dM/dr). The homonuclear diatomic molecules such as
H2, N2, O2, etc., are IR inactive since their dipole moment
is zero. Vibrational data for such molecules are determined re
Distance, r
from Raman and electronic spectra. However, if the centre of
Fig. 3.2 Change in the dipole moment of a bond A—H
symmetry of such molecules gets distorted some way or the with internuclear distance.
other, the infrared inactive vibration becomes infrared active.
It is thus possible to observe infrared inactive frequencies at high pressures, when due to collisions, the sym-
metry of the molecule is perturbed, e.g. at pressures of 150 atmospheres and higher, we observe absorption
bands for H2, N2 and O2. At pressures up to 25 atmospheres, the absorption spectrum of C2H2 exhibits a band
at 1974 cm−1 which is infrared inactive at ordinary conditions. The increase in intensity of the band is propor-
tional to the square of the gas density which indicates that activity is due to deformation of the C2H2 molecule
in collisions. Adsorption also causes deformation of the molecule since optically inactive frequency of H2
observed at 4161 cm−1 in Raman spectrum also appears at 4131 cm−1 in IR spectrum during its adsorption on
solid surfaces.
Further, molecular vibrations which are infrared active may or may not be Raman active. The necessary
condition for a particular vibration in a molecule to be Raman active is that its polarisability should change
when the molecule is exposed to an electromagnetic wave of light. The mechanism of change in polaris-
ability of molecule will be explained in Chapter 4 on Raman spectroscopy. The vibrations which do not
lead to change in polarisability or dipole moment are both IR and Raman inactive.Thus, from the foregoing
discussion it is clear that in all the molecules which have centre of symmetry, the infrared active vibrations
are Raman inactive and vice versa. This is called exclusion principle, e.g. CO2 has centre of symmetry. Two
of its fundamental vibrations are IR active only while the third one is Raman active. On the other hand, all
the three normal modes of vibrations in H2O are Raman as well as IR active. To summarise, we can say that
Raman spectra significantly supplement IR spectra and simplify the problem of analysing the vibrational
terms of complex molecules. The Raman spectrum of molecules will be discussed in detail in Chapter 4 on
Raman spectroscopy.
Infrared Spectroscopy 85
3.3 INSTRUMENTATION Entry

Slit
Exit
Analyser
The instrument employed for recording Continuous Sample Monochromator Detector
Source
the spectra in the IR region is called an
IR spectrophotometer. There are generally
two types of spectrometers, viz single- and Fig. 3.3 Block diagram of a single-beam spectrometer.
double-beam spectrometers. The block dia-
gram of a single-beam spectrometer is shown in Fig. 3.3.
When a single-beam spectrometer is used for record-
ing the spectrum, the IR absorption bands due to H2O and
CO2 present in the air appear in the spectrum as shown in
Fig. 3.4. The background spectrum due to CO2 and H2O
is, therefore, to be subtracted from the spectrum of any
sample under investigation in order to have the spectrum
of the sample. H2O CO2 H2O
Since H2O and CO2 may create difficulties in the spec- 4000 3500 3000 2500 2000 1800 1600
troscopic studies in their respective absorption regions,
they should either be removed from air or ways and means
should be found out to avoid overlapping due to CO2 and
H2O of the spectrum. One way of dealing with this prob-
lem is the complete evacuation of the spectrometer. This
is not an easy task because the volume of a modem spec-
trometer is of the order of 0.3 m3 and even the chances of
leakage in such a large instrument are not ruled out. Fur-
ther, time limit is also a problem. The evacuation process
needs a week’s time for the complete desorption of water
vapours from the surface of mirrors inside the instrument. H2O CO2 H2O
The second method of tackling the problem is to flush the 1600 1400 1200 1000 800 600 400
instrument either with dry nitrogen or dry CO2 free air. But ν (cm−1)
both these methods are not very effective since they are
tedious and time consuming. However, the introduction Fig. 3.4 The spectrum of atmospheric water vapour and car-
of double-beam spectrometer proved very fruitful since it bon dioxide.
eliminates the problems which are encountered in a single-
beam spectrometer. The block diagram of a double-beam
spectrometer is presented in Fig. 3.5.
Alternator
3.3.1 Working of a Double- I
Sample
It Exits
slits
Beam Spectrometer Continous Optical Monochromator
The infrared radiation from the source is Source splitter
split into two beams, one passing through the Optical

I attenuator I0 Detector
sample and the other serving as reference.
Entry
Therefore, when spectrum of a solution is to slits
be recorded, the reference beam should be
passed through the solvent. The function of
Analyser
the attenuator is to reduce the intensity to the
Fig. 3.5 Block diagram of a double beam spectrometer.
required level. The two beams are fed alter-
nately to the monochromator through the entry slits by an oscillating mirror assembly called alternator. The
width of entry slits varies between about 10 and 100 m. After passing through the optical portion of the spec-
trometer, the radiation passes through the second set of slits, called exit slits (width 10−100 m), to the detector
and then to the analyser. The analyser records per cent transmission of the sample as a function of wavelength or
wave number. It should be borne in mind that double-beam instrument only removes the spectral traces of CO2
and H2O. Therefore, for very accurate spectroscopic work, the instrument must be flushed thoroughly with some
dry inert gas since the very strong absorption energy by CO2 and H2O still remains in both beams and reduces
their energy.
3.3.2 Sources
The two most common sources for IR radiations are the Nernst glower and the globar. The Nernst glower is
a hollow rod of zirconium, yttrium and thorium oxides heated electrically around 1500°C. It gives an intense
radiation of up to about 1000 cm−1. Its normal life is several hundred hours. Globars are made from silicon
carbide and operate at about 1400°C. They are more stable than Nernst glowers. The radiation emitted approxi-
mates that of a black body. Globar life is relatively short and power consumption is relatively high. The other
sources are heated nichrome wire, heated alumina tube or Welshbach mantle which is good for below 400 cm−1
and tunable lasers.
The wavelength of any black-body radiation is given by the Planck equation
C
Wλ = 1
C 2 / λT
(3.1)
λ (e
5
− 1)
where Wl is the amount of radiation emitted by the black body per unit surface area, T is the absolute temperature
of the black body, l is the wavelength of the emitted radiation in centimeters, and C1 and C2 are the constants;
C1 = 3.7402 × 10−12 watts/cm2 and C2 = 1.43848 cm deg−1. The displacement of maximum intensity of emitted
radiation with temperature is given by Wien’s displacement law
lmax T = K (3.2)
where the constant, K = 2897 m deg−1 when lmax, is in microns. At the lowest temperatures, the maximum lies in
the infrared region. That is precisely why infrared radiation is also termed thermal radiation.
3.3.3 Monochromators
The polychromatic radiations emitted from or passed through the sample under investigation are dispersed into
its constituent wavelengths with the help of monochromators. Prisms of suitable material as well as gratings or
a combination of both are used for dispersion purposes. Gratings are especially useful for higher resolution. The
suitable prisms along with their spectral ranges are listed in Table 3.1.
Table 3.1 Prism materials and their spectral ranges.

Prism Material Spectral Range, cm−1
LiF 4000−1400
NaCl 1700−700
KBr 800−400
Csl 400−250
TlI-TlBr 5000−250
CaF2 50000−1200
3.3.4 Detectors
Thermocouples are the most common type of detectors employed. Photoconductive cells, i.e. PbS, PbTe and
semiconductor types such as Bolometers and Golay pneumatic detectors, are generally employed.
3.4 SAMPLE PREPARATION AND SAMPLE CELLS

IR spectra are generally recorded in the gaseous, liquid (neat and solution) and solid phases. The most important
part is the sample preparation. Any solvent is a chemical compound and will absorb in the IR region. Therefore,
the choice of the solvent is decided by the region to be studied. Various solvents and their regions of transparency
are known and the solvents employed are CC14, CS2, or CHC13. 1 to 5 per cent solution of the compound whose
spectrum is to be recorded is prepared in some suitable solvent. The sample cell of some suitable material of
thickness 0.1 to 1 mm has been used. A sample whose spectrum is to be recorded should be dried before use since
water absorbs at ≈3710 cm−1 and 1630 cm−1. If the compound is not soluble in some suitable solvent suggested
then other methods are adopted. Liquid samples are held between plates of suitable window material. The liquid
film thickness varies and its values are 0.025, 0.1 and 0.5 mm. Cells in which path length can be adjusted are also
available these days. Suitable gas cells fitted with optical windows are used to record the gaseous spectra. The
path length of the gas cells varies from few centimetres to metres. Metal cells are used when spectra of gases are
to be studied under high pressure.
Spectra of solids are not recorded because of the problem encountered due to scattering. The important factors
in determining the infrared spectra of solids are (i) the differences in the refractive indices of the material under
study and of the dispersion medium, and (ii) the dimensions of the individual particles under study. The second
effect is especially important in the study of infrared spectra of adsorbed molecules, since excessive scattering by
adsorbent causes a large reduction in I0 value that can be obtained. Also the effect of large particle size is to reduce
both the apparent intensity of absorption and resolution of spectrum. This can be demonstrated by recording the
spectra of CaCO3 (Calcite) in KBr discs as a function of particle size. According to Rayleigh, the amount of scat-
tering, S, is related to the wavelength of incident radiation by the expression
d3
S ∝
λ4
where d is the particle diameter.
Thus, as we move to higher frequencies, the amount of scattering also increases, e.g. surface OD groups
(~2700 cm−1) are preferred over surface OH groups (~3600 cm−1) for infrared spectral studies. It is to be noted
that while discussing the effect of particle size on light scattering it was assumed that the material was embedded
in a medium of equal refractive indices which is not possible in practice. However, whereever possible, the mate-
rial under investigation should be immersed in a liquid or solid whose refractive index is as close as possible to
that of the particle under investigation. This reduces scattering and enhances the attainable resolution, e.g. when
a pressed disk of silica gel is immersed in a fluorolube oil, the transmittance which initially may be 20 per cent is
increased to about 80 per cent. This is because of the effectiveness of oil in reducing the light scattering. The use
of such oil in surface studies is avoided since it may contaminate the surface.
It is seen that maximum transmission is obtained when particle size is less than one micron, i.e. less than the
wavelength of incident light. As the particle size is increased, a stage may come where the particles become so
large that they approach the bulk crystalline material in their absorption properties. The transmittance must go
through a minimum. It is because of this effect that materials such as porous glass and silica-alumina gels are so
unique in their light-transmitting qualities compared with pressed silica discs containing material of apparently
similar fundamental particle size. Therefore, solids are studied in nujol mull which is a saturated paraffin hydrocar-
bon. The C—H region is lost since nujol absorbs in this region and spectrum of nujol is shown in Fig. 3.6. In that
case perfluorinated hydrocarbon, fluorolube is used. Generally one milligram of the solid is homogenised in a drop
of the liquid hydrocarbon. In some cases especially, inorganic, polymers, etc., KBr discs are used. The mixture
of KBr and solid sample is in the ratio of 1:10 or 1:100 is pressed under pressure of the order of 20,000 lb/in2 or
13.7894 × 104 kN/m2 into an optically acceptable disc using a special mould and a hydraulic press. The use of KBr
eliminates the problems encountered in liquid paraffin but a band at 3450 cm−1 called K band due to the OH group
of traces of water always appears.
Polarised infrared spectra of the sample can be Wave number in cm−1
recorded in liquid crystal solvents, provided the
5000
4000
3000
2500
2000
1500
1400
1300
1200
1100
1000
900
800
700
625
background absorption of the liquid crystal sol-

Percent transmittance
100 100
vent does not interfere in the studies. When the
80 80
absorption bands due to the sample fall outside
60 60
the region of the liquid crystals, such studies can
40 40
be carried out easily, e.g. p-(p′-ethoxybenzoxy)
phenyl carbonate is a weak absorber in the fre- 20 20
quency region 1900−2800 cm (OD < 0.15). The

−1 0
2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
0
terminal C ≡ O stretching bands usually occur in Wavelength in microns
the region 1900—2200 cm−1. Hence, this liquid Fig. 3.6 The infrared spectrum of nujol mull.
crystal can be used to determine the orientation of
C ≡ O vibrations.
The major hurdle in the measurement of IR spectra of biologically important molecules is strong absorption
due to water in the regions of spectrum that are of interest to the biochemist. However, the aqueous interference
could be avoided by the following methods: (i) H2O may be replaced by D2O since it does not interfere in the
Table 3.2 Infrared spectra (cm–1) of H2O and D2O.
H2 O D2O Assignment
Very broad band with two main O—X str (ns)
maxima and a shoulder (sh) at 25°C
3920 sh (0.83) 2900 sh (0.60)
3490 (62.7) 2540 (59.8)
3280 (54.4) 2450 (55.2)
2125(3.23) 1515(1.74) Association (nA)
1645(20.8) 1215(16.1) X—O—X bend (n2)
Broad band between 300 cm−1 and 900 cm−1 Libration (nL)
Prominent shoulder on nL band at ∼ 190 cm−1, 30°C Hindered translation (nT)
str-stretch. The quantities within parentheses are the values of extinction coefficients (in units of l03 cm−2 mole)
at the absorption maximum.
regions of spectrum of biological importance. The IR spectra of H2O and D2O are compared in Table 3.2;
(ii) evaporation to dryness of an aqueous layer on a disk of AgCl; (iii) formation of a KBr pellet or oil suspension
with the dried sample. In addition, cooling of the sample to ∼ −180°C sharpens water absorption bands and allows
discrimination of otherwise obscured components. Multiple frustrated internal reflection and refined differential
IR spectroscopy may also be employed to record the spectra of biological molecules.
Further, when recording the spectra, compensation should be allowed in the reference beam. A solvent is used
in solution work while a plane window of the same thickness as that of the sample beam is put across the refer-
ence beam in the case of solid mulls or liquids. The optical windows generally used in recording the spectra in the
various IR regions are listed in Table 3.3.
Table 3.3 Optical materials useful in infrared studies.

Material Useful Range Refractive Comments
(cm−1) Index at 0.54m
NaCl >500 1.54 Widely used, cheap and easily worked; must be kept dry, however.
Soft and easily scratched. Similar to NaCl, but has greater range.
KBr >310 1.53
Used as powder for pressed-disc technique.
Easily worked, but harder than other halides. Expensive.
CsBr >240 1.69
Not hygroscopic.
Insoluble and easy to work in fused form. Useful for high-temperature
SiO2 >2500 ∼1.5
work and in the overtone region.
Al2O3 >1600 1.77 Has high mechanical strength but is expensive.
MgO >1200 1.74 Hard and costly. Can be sealed to high expansion glass.
Very useful in the near infrared because of good dispersion,

LiF >1200 1.38
Scratches easily.
Inert to most chemicals, but tends to be costly. Good from −200°C

CaF2 >900 1.41
to 1000°C. Has low solubility in all except NH 4+ salts.
AgCl >400 2.07 Soft material. Has low melting point, is corrosive and photosensitive.
MgF2 >1333 1.38 Strong. Will withstand temperature of 900°C. Chemically durable.
ZnS >714 2.37 Chemically durable, good up to 800°C. Strong, very useful.
ZnSe >500 2.89 Useful up to 300°C. Soluble in acids.
3.5 CALIBRATION OF SPECTRA

For authentic IR results, the calibration of spectra, are important. The spectra are calibrated using standards such
as methane, water, carbon monoxide, ammonia, indene, toluene and some substituted benzenes. The most com-
monly used calibrant is polystyrene with bands at 3.509, 6.238 and 11.035 microns.
3.6 BANDWIDTH
The shape of an IR curve may be approximated by a Lorentzian curve of the form
I0 a
ln = (3.3)
I t (v v 0 ) + b2
2
where ln I0/It is the optical density, a and b are constants. For maximum optical density, v = v0 at the band centre,
⎛I ⎞ a
ln ⎜ 0 ⎟ = 2 (3.4)
⎝ It ⎠ v b
0
Then, bandwidth (recall from Chapter 1) is defined as the full width Δν1/2 or (Τ)
of the absorption band at half the maximum absorbance or extinction
(Fig. 3.7) and is given by
Δv1/2 = 2b (3.5)
The above formulae hold if the instrumental inadequacies are ignored.
If the slit function is incorporated in Eq. (3.3), a difference is noticed
between the observed and theoretical values of Δv1/2. Δv1/2 is independent Fig. 3.7 Spectral band having width Δν1/2 at
half height.
of the number of molecules particularly in an absorption process while
peak height is dependent on it. Therefore, it is good to report Δv12 for standard peak heights. Under standardised
conditions, Δv1/2 is as useful as the extinction coefficient in identifying the molecule.
3.7 POLARISATION OF INFRARED BANDS

The degree of absorption polarisation rir is measured in term of the ratio of the optical densities (OD)
OD||
ρir = (3.6)
OD ⊥
where OD|| and OD⊥ are the optical densities obtained if the polarisation direction of the measuring infrared radia-
tion is oriented parallel and perpendicular to the optic axis or the electric field respectively. rir > 1 for parallel
while < 1 for perpendicular bands.
3.8 VIBRATIONAL SPECTRA OF DIATOMIC MOLECULES

The observed infrared vibrational spectra of diatomic molecules such as HC1, CO, HBr, HI, NO, etc., consist
of a series of equally distant lines but with diminishing intensity. The vibrational spectra may also be obtained
theoretically, from mathematical models describing the vibrational motion of the molecules. The three well-
known models to predict the vibrational spectra of diatomics, are (i) the classical harmonic vibrational model;
(ii) the quantum mechanical harmonic vibrational model, and (iii) the quantum mechanical anharmonic
vibrational model. Let us see how these models have been developed and what has led to their failure and
success?
3.9 DIATOMIC MOLECULE AS A HARMONIC VIBRATOR

This problem is treated by considering the diatomic molecule to be always in one of its lowest vibrational state at
room temperature, and on absorption of radiation, it behaves as a simple harmonic oscillator. Classically, assume
the molecule to be a dumb bell and let re be the equilibrium distance between the atoms of masses m1 and m2 as
shown in Fig. 3.8.
re
m2 m1
r2 r1
Cm
r
Fig. 3.8 The dumb-bell model of a diatomic
molecule.
As a result of absorption, the atoms oscillate about their fixed mean positions and let ± Δr = r − re be the amount
of bond-distance difference from the equilibrium distance.
The potential energy U of the system is then given by
1
k (r re )
2
U (3.7)
2
where k is a constant. Equation (3.7) represents a parabola shown in Fig. 3.9. From
classical mechanics, the restoring force
⎡1 2⎤
−d
d k ( r re ) ⎥
f = ⎣2 ⎦ U(r)
d ( r − re )
f = − k (r − re) (3.8)
Thus the constant k is the restoring force required to bring the molecule back re
r
to its equilibrium position and is called force constant. It is generally expressed in
Fig. 3.9 Potential energy inter-
millidynes/Å. nuclear distance curve for simple
If Δ r1 and Δr2 are displacements of the masses from their equilibrium positions harmonic oscillator.
then
(r2 − r1) = r − re Since force = mass × acceleration
Therefore,
d 2 r1 m 2d 2 r2
k ( r2 − r1 ) m1 and − k ( 2 1)
r − r = (3.9)
dt 2 dt 2
The force on m1 and m2 is in the positive and negative directions respectively when r2 − r1, is positive. The solu-
tions of set of equations of motion given in Eq. (3.9) are
r1 = A1 cos (2pv + j) and r2 = A2 cos (2pv + j) (3.10)
These equations describe the simple harmonic motion for both masses, each oscillating with the same fre-
quency v and phase constant j along the molecular axis, but with different maximum amplitudes A1 and A2.
Combining Eqs (3.9) and (3.10),
k(A2 − A1) = −m1A14p2v2 (3.11)
k(A2 − A1) = m2A2 4p2v2

Rearranging Eqs (3.11), we get
A1 −k A 1 4π 2 m 2 v 2 − k
= 2 and = (3.12)
A 2 4π m1 v 2 − k A2 −k
Eliminating A1/A2 from the set of equations in Eq. (3.12)
⎛ k ⎞ ⎛ m + m2 ⎞
v4 v2⎜ 2⎟⎜ 1 =0
⎝ 4π ⎠ ⎝ m1m 2 ⎟⎠
⎡ k ⎛ m + m2 ⎞ ⎤
or v 2 ⎢v 2 − 2 ⎜ 1 ⎥=0 (3.13)
⎣ 4π ⎝ m1m 2 ⎟⎠ ⎦
Equation (3.13) has two solutions: (i) v = 0, i.e. A1 = A2 from Eq. (3.12) and for any time t, r1 = r2
from Eq. (3.10). This corresponds to translational motion of masses without vibration in the molecular axis
direction.
1 ⎛ 1 1 ⎞ 1 k
(ii) v k + ⎟ = (3.14)
2π ⎝ m1 m 2 ⎠ 2π μm
m1m 2
where μm =
m1 + m 2
It represents the vibration of masses without translational motion. Further, it is evident from Eq. (3.14) that
v is independent of amplitude of vibration.
Combining Eq. (3.14) with Eqs (3.12) and (3.10)
A1 m
=− 2 (3.15)
A2 m1
r1 − m 2
or at any time =
r2 m1
This indicates that centre of gravity of nuclei remains stationary during the vibration. Thus, according to classical
model of the harmonic diatomic vibrator, all energies are possible with frequency of vibration
1 k
v=
2π μm
However, this is contrary to the experimental facts. The vibrational spectrum of diatomics is not continuous
but has a discrete structure. The discrete character of the vibrational spectrum could be explained on the basis of
quantum mechanical treatment of vibrational motion of the diatomics.
Quantum mechanically, the vibrational motion of the diatomic molecule is reduced to the motion of a single
particle of reduced mass mm. Thus, the total energy of vibration
Evib = Kinetic energy + Potential energy
1 d ( r − re ) 2 1
= μm + k ( r − re ) 2 (3.16)
2 dt 2
The Schrödinger equation then takes the form
d 2 Ψ( ) 8π 2 μ m ⎡ 1 2⎤
+ EV k (r re ) ⎥ Ψ( r ) = 0 (3.17)
dr 2 h2 ⎣ 2 ⎦
The solution of Eq. (3.17) gives [For solution see Appendix 3.1A]
⎛ 1⎞
EV hν V + ⎟ (3.18)
⎝ 2⎠
⎛ 1⎞ ⎡ EV ⎛ 1⎞ ⎤
= hcω e V + ⎟ ⎢ G= = ω e ⎜V + ⎟ , cm −1 ⎥
⎝ 2⎠ ⎣ hc ⎝ 2⎠ ⎦
1 k k ( millidynes/A )
where ωe d ω e (cm −1 ) = 1303.16 (3.19)
2π c μm μ m (amu )
Here v/we, is the frequency of vibration in Hz/cm−1 and V, the vibration quantum number, and can have values
0, 1, 2, 3,....
3.9.1 Spacing Between Two Successive Vibration Levels

⎛ 1⎞
EV +1 kcω e V + 1 + ⎟ (3.20)
⎝ 2⎠
⎛ 1⎞
EV hcω e V + ⎟ (Eq. 3.18)
⎝ 2⎠
From Eqs (3.20) and (3.18) we obtain
EV + − EV
Δd = = ωe (3.21)
hc
Thus, the spacing between successive harmonic levels is always constant.
3.9.2 Selection Rules of Harmonic Vibrator

The selection rules have been deduced quantum mechanically from the solution of the integral.
M = ∫ ( Ψi*M̂ Ψj ) dt (3.22)
and are the following:

(i) Diatomic molecules must possess permanent dipole moment, i.e. M ≠ 0.
(ii) The oscillating dipole moment should not be zero, i.e. Mt ≠ 0.
(iii) Only those transitions are permissible for which ΔV = ±1. (3.23)
Let the transition electric–dipole–moment integral be defined as
M ij = ∫ Ψi*M̂ Ψj dt (3.22a)
Here, i corresponds to V ″ and j corresponds to V ′.

For pure rotational motion, M is constant while for vibrational motion it changes because of change in bond
length, provided the molecule has a permanent dipole moment. Thus, the dipole moment of a oscillating molecule
may be expressed by a Maclaurin series about Q = 0 by
⎛ dM ⎞ 1 ⎡ d 2M ⎤
MQ M Q =0 + ⎜ Q + ⎢ ⎥ Q + ....
2
(3.22b)
⎝ dQ ⎟⎠ Q = 0 2! ⎣ dQ 2 ⎦Q = 0
where Q is the weighted average Cartesian displacement coordinate Q, i.e. Q = r − re. Combining Eqs (3.22a) and
(3.22b) and ignoring the third and higher terms of Eq. (3.22b), we get
⎡ ⎛ dM ⎞ ⎤
ij ∫ Ψ* ⎢⎢ M
i Q =0 +⎜ ⎟ Q⎥
⎝ dQ ⎠ Q = 0 ⎥⎦
j dτ
⎣
⎛ dM ⎞
= M Q = + ∫ Ψi* ⎜ Q Ψj dτ (3.22c)
⎝ dQ ⎟⎠ Q = 0
Equation (3.22c) reveals that there will be no transition if the permanent dipole moment of molecule does not
change during a vibration. The second term gives rise to the selection rule ΔV = ±1 and is responsible for the fundamen-
⎛ d 2M ⎞
⎝ dQ 2 ⎟⎠ Q = 0 ∫
tal vibrational mode in the IR spectrum of the molecule. The inclusion of higher terms, i.e. ⎜ Ψi* 2Ψ j dτ ,
⎛ dM ⎞ ⎛ d 2M ⎞ ⎛ d 3M ⎞
etc., lead to the selection rules ΔV = ±2, ±3,.... Since ⎜ >> >> ⎜ 3 ⎟ . . ., the additional
⎝ dQ ⎟⎠ Q = 0 ⎜⎝ d 2Q ⎟⎠
Q
⎝ d Q ⎠ Q =0
lines due to overtones and combination bands will in general be very weak, i.e. I (V = 0 → 1) >> I (V = 0 → 2)
>> I (V = 0 → 3).
3.9.3 Transition Between Two Successive Vibrational Levels

Let V ′ and V ″ be the vibrational quantum numbers of the lower and upper levels respectively and EV′ and EV′′ be
the energies of the respective levels. According to Eq. (3.18),
EV′′ = hcwe(V′′+ 1/2) (3.24)
EV′ = hcwe(V′+1/2) (3.25)

Lumping together Eqs (3.24) and (3.25),
EV EV ′
ΔG = = ω e (V ″ − V ′ ) (3.26)
hc
Applying the selection rule, ΔV = +1 to Eq. (3.26) we obtain

ΔG = we
Thus, the spectrum of energy values according to Eq. (3.18) is a discrete one, as distinct from the continuous
spectrum permitted by classical harmonic oscillator. The difference between energy levels has always a constant
value we. The other important result from quantum mechanical oscillator is that the lowest energy of vibrational
level is hv0/2 but not zero. This is called the zero point energy of the harmonic vibrator which means that the
molecule is always vibrating.
To summarise, according to harmonic vibrator model of the diatomics, the vibrational spectrum consists
of a single band as predicted by the selection rule ΔV = ±1. Recalling that the experimentally observed
spectrum of diatomics consists of a series of nearly equally spaced lines with diminishing intensity, the
discrepancy between the experimental and theoretical spectra may be because of anharmonic nature of the
diatomic vibrator.
Problem 3.1: Determine the force constant for CO vibrator provided we = 2170 cm−1.
Solution According to Eq. (3.19),
k=
w e2
×
m1m 2
=
(2170) × 12 × 16
2
(1303.16)2 m1 + m 2 (1303.16 )2 28
= 19.01 millidynes/Å
Problem 3.2: Determine the frequency of OD vibrator if we for OH vibrator is 3738 cm−1.
Solution According to Eq. (3.19), we write
we(OH) = 1303.16 k OH ÷ μOOH
we(OD) = 1303.16 k OD ÷ μOOD
Lumping together these expressions,
we( ) μOOH
= [i k k ]
we( ) μOOD
we( ) = we( ) μOH ÷ μOOD
16 18
= 3738 ×
17 32
= 2720 cm−1
3.10 DIATOMIC MOLECULE AS

AN ANHARMONIC
VIBRATOR 13
12
11
Potential energy (cm−1)

The potential function of an anharmonic vibrator 10
9
may be represented in several ways. 8
(i) The potential energy may be described by power 7
series in k, 6
5 Do De D1
1 4
U(r) = kl (r−re)2 + a(r−re)3 + b(r−re)4 +… (3.27)
2 3
2
where a and b are constants, and (ii) the potential
1
energy is more or less adequately described by the
Morse potential function. The Morse function gives V=0
an idea of the dependence of the potential energy of
Bond distance (Å)
a molecule on the bond length as shown in Fig. 3.10.
Fig. 3.10 Morse potential energy-internuclear distance curve for an
It is clear from Fig. 3.10 that U(∞) − U(0) = De.
anharmonic oscillator.
The Morse function for a diatomic vibrator may
now be written as
U(r) = De [1− exp (−aa (r − re ))]2 (3.28)
where De is the depth of potential well, aa is adjustable constant and by changing aa, the degree of curvature on
the right-hand side of potential well can be changed.
hcw
we
De = (3.29)
4xe
α a = 8π 2 μ mwe cxe /h (3.30)
xe is the anharmonicity constant.

U(r) = 0 at r = re.
This is the reference point to measure U according to Eq. (3.28)
The Morse function may also be written as
U ( r ) = De ⎡⎣e −2αa (r re ) − 2e − αa (r re ) ⎤⎦ (3.31)
U(r) = 0 at r = ∞
This is the reference point to measure U according to Eq. (3.31).
The Schrödinger wave equation for a diatomic anharmonic vibrator may now be written as,
d 2Ψ ( ρ ) 8π 2I e ⎡ r re ⎤
+ 2 ⎡⎣ EV U ( ρ )⎤⎦ Ψ ( ρ ) = 0, ⎢ ρ = (3.32)
d 2ρ h ⎣ r ⎥⎦
The solution of the wave equation (3.32), gives the expression for the energy of the vibrational levels of an
anharmonic oscillator. (Appendix 3.2A)
2 3
⎛ 1⎞ ⎛ 1⎞ ⎛ 1⎞
EV hhcwe V hcw e x e V + ⎟ + hcw y e V + ⎟ + ...
⎝ 2⎠ ⎝ 2⎠ ⎝ 2⎠
2 3
⎛ 1⎞ ⎛ 1⎞ ⎛ 1⎞
GV we V we x e V + ⎟ + w y e V + ⎟ + ... (3.33)
⎝ 2⎠ ⎝ 2⎠ ⎝ 2⎠
where we >> wexe >> we ye …, etc. In practice, only the first three terms are within the precesion of the spectrometers.
For V = 0
hcw e hcw e x e hcw e y e
EV ,0 = − +
2 4 8
hcwe ⎛ x e y e ⎞
= 1− + ⎟
2 ⎝ 2 4⎠
we ⎛ x e ye ⎞
or GV ,0 = 1− + ⎟ (3.34)
2 ⎝ 2 4⎠
EV = 0 is termed as the zero point vibrational energy of the anharmonic vibrator.
Combining Eqs (3.33) and (3.34), we get the expression for the energy levels of the anharmonic oscillator with
respect to the zero point energy.
GV(V,0) = GV− GV,0 = w0V− w0x0V 2+ w0y0V 3 (3.35)
3
where w0 w e − w e x e + w y e + ... (3.36)
4
3
w 0 x 0 = w e x e − w y e + ... (3.37)
2
w y 0 = w y e + .... (3.38)
Problem 3.3 Using the Morse function,

aρ 2
U ( ρ) D(
D( e ) ,
r − re
where ρ = ; prove that in case of a diatomic vibrator w2 ~ D.
r
aρ 2
Solution Given U ( ρ ) D(
D( e ) ,
For small (r) (small vibrations), the function can be written as
2
⎡ ⎧ (aρ ) 2 ⎫ ⎤
U ( ρ) D ⎢1 ⎨1 − aρ + − ⎬ ⎥ =D (aρ ) 2 D
Da2 ρ 2
⎣ ⎩ 2 ⎭⎦
Thus, the quasi-elastic force

F = −2 Da2 r/re = −kre r
where k Da2 / re2 .

Comparing U(r) and k, we get
krre2 ρ 2
U ( ρ) D 2ρ 2 =
Da
2
Solution of Schrödinger wave equation for a diatomic vibrator gives
⎛ 1⎞
EV hcw V + ⎟
⎝ 2⎠
where
1
w= k / μm
2π c
Substituting the value of k, we write

1 2Da2 a
w= = 2D / I e
2π μ m re2
2π c
a2
or w2 = D.
2π 2I e c 2
a2
Since the factor is very small, the expression can safely be approximated to
2π 2I e c 2
w2 ≅ D.
The greater the D, the greater should be the vibrational frequency. Experimental observations show that the
relationship is qualitatively fulfilled in practically all cases.
3.10.1 Spacing Between Two Adjacent Vibrational Energy Levels

From Eq. (3.33), the energy of the two successive vibrational levels of an anharmonic oscillator may be written as
2
⎛ 1⎞ ⎛ 1⎞
EV h e V
hcw hcw e x e V + ⎟
⎝ 2⎠ ⎝ 2⎠
2
⎛ 3⎞ ⎛ 3⎞
EV hhcw e V hcw e x e V + ⎟ (3.39)
⎝ 2⎠ ⎝ 2⎠
From Eqs (3.33) and (3.39), we write
EV + − EV
Δd = = w e ⎡⎣1 − 2x e (V + 1)⎤⎦ (3.40)
hc
where Δd is the spacing between the two successive vibrational energy levels.
For V d ⎫
⎪
0 w e (1 2xe ) ⎪⎪
⎬ (3.41)
1 w e (1 4 xe ) ⎪
⎪
2 w e (1 6 xe ) ⎪⎭
Set of Eqs (3.41) indicates that the spacing between two consecutive anharmonic vibrational levels decreases
as the vibrational quantum number grows.
3.10.2 Selection Rules

The selection rules worked out in the usual way, to predict the spectrum of real molecules, are (i) M ≠ 0, (ii) Mt ≠ 0,
and (iii) ΔV = ±1, ±2, ±3,..., IΔV = ±1 >> IΔV = ±2 >>IΔV = ±3 .
3.10.3 Transition Between Two Adjacent Vibrational Levels

Let V′ and V″ be the vibrational quantum numbers of the lower and upper vibrational levels respectively and Ev ′.
and Ev″ be the respective energies. From Eq. (3.33) we may write
2
⎛ 1⎞ ⎛ 1⎞
EV hcw e V ′
hcw hcw e x e V ′ + ⎟
h (3.42)
⎝ 2⎠ ⎝ 2⎠
2
⎛ 1⎞ ⎛ 1⎞
EV h e V″
hcw h exe V ″ + ⎟
hcw (3.43)
⎝ 2⎠ ⎝ 2⎠
Using Eqs (3.42) and (3.43) we obtain,

EV − EV ′
ΔG = (3.44)
hc
= (V″−V′)[we (1−xe) − wexe (V″+V′)]
Applying selection rule ΔV = +1 we get
ΔV = V″−V′ = +1, V″ = 1 + V′
Therefore, for ΔV = +1, ΔG = we(l − xe) − wexe (2V′ + 1) Consequently for,
V ′ Transition
a ΔG ⎫
⎪
0 0 → 1 w e (1 2x e ) ⎪⎪
⎬ (3.45)
1 1 → 2 w e (1 4 x e ) ⎪
⎪
2 2 → 3 w e (1 6 x e ) ⎪⎭
Thus, it is evident from the set of Eqs (3.45) that as the vibrational quantum number increases, the anhar-
monicity also increases as expected. The transitions 1 → 2, 2 → 3,... are observed only at higher temperatures.
Therefore, vibrational bands due to such transitions are termed hot bands. At higher temperature, the intensity of
0 → 1 transition depletes because under these conditions, the molecules jump to higher levels. However, the band
due to 0 → 1 transition will still be more intense as compared to the hot bands. A comparison of Eqs (3.41) and
(3.45) suggests that separation between successive energy levels (cm−1) is equal to the transition energy between
the corresponding successive energy levels.
For ΔV = +2, V″ = V′ + 2,
ΔG = 2we (l − xe) − 4wexe (V′ + 1) (3.46)
Since V′ = 0, 1, 2,...
Therefore, for V′ = 0
ΔG0 → 2 = 2we (1−3xe)
for V′ = 1
ΔG1 → 3 = 2we (1−5xe) (3.47)
and for V′ = 2,
ΔG2 → 4 = 2we (1−7xe).

The 1 → 3 and 2 → 4 transitions occur only at higher temperatures and are termed as hot overtones. Similarly,
we arrive at the following set of equations for ΔV = +3.
ΔG0→3′ = 3we(1−4xe )
ΔG1→4′ = 3we(1−6xe ) (3.48)
ΔG2→5′ = 3we(1−8xe )
Thus, at room temperature, the following bands are observed in the vibrational spectrum of a diatomic mol-
ecule.
1 ΔG 0→1 = we (1 2x e )
v2 ΔG 0→ 2 = 2w
w e (1 3x e ) (3.49)
v3 ΔG 0→3 = 3w e (1 4 x e )
Here, ν1 ν 2 , ν 3 , are the wave numbers of the fundamental absorption band, first and second overtones respec-
tively. For small values of V, the approximation of harmonic oscillator is quite good and becomes significant at
large values of V. Thus, for small values of V, the bands will appear at we, 2we, 3we,.... The fundamental band is
always intense as compared to overtones and intensity follows the trend, IΔV = ±1 >> IΔV = ±2 >> IΔV = ±3.
3.10.4 Determination of Electrical Anharmonicity

The harmonic oscillator model can account for the fundamental vibration that occurs between nonvibrating state
and the next vibrationally excited state, but it cannot account for the presence of overtones and combination
bands. Overtones and combination bands may be explained by introducing mechanical anharmonicity into the
potential energy, i.e. by expanding the potential energy of an anharmonic oscillator about the equilibrium position
(Q = 0) in the Maclaurin series
1 ⎛ d 2U ⎞ 1 ⎛ d 3U ⎞
UQ = ⎜ ⎟ Q 2
+ ⎜ 3⎟
Q 3 + ... (3.50)
2! ⎝ dQ ⎠ Q
2
3! ⎝ dQ
Q ⎠ Q =0
and in which the energy scale is chosen in such a way that U(0) = 0. The quantum mechanics of an anharmonic
oscillator has been described in Appendix 3.2 A and it expresses the total energy of the anharmonic oscillator by
the expression
2 3
⎛ 1⎞ ⎛ 1⎞ ⎛ 1⎞
GV we V w e x e V + ⎟ + w y e V + ⎟ + .... (Eq. 3.33)
⎝ 2⎠ ⎝ 2⎠ ⎝ 2⎠
GV (V, 0) = w0V−w0x0V 2 + w0y0V 3

where we is the harmonic oscillator frequency, we xe and we ye are anharmonicity constants, we ye << we xe << we.
The net effect of expressing the potential energy of anharmonic oscillator by Eq. (3.50) is that the instantaneous
dipole moment, MQ of an oscillating molecule may be expressed by a Maclaurin series about Q = 0 by
⎛ dM ⎞ 1 ⎛ d 2M ⎞
MQ M Q =0 + ⎜ ⎟ Q+ ⎜ 2 ⎟
Q 2 + ... (3.51)
⎝ dQ ⎠ Q = 0 2! ⎝ dQ ⎠ Q = 0
where Q is the weighted average Cartesian displacement coordinate.
The first term in Eq. (3.51) expresses the permanent dipole moment and is responsible for the pure rotational
infrared spectrum only. The second term is responsible for the appearance of vibrational infrared spectrum while
the higher terms provide intensities to overtones and combination bands.
The energy difference between two successive vibrational levels V and V + 1 is
⎛ 1⎞
ΔG V + ⎟ = G (V + ) − G (V ) = G (V + )(V = 0 ) − G (V ) (V = 0 ) (3.52)
⎝ 2⎠
which because of Eq. (3.33) yields
2 2
⎛ 1⎞ ⎛ 1⎞ ⎛ 1⎞ ⎛ 1⎞ ⎛ 1⎞
ΔG V + ⎟ = w e V + 1 + ⎟ − w e x e V + 1 + ⎟ + ... −w e V + ⎟ + w e x e V + ⎟ + ...
⎝ 2⎠ ⎝ 2⎠ ⎝ 2⎠ ⎝ 2⎠ ⎝ 2⎠
(3.53)
= we−2wexe−2wexeV
= w0−2w0x0−2w0x0V
Similarly, the energy difference between vibrational levels V + 1 and V + 2 would be
ΔG (V + 3/2) = we − 4wexe − 2wexeV (3.54)
so that
Δ2G (V+1) = ΔG(V+3/2)−ΔG(V+1/2) (3.55)
= − 2wexe = −2w0x0
However, in most of the cases the frequency of the fundamental and first overtone is available. In such cases
the energy corresponding to the lowest vibrational level (V = 0) is taken as zero so that energy difference between
the vibrational levels 0 and V is
ΔG ⎛ 1⎞ = G (V ) − G (V = 0 ) = w eV − w e x eV 2 + ... (3.56)
⎜⎝V + ⎟⎠
2
Consequently, the frequency of the fundamental, G0(1) and first overtone G0(2) would be expressed by
G0(1) = w0− w0 x0 (3.57)
and G0(2) = 2 w0 − 4 w0 x0 (3.58)

Hence, both w0 and w0 X0 can be obtained from Eqs (3.57) and (3.58) once the experimental values of G0 (1)
and G0 (2) are known.
Problem 3.4: The laser Raman and infrared spectra of tribromoacetyl chloride reveal that the position of the
C = O fundamental vibration is at 1772 cm−1 and that of its first overtone is at 3540 cm−1. Calculate the anharmo-
nicity constants for the C = O stretching frequency.
Solution
Here G0(1) = 1772 cm−1
and G0(2) = 3540 cm−1
Consequently, Eqs (3.57) and (3.58) yield
1772 cm−1 = w0 − w0x0
3540 cm−1 = 2w0 − 4w0x0

Hence, w0 = 1774 cm−1 and w0x0 = 2 cm−1.
Problem 3.5: The low-temperature infrared study of the self-association of thiolacetic acid reveals that the
fundamental and the first overtone for a free and associated S—H vibrations appear at 2588 cm−1 and 5078 cm−1
and at 2530 cm−1 and 4910 cm−1, respectively. Calculate the anharmonicity constants for both the free and the
associated S—H groups.
Solution For free S—H group, we have
G0(l) = 2588 cm−1
G0(2) = 5078 cm−1

From expressions (3.57) and (3.58), we write
2588 = w0 − w0x0
5078 = 2w0 − 4w0 x0

Therefore, w0x0 = 49.5 cm−1 and w0 = 2637 cm−1.
Similarly, the constants for associated S—H group have been found to be w0x0 = 75 cm−1 and w0 = 2605 cm−1.
Problem 3.6: Determine the values of anharmonic constants w0x0 and wexe from the following infrared spectral
data on HCl,
V 0 1 2 3 4
1
v( ) 0 2885.9 5668.2 8347.4 10923.7
Solution In order to determine w0x0 and wexe, the infrared spectral data are arranged as follows.
V ν (cm−1) ΔG(cm−1) −Δ2G(cm−1)

0 (0)
1 2885.9 2885.9 103.6
2 5668.2 2782.3 103.1
3 8347.4 2679.2 102.9
4 10923.7 2576.3
Δ2G is constant within the permissible limits. The average value of Δ2 G = −103.2 cm−1. Using Eq. (3.55)
103.2
w e xe = w 0x 0 = = 51.6 cm −1
2
The value of w0 can be obtained either graphically or analytically. Graphically, the intercept of the plot of V
against ΔGV+1/2 [Eq. (3.53)] will correspond to w0. To determine w0 analytically, use is made of Eqs (3.57) and
(3.58), etc.
G0 (1) = w0− w0 x0
G0 (2) = 2 w0 − 4w0x0
2885.9 = w0−51.6
5668.2 = 2w0 − 4 × 51.6
w0 (1) = 2937.5 cm−1

w0 (2) = 2937.3 cm−1
w 0 (1) w 0 ( 2)
w0 =
2
= 2937.4 cm−1
Therefore, from Eq. (3.36)
2937.4 = we −51.6
we = 2989 cm−1
3.10.5 Maximum Value of Vibrational Quantum Number and Energy

The maximum value of the vibrational quantum number, Vmax is calculated from Eq. (3.40) as follows.
Δd = we[1− 2xe(V+1)]
For the maximum value of quantum number, i.e. V = Vmax. Δd, the spacing between consecutive levels will tend
to zero, i.e. Δd → 0. Consequently,
1 − 2x e
V max = (3.59)
2x e
The distinct nature of the spectrum vanishes and continuous spectrum is obtained.
The vibrational energy of the maximal vibrational quantum level is determined from Eqs (3.59) and (3.33).
EV , max =
hcw e
4x e
( − xe )
Since x e2 << 1
Therefore,
hcw e
EV, max = = De (3.60)
4x e
3.10.6 Bond Energy of Diatomic Molecule

The Bond Energy (BE) is the product of the difference between the vibrational energies on the maximal and
zeroth vibrational levels and the Avogadro number, NA.
Therefore, using Eqs (3.60) and (3.34),
BE = (EV,max − EV,0)NA
⎡ hcw e hcw e ⎛ xe ⎞ ⎤
=⎢ − ⎜⎝1 − ⎟⎠ ⎥ N A
⎣ 4x e 2 2 ⎦
hcw e ⎛ 1 ⎞
= N A xe + − 2⎟ (3.60a)
4 ⎝ xe ⎠
Problem 3.7: (a) Calculate the maximal vibrational quantum number for C1F, if we = 786.15 × 102 m−1.
xe = 7.84 × 10−3 and re = 1.63 × 10−10 m

(b) Using the data from part (a) determine the vibrational energy of molecule ClF on the zeroth and maximal
vibrational quantum levels. Determine also the bond energy. (c) With the aid of data from part (b), find the vibra-
tional energies of the molecule ClF on the vibrational levels V = 0, 1, 2, 3, 5, 8, 12, 20, 30 and 63; (d) Using the
data from part (c) compute the minimal and maximal distances between the atoms Cl and F on the vibrational
levels V = 0, 1, 2, 3, 5, 8, 12, 20, 30 and 63.
Solution (a) According to Eq. (3.59)
1 − 2x e 1 − 2 × 7.84 × 10 −3
V max = =
2x e 2 × 7.84 × 10 −3
0.9843
= = 63
0.01568
(b) The maximum value of vibrational energy corresponding to Vmax = 63 can be determined from Eq. (3.60).
hcw e 6.62 10 −34 3 × 108 × 786.15 × 10 2

EV, max = = 8 × 10 −20 J
= 49.786
4x e 4 7.84 × 10 −3
The zero point vibrational energy is determined from Eq. (3.34).
hcw e ⎡ xe ⎤ 6.62 × 10 − 34 × 3 × 108 × 786.15 × 10 2 ⎡ 7 84 × 10 −3 ⎤

EV , 0 = ⎢1 − = ⎢1 − ⎥
2 ⎣ 2 ⎥⎦ 2 ⎣ 2 ⎦
= 0.777 × 10−20 J
Therefore, by definition from Eq. (3.60a)

BE = (49.786 − 0.777) × 10−20 × 6.023 × 1023
= 295.18 × 103 J mole−1 = 295.18 kJ mole−1

(c) Here
we = 786.15 × 102 m−1
xe = 7.84 × 10−3
c = 3 × 108 m/s
h = 6.62 × 10−34 J s
In order to determine the vibrational energies on the different vibrational levels, determine the terms hcwe and
hcwexe .
hcwe = 786.15 × l02 × 3 × 108 × 6.62 × 10−34
= 156.12 × 10−22 J
hcwexe = 156.12 × 10−22 × 7.84 × 10−3 = 1.22 × 10−22 J

Now by making use of Eq. (3.33), i.e.
2
⎛ 1⎞ ⎛ 1⎞
EV V + ⎟ hcw e V + ⎟ hcw e x e
⎝ 2⎠ ⎝ 2⎠
EV for different values of V have been determined and are listed in Table 3.4.
Now we see from Table 3.4 that as the vibrational quantum number increases the contribution of anharmo-
nicity also increases. Also, we see that for the vibrational quantum numbers 0 and 1, the vibrations are almost
harmonic.
(d) Solve Eqs (3.33) and (3.28) together.
2
⎛ 1⎞ ⎛ 1⎞
⎜⎝V hcw e − V hcw e x e De [1 − exp(α a ( r re ))]2
2⎠ ⎝ 2⎠
Compute the constants De and aa, using Eqs (3.29) and (3.30), respectively.
Table 3.4 Vibrational energy of CIF vibrator on different vibrational levels.
2
⎛ 1⎞ ⎛ 1⎞
V ⎜⎝V hcw e × 10 +22 J ⎜⎝V hcw e x e 10 +22 J EV × 10+22 J
2⎠ 2⎠
0 78.06 0.3060 77.76

1 234.19 2.75 233.44
2 390.32 7.65 392.67
3 546.45 14.99 531.46
5 858.71 37.02 821.69
8 1327.09 88.43 1231.66
12 1951.61 191.25 1760.36
20 3200.65 514.40 2686.25
30 4761.94 1138.67 3623.27
63 9913.62 4919.34 4994.27
hcw e 156.12 × 10 −22

De = = = 4.97 × 1
100 −19 J
4x e 4 × 7.84 × 10 −3
8 × (3.14 ) × 3 × 108 × 786.15 × 10 2 × 7.84 × 10 −3 × 20.42 × 10 −27

2
8π 2 μm
αa = = = 2.1213 × l010 m−1
h 6 62 × 10 −34
EV
Transform Eq. (3.33) to 1 − ey =
De
where y = − aa(r − re).

The values of each factor in the transformed equation are recorded in Table 3.5. In Columns 7 and 8 of Table 3.5,
y1 y2
r1 re − and r2 re −
αa αa
are the maximal and minimal distances between the atoms Cl and F on the different vibrational levels respectively.
Table 3.5 Maximal and minimal distances between CI and F atoms on the different vibrational levels.
V ± (1 − e y) e y1 e y2 y1 y2 r1 × 1010m r2 × 1010m
0 0.125 0.8749 1.1251 −0.1336 0.1178 1.6927 1.5745
1 0.216 0.7833 1.2167 −0.2442 0.1961 1.7446 1.5381
2 0.281 0.7189 1.2811 −0.3300 0.2477 1.7851 1.5136
3 0.327 0.6730 1.3270 −0.3960 0.2829 1.8167 1.4971
5 0.406 0.5934 1.4066 −0.5219 0.3412 1.8751 1.4697
8 0.497 0.5022 1.4978 −0.6888 0.4040 1.9539 1.4400
12 0.595 0.4049 1.5951 −0.9041 0.4669 2.0557 1.4103
20 0.735 0.2648 1.7352 −1.3288 0.5511 2.2560 1.3703
30 0.853 0.1462 1.8538 −1.9228 0.6172 2.5337 1.3396
63 1.000 0 2.0 0.6931 1.3033
y1 y2 −10
where r1 re − , r2 = re − and re = 1.63 10 m.
αa αa
3.10.7 Determination of Dissociation Energy of Diatomics

The dissociation energy of diatomic molecules can be determined from the spectroscopic data by the following
three methods:
(i) The values of D0 and De can be determined from the vibration constants of the diatomic molecule, i.e. (Eq. 3.60).
w e2 w 02
De = ; D0 =
4w e x e 4w 0 x 0
These relations do not give accurate values of De and D0.
Problem 3.8: Determine the values of De and D0 for FO molecule if
we = 1060 cm−1 and xe = 14.6 × 10−3

Solution From Eq. (3.60)
1060 × 1060
De = = 18.15 × 103 cm −1
4 × 1060 × 14.6 × 10 −3
= (18.15 × 103 cm−1) × (3×1010 cm/s) × (6.62×10−34 Js) × (6.023 × 1023 mole−1)
= 2171.12 × 102 J mole−1 = 217.112 kJ mole−1

By Eq. (3.36)
w0 = we − wexe = 1060 − 1060 × 14.6 × 10−3 = 1045 cm−1
Since wexe = w0x0 (By Eq. 3.37)

Therefore,
1045 × 1045
D0 = −3
= 17.64 × 103 cm −1 = 211.018 kJJ molee −1
4 × 1045 × 14.6 10
(ii) The band convergence method is the most accurate method since the discrete spectrum is joined by the con-
tinuous spectrum. The position of convergence limit corresponds to De and D0. The method fails if convergence
limit is not achieved. In such cases, the Birge–Sponer method is used.
We have already proved that the maximum value of vibrational quantum number is given by
1 − 2x e
V max = (Eq. 3.59)
2x e
The vibrational energy corresponding to Vmax will then correspond to De as shown earlier.
EV ,max =
hcw e
4x e
( − xe ) (3.60b)
hcw e
= = De (3.60c)
4x e
It is evident from this expression that vibrational energy of a diatomic molecule cannot exceed its dissociation
energy.
(iii) The Birge–Sponer method gives accurate values of D0 and De if large number of vibrational bands is
observed. This method is based on the transition energy between two successive quantum vibration levels.
ΔG = we − 2xewe (V+1) (Eq. 3.44)
A graph is plotted between ΔG (cm−1) and V. If the molecule is a Y

Morse oscillator, such a plot will be a straight line. Otherwise it is
always curved as shown in Fig. 3.11.
ΔG
The linear extrapolation to X′m always overestimates Vmax and (cm−1)
hence dissociation energy. The curve is extrapolated to meet the
X-axis at Xm. This point corresponds to maximum value of vibra-
tional quantum number, i.e. Vmax, since at this point ΔG = 0. To X
determine De, substitute the value of Vmax into the energy expression 0 V xm x¢m
(3.33) for the vibrational level. Fig. 3.11 Plot of ΔG versus V.
Problem 3.9: (a) The following data are obtained from the electronic spectrum of ClO radical:
Band V′, V″ n , cm−1 Band V′, V″ n , cm−1
5 33406.6 17 37264.8
6 33844.4 18 37425.1
9 35053.6 19 37561.9
10 35415.8 20 37678.7
11 35754.5 21 37772.9
12 36066.5 22 37849.5
13 36355.8 23 37917.8
14 36619.5 24 37960.0
15 36864.3 25 37997.4
16 37072.8 — —
(i) Compute the values of the constants we, xe for the upper level and determine the dissociation energy from these
constants, and (ii) calculate the value of dissociation energy from Birge–Sponer plot and compare it with the one
determined in Part (i).
(b) The average value of Δ2G as deduced from the electronic spectrum of BrO radical is −16.40 cm−1. ΔG for
the first band is 477 cm−1. The value of Vmax found from the Birge–Sponer plot is 24. Calculate the energy of the
vibrational state for Vmax = 24, w0, we, D0 and De.
Solution (a) In order to determine the constants we, xe and the dissociation energy, the values of ΔG and Δ2G
mentioned in Eqs (3.53) and (3.55) respectively have been calculated and recorded in the tabular form as follows.
Band ν , cm−1 ΔG, cm−1 −Δ2G, cm−1
5 33406.6
6 33844.4 437.8
9 35053.6 not allowed
10 35415.8 362.2 23.5
11 35754.5 338.7 26.7
12 36066.5 312.0 22.7
13 36355.8 289.3 25.6
14 36619.5 263.7 18.9
15 36864.3 244.8 36.3
16 37072.8 208.5 16.5
17 37264.8 192.0 31.7
18 37425.1 160.3 23.5
19 37561.9 136.8 20.0
20 37678.7 116.8 22.6
21 37772.9 94.2 17.6
22 37849.5 76.6 8.3
23 37917.8 68.3 26.1
24 37960.0 42.2 4.8
25 37997.4 37.4
The average value of −Δ2G has been found to be 21.65 cm−1. (i) Proceeding as in Problem 3.5, we get
21.65
w e xe = w 0x 0 = 83 cm −1
= 10.83
2
With the aid of Eq. (3.53), for V = 5, we obtain

ΔG(V+1/2) = w0 − w0x0 − 2w0x0 × 5 or 437.8 = w0 − 10.83 − 5 × 21.66
Hence, w0 = 546.88 cm−1
From Eq. (3.36) we get
546.88 = we − 10.83; we = 567.71 cm−1
Since wexe = w0x0, therefore
xe = 19.07 × 10−3
w e 567.71 × 103
and De = = = 7442.45 cm−1 = 88.98 kJ mole−1
4x e 4 × 19 07
(ii) In order to determine dissociation energy from the Birge–Sponer method, the maximum value of vibra-
tional quantum number, i.e. Vmax, is determined from the plot of ΔG against V shown in the diagram.
Here Vmax = 28
Using Eq. (3.33) we get
2
⎛ 57 ⎞ ⎛ 57 ⎞
G( ) = 567.71⎜ ⎟ − 10.83 ⎜ ⎟
⎝ 2⎠ ⎝ 2⎠
= 7383.06 cm−1
De = 88.27 kJ mole−1
The error with respect to G(28)
88.27 − 88.98
= × 100 = 0.79
79 per cent
88.27
(b) Proceeding as in Problem 3.5
16.4
w e xe = w 0x 0 = 20 cm −1
= 8.20
2 500
450
From Eq. (3.53), for V = 1, we get 400
ΔG(V+l/2) = G(1+1) −G(l) 350
DG (cm−1)
300
477 = we− wexe − 2 wexe 250
= w0 − 8.20 −16.4 200

150
w0 = 501.6 cm−1 100
So from Eq. (3.36) 50 V max = 28
0
501 = we − 8.2, we = 509.8 cm−1 2 6 10 14 18 22 26 30
V
The energy corresponding to the maximum value of vibrational quantum Fig. 3.12 Plot of V against ΔG.
number, i.e. 24, is determined from Eq. (3.33).
2
⎛ 49 ⎞ ⎛ 49 ⎞
G(24) = 509.8 ⎜ ⎟ − 8.22 ⎜ ⎟ = 7568 cm −1
⎝ 2⎠ ⎝ 2⎠
2
w 02 (501.6)
D0 = = 80 cm −1 = 91.70 kJ mole −1
= 7670.80
4w 0 x 0 4 8.2
2
w e2 (509.8)
D0 = = 65 cm −11 = 94.73 kJ mole
= 7923.65 1
4we x e 4 8.2
3.11 THE VIBRATIONAL PARTITION FUNCTIONS OF MOLECULES

The vibrational component of partition function for diatomic molecules is expressed as
1 1
Z vib = = (3.61)
1− e −hcw
hcw / T
1 e −θ / T
where we is the frequency of vibration (in cm−1 ) and q is the characteristic temperature,
⎛ hc ⎞
θ = ⎜ ⎟ w e = 1 4387 we (3.62)
⎝κ⎠
The same function for polyatomic molecules is
3 n−6 ( 5 )
1
Z vib = ∏i =1 1 − e −θi / T
(3.63)
Problem 3.10: Calculate the vibrational partition function for NO molecule at 500 K if the frequency of vibration
is 1904 cm−1,
Solution Using Eq. (3.62),
q = 1.4387 × 1904 = 2738

θ 2738
= = 5.476
T 500
1 1
From Eq. (3.61) Z = −5.476
= = 1.0041
1− e 1 − 0 .0041
Similarly, we can determine Zvib for other diatomics such as CO (2170 cm−1), OH(3738 cm−1), FO(1060 cm−1),
Cl2 (560 cm−1) etc.
3.12 POPULATION DISTRIBUTION AMONG THE VARIOUS

VIBRATIONAL STATES
The relative population of molecules at a particular temperature on any vibration state is governed by Boltzmann
law of the form
NV g
= V e − EV / T
(3.64)
N A Z v ib
where gV, the degree of degeneracy of any vibrational state, is unity. Y
The relative population varies as e − EV / T and is shown in Fig. 3.13.

NV
Problem 3.11: Using Boltzmann law for the population distribu- NA
tion among the vibrational states, derive the expression for the maxi-
mum value of vibration quantum number.
X
Solution Lumping together Eqs (3.64) and (3.33), 0 V
Fig. 3.13 Plot of NV /NA versus V.
NV g x [ − hc (V +1// )+ h x (V / 2) ]/ T
= V exp e e e (3.65)
N A Z vib
Differentiating Eq. (3.65) with respect to V at constant temperature and applying the condition maxima, i.e.
⎛ dN V ⎞
⎜⎝ ⎟ =0
dV ⎠ T
we get −1 + xe (2Vmax+ 1) = 0
1 − xe
V max = (3.66)
2x e
Problem 3.12: Determine the value of Vmax for OH, if we = 3738 cm−1, xe = 29.7 × 10−3.
Solution Using Eq. (3.66)

1 − 29.7 × 10 −3
V max = = 16
2 × 29.7 × 10 −3
Problem 3.13: Find the fraction of FO molecules at 1000 K on the vibrational quantum level V = 1 and on the
zeroth rotational quantum level if we = 1060 cm−1 and xe = 14.6 × 10−3.
Solution To determine the fraction of FO on the vibrational level V = 1, the vibrational partition function needs
to be defined first. Using Eqs (3.61) and (3.62)
q = 1.4387 × 1060 = 1525
θ 1525
= = 1.525
T 1000
1 1
Z vib = −1.525
= = 1.278
1− e 1 − 0.2176
To find the FO molecules on the vibrational level V = 1, use Eq. (3.64). To do this, calculate the value of
exponent
EV 1 5hcw e . 5hcw e x e
=
κT 1.38054 × 10 −23 × 1000
6.62 10 −34 3 × 108 1060 10 2 (1.5 − 2.25 14.6 10 −3 )

=
1.38054 × 10 −20
= 2.2372
N 1 e −2.2372 0.1067
= = = 0.2925
NA Z vib 0.3646
3.13 ROTATION–VIBRATION SPECTRA OF DIATOMIC MOLECULES

The rotational spectra can be observed in pure form while the purely vibrational spectra are possible only in
extremely rare and special cases (except liquids and solids). There is a simultaneous change in the vibrational and
rotational states of the molecule. The spectrum produced due to change in both forms of energy⎯vibrational and
rotational⎯of the molecule is known as rotation-vibration spectrum and generally appears in the region 1−23 m.

The transition dipole moment integral for a vibrating rotator is defined as
Mv v ∫ Ψ * (V ″)ΨM̂ Ψ
Vr Vr (V ′) d τ
where ΨV r is given by
r − re
ΨV r Ψvib Ψrot (θ , ϕ )
r
and M̂ is the transition dipole moment operator and is related to the permanent dipole moment M. The expanded
form of the dipole moment integral is
MV J ″ M J V ′JJ M J ∫ V″ r V ′ dr
θ φ
∫Ψ rot (J ″ M J″ ) M (θ , φ ) (J ′ M J′ )d τ
where dτ = sin q dq df and MJ’s are the components of J the direction of an external field.
In polar coordinates, we have
M(r, q, f) = M(r) M(q, f).
Neglecting the anharmonicity of M or the vibration, we obtain the following selection rules.
ΔV = ±1, ΔJ = ±1, ΔMJ = 0, ±1
3.13.2 Transition Between two Adjacent Rotation−Vibration States

Ignoring the interaction of vibrations with rotations, the total energy E of a molecule can be written as
E = Eelec + Evib + Erot (3.67)
where Eelec, Erot and Erot have their usual meanings. Since the total electronic energy of the molecule does not
change in the rotation–vibration spectrum, we can write Eq. (3.67) as
EV,J = EV + EJ (or T = GV + FJ cm−1) (3.68)
or TV,J = (GV + FJ), cm−1, and is called the vibrating rotator term value. It can be obtained from the solution of
Schrödinger wave equation. (For solution see Appendix 3.3A).
EV EV ′ EJ EJ ′
v = + = v rot + v vib (3.69)
hc hc
The single and double primes here stand for lower and upper vibrational and rotational states respectively.
If the anharmonicity of diatomic vibration is taken into account and the rotational constant is dependent on the
vibrational state of the molecule, the energy of the rotation-vibration levels may be expressed as the minimum in
the potential curve.
2
⎛ 1⎞ ⎛ 1⎞
BV B e − α e ⎜V γe V + ⎟
⎝ 2⎠ ⎝ 2⎠
⎛ 1⎞
De − β ⎜ V + ⎟
DV (3.70)
⎝ 2⎠
Here, Be and De are the rotational constant and dissociation energy respectively for the hypothetical vibration-
less molecule and correspond to the position represented by
2
⎛ 1⎞ ⎛ 1⎞
EV,J hcw e V h e x e V + ⎟ + BV hcJ (J
hcw (J )
⎝ 2⎠ ⎝ 2⎠
2
⎛ 1⎞ ⎛ 1⎞
G V,J we V w e x e V + ⎟ + BV J (J
(J ) (3.71)
⎝ 2⎠ ⎝ 2⎠
Taking into account the anharmonicity of the oscillations, and assuming rotational constant to be independent
of vibrational state, it follows from Eq. (3.71) that for a simultaneous vibrational and rotational transitions from
V″ to V′ and J″ to J′.
ΔE = EV ″,J ″ − EV ′, J ′ (3.72)
= hcwe(V″−V′) + BehcJ″ [(J″ + 1) − J′ (J′ + 1)]

For a fundamental band V″ − V′ = 1, J″ − J′ = 1 or J″ − J′ = −1
Therefore, for J″ − J′ = 1, we get from Eq. (3.72)
ΔE = hc (we + 2BeJ″) (3.73)
where J″ = 1, 2, 3,... but not zero.
The corresponding frequency is, therefore
v = we + 2BeJ″ [R-branch] (3.74)
Similarly for J″ − J′ = −1, we have

ΔE = hcwe − 2Behc (J′ + 1) (3.75)
and v = we − 2Be (J′ + 1) [P-branch] (3.76)
where J′ = 0, 1, 2, 3,...
The second term in Eqs (3.74) and (3.76) yields the rotational fine structure of the vibration–rotation band.
For definite values of numbers V″ and V ′, i.e. we = constant, therefore, only the second term is responsible for the
fine rotational structure of the band. Hence if J ″ = 1, 2, 3,... we get a group of lines called positive or R-branch of
the band and is characterised by the values v > we. Similarly when J′ = 0, 1, 2, 3, a group of lines called negative
or P–branch is observed and is characterised by the values v < we. We thus obtain the following lines that form
R–and P–branches of the given band.
The lines of R–branch begin with J ′ = 0 while for P–branch J ′ = 1 because J ′ = 0 corresponds to J ″ = J ′ − 1 = −1,
which has no physical meaning. Thus, the following lines form the P– and R–branches of the given vibration–
rotation band.
J′ 0 1 2 3
P ⎯ we − 2Be we − 4Be we − 6Be
R we +2Be we + 4Be we + 6Be we + 8Be
The band, therefore, consists of equally spaced lines with a zero gap we, except when the molecule has an
angular momentum about the internuclear axis, called the zero line of the band. The distance between adjacent
lines is of the order of 2B. The graphical representation of the expected vibration–rotation spectrum is shown in
Fig. 3.14. In reality due to interaction between rotation and vibration, the straight lines become curved.
If anharmonicity and nonrigidity of a diatomic oscillator and rotator respectively are taken into account, then
for the P-branch we shall have
v = we − 2wexe − 2BeJ − 1.5ae J(J−1) (3.77)
and for the R-branch
v = we −2wexe + 2Be(J + 1) −1.5ae(J +l)(J +2) (3.78)
The spectrum of HCl under moderate resolution is shown in Fig. 3.15. The spectrum consists of two peaks of
different intensities and shapes of the peaks are temperature dependent. The high resolution spectrum suggests that
J
4
3
V =1 2
1
0
Absorption
3
V =0
2
1
0
3333 2886 2500

P R ν (cm−1)
-
n Fig. 3.15 The fundamental absorption band of HCI
Fig. 3.14 Vibrational–rotational energy level diagram for a diatomic molecule. in the near infrared under moderate dispersion.
these peaks are simply the envelop of a series of

approximately equally spaced lines apart from
the gap in middle as shown in Fig. 3.16.
Let us now consider transition between the
ground and first excited vibrational level of the
molecule. Then, from Eq. (2.33) we have
E0 = B0J(J + 1) (3.79) 2728 2865 2906 3014
ν (cm−1)
and
Fig. 3.16 High-resolution spectrum of the fundamental vibrational band of
HCI showing rotational fine structure and isotopic splitting due to H35 CI and
v v b + B1J (J + 1)
H37Cl.
E1 (3.80)
where B0 and B1 have their usual meanings.

Since, for P–branch J′ = J″ + 1 and for R–branch J″ = J′ + 1, the energy terms of Eqs (3.79) and (3.80) can be
written as
For P–branch, E0 = J(J + l)B0 (3.81)
E1 v vib + (J l ) JJB
B1 (3.82)
For R–branch, E0 = J(J+l)B0 (3.83)
E1 v vib + (J 1))((J + 2) B1 (3.84)
Thus, the frequencies of the lines in P- and R-branches of the spectrum can be obtained from the expressions
P (J )…v vib + (J − 1) JB1 − J (J + 1) B 0 f J ≥1 (3.85)
and R (J )…v vib + (J + 1)(J 2) B1 − J (J + 1) B 0 f J ≥0 (3.86)
Thus,
⎛ 1⎞
R (J ) − P (J ) = 4 J B1 (3.87)
⎝ 2⎠
and similarly we can deduce

⎛ 1⎞
R (J ) − P(J
(J ) = 4⎜J B0 (3.88)
⎝ 2⎠
and P(J) + R(J −1) = 2 νvib − 2(B0−Bl) J2 (3.89)
B1 and B0 can be obtained from Eqs (3.87) and (3.88) respectively and hence the corresponding moments of
inertia, i.e. I and I0.
The value of νvib is determined graphically by means of Eq. (3.89). The plot of P(J) + R(J − 1) versus J2 will
yield a straight line with intercept = 2 νvib and slope = −2 (B0 − B1). The value of νvib can thus be evaluated from
the intercept.
The internuclear distance (re) is determined from the relation
I μ m re2
and force constant from
we k / μm
Problem 3.14: Calculate the number of absorption maxima in the purely rotational spectrum of CO.
we = 2170 cm−1, xe = 6.11 × 10−3, Be = 1.931 cm−1.

Solution Using Eq. (3.49), the wave number of fundamental band of CO is
ν = we (1−2xe) = 2170(1 − 0.01222)

= 2143 cm−1
The rotational quantum number of maxima in the pure rotational spectrum with wave number ν = 2143 cm−1, is
ν = 2Be (J+ 1), 2143 = 2 × 1.931(J + 1), J = 554
Problem 3.15: The wave numbers of the lines in the P-branch of the rotation-vibration spectrum of methane
are 3032.30, 3043.15, 3054.00, 3064.85 and 3075.70 cm−1. Determine the moment of inertia and bond distance
of methane.
Solution The average value of spacing between the successive lines in the P-branch of the rotation-vibration
spectra of methane is 10.85 cm−1, i.e.
2B = 10.85 cm−1
or B = 5.42 cm−1
Using Eq. (2.34)
16.8575
I= = 3.11 au
5 42
From Eqs (2.17) and (2.18)
r1 = 3 11 = 1.763 Å
3.11
r2 = 3 / 8 r1 1.079 Å
Thus, C—H bond length of CH4 is 1.079 Å. The literature value is of the order of 1.09Å.
Problem 3.16: Determine the wave numbers of the first three lines in the P- and R-branches of the rotation–
vibration absorption spectrum of ClF if we = 786.15 cm−1; xe = 7.84 × 10−3, Be = 0.516 cm−1.
Solution The wave numbers of P- and R-branches are calculated from Eqs (3.77) and (3.78), respectively.
v w e − 2w e x e − 2B e J − 1.5 e (J 1)J
v w e − 2w e x e + 2B e (J + 1) − 1.5 e (J 1)(J + 2)
we −2wexe = 786.15 − 2 × 786.15 × 7.84 × 10−3 = 773.83 cm−1
Ignoring the contribution of ae, the wave numbers for the first five levels are recorded in the tabular form as
follows.
J P-branch R-branch
2Be(J + I),
2BeJ,cm−1 ν , cm−1 ν , cm−1
cm−1
0 — — 1.032 774.862
1 1.032 772.798 2.064 775.894
2 2.064 771.766 3.096 776.926
3 3.096 770.734 4.128 777.958
4 4.128 769.702 5.160 778.990
5 5.160 768.670 6.192 780.022

Problem 3.17: The frequencies (in cm−1) of HI lines arranged in order of their magnitudes, starting with the
highest value in the region 2200−2300 cm−1 are 2117.7, 2132.7, 2147.5, 2161.7, 2176.7, 2176.0, 2190.0, 2203.6,
2216.7, 2242.2, 2254.3, 2266.1, 2277.5 and 2288.5. Calculate the moment of inertia, intenuclear distance and
bond strength of HI.
Solution The division of the branches on the basis of gap between the lines at 2216.7 and 2242.2 cm−1 is as
follows.
J R(J), cm−1 P(J), cm−1
0 2242.2 —
1 2254.3 2216.7
2 2266.1 2203.6
3 2277.5 2190.0
4 2288.5 2176.0
5 — 2161.7
6 — 2147.5
7 — 2132.7
8 — 2117.7
The values of a0, b0 and c0 are calculated from Eqs (3.87), (3.88) and (3.89) in order to determine B1, B0 and ν vib,
R (J ) − P (J ) R (J − 1) P (J 1)
where a0 = ; b0 = ; and c 0 = P (J ) + R (J − 1)
⎛ 1⎞ ⎛ 1⎞
⎜⎝ J + ⎟⎠ ⎜⎝ J + ⎟⎠
2 2
The calculated values of B1 and B0, a0, b0 and c0 (in cm−1) are
J a0 B1 b0 B0 c0
1 25.06 6.27 25,73 6.43 4458.9
2 25.00 6.25 25.72 6.43 4457.9
3 25.00 6.25 25.74 6.44 4456.1
4 25.00 6.25 25.73 6.43 4453.6
5 — — 25.64 6.41 4450.2
The difference in the values of B0 and B1 is due to centrifugal stretching. Taking B0 = 6.43 and Bl = 6.26 cm−1,
we get
I0 = 2
h
= 2
6.624 × 10 −27 2
( )
= 4.36 × l0−40 g cm2 = 2.625 au
8π cB 0 8π 2.998 × 1010 1
× 6 43 cm (
−1
) ( )
4460.0
and 4458.0
( )
P(J) + R(J − 1)
h 6.624 × 10 −27 2
I1 = = = 4.48 ×10−40 g cm2 4456.0

8π cB1 8π 2
2
2.998 × 10 10
( (
/ s ) × 6.26 cm −1 ) 4454.0
= 2.698 au
4452.0
Then plot between P(J ) + R(J − 1) = c0 and J 2 is shown in Fig. 3.17. 4450.0
The intercept at J = 0 is 4459.3 cm−1. From the intercept we get 0 5 10 15 20 25
J2
1
v vib = × 4459.3 = 2229.7 cm −1 Fig. 3.17 The fundamental vibrational
2 frequency of HI.
The bond lengths in both the states are determined from the expression I = mm r2 as follow.
We know mH = 1.663 × lx10−24g = 1 amu
and mI = 126.92 × 1.663 × 10−24g = 126.92 amu

Hence, mHI = 1.6486 × 10−24g = 0.99218 amu
Thus, from
I0 μ m r02
2.625 = 0.99218 r02
r0 = 1.63 Å
and
I1 μ m r12
2.698 = 0.99218 r12
r1 = 1.65Å
Force constant is evaluated from Eq. (3.19).
Since νvib = (1 − 2xe)we, where
xe = 0.0172 and νvib = 2229.7 cm−1
2229.7
Therefore, we = = 2309 cm−1
(1 2 × 0.0172)
k
and 2309.1 = 1303.16
0.99218
2
⎛ 2309.1 ⎞
Therefore, k= ⎜ × (0.99218) = 3.12 millidynes/Å
⎝ 1303.16 ⎟⎠
3.13.3 Intensity Distribution in Rotation–Vibration Absorption Spectrum

The maxima in the rotation–vibration absorption spectra of diatomics are because of intensity distribution of
rotational lines. The maxima can be determined from population distribution among the rotational levels and the
probabilities of appropriate transitions between the levels.
Recall from Chapter 2 that the relative population among the rotational levels is governed by the Boltzmann
law of the form
N J ( 2 J + 1) e −BBe hcJ (J 1) / T
= (3.90)
NA σ Z rot
The intensity of the absorption line is measured in terms of its absorption coefficient,
Av = Av0 NJ (3.91)
0
where A is a constant proportional to the probability of the appropriate transition. Combination of Eqs (3.90)
v
and (3.91) yields
2 J 1 e −BBe hcJ (J 1) / T
A ν A v0 N J A v0 NA (3.92)
σ Z rot
The relative probabilities in which we are mainly interested can easily be worked out in case of the terms 1 ∑ .
In this case two transitions, J → J + 1 with probability P+ and J → J − 1 with probability P− except for J = 0 are
possible.
The total probability due to these two transitions may be expressed as

P0 = P+ + P− (3.93)
Theoretically, P0 remains constant within the limits of a given band.
The relative probabilities of transitions are expressed in terms of intensity factor; and is determined from the
equality;
iP0 = (2 J + l)P (3.94)
It follows from Eq. (3.94) that
∑i i+ + i− 2J + 1 (3.95)
Combining Eqs (3.92) and (3.95), we get

1 e −BBe hcJ (J 1) / T
Av ~ iN A (3.96)
σ Z rot
J′− J″ and J″ − J′ belong to the same value of intensity factors because J″ and J′ enter symmetrically into the
expression of the appropriate quasi-moments. Due to this reason, we must assign intensity factors to the transi-
tions. The assigned values to different transitions are as follows:
Transition Intensity Factor
0 →1′ , 1′ → 0 i(0) = i+ (0) = 1 (i− in this case is zero)

1 → 0′, 0′ → 1 i(1) = i− (1) = 1
[1→ 2] is also possible in this case from J = 1 [i +(1) = 3 − 1 = 2]
2 → 1′ i− (2) = 2
2 → 3′, 3 → 2′ i+ (2) = i− (3) = 5 − 2 = 3
From the above data, the general expressions deduced for the intensity factors of the lines belonging to the P-
and R-branches of the band are
P: i− (J) = J (3.97)
R: i+ (J) = J + 1
Taking advantage of Eqs (3.97) we can, from expression (3.96), find the value of J and consequently, the fre-
quency (or wavelength) to which at the given temperature, there corresponds a maximum of intensity in each
branch of the band.
The following conditions of maxima have been deduced from Eq, (3.96)
⎛ di ⎞ B hc
⎜⎝ ⎟⎠ = i(
i J+ ) e (3.98)
dJ T κT
which for the R- and P-branches will be written in the form
κT
R-branch: 2J +2 3J + + 1 = (3.99)
B e hc
κT
and P-branch: 2J −2 J − = (3.100)
B e hc
Assuming that at room and higher temperature Behc/κTl << 2, and solving Eqs (3.99) and (3.100) for J+ and
J− respectively, we arrive at the following expressions:
2κT
J+ J −1 + 1 = (3.101)
B e hc
For different molecules at room temperature, the value of Behc/κT lies in the range of 0.1 − 0.001.
Solution of Eqs (3.99) and (3.100) gives the values of J+ and J− that correspond to the maximum intensity of the
line in the R- and P-branches of the band, respectively. From these values we can find the frequencies v+ and v− that
correspond to the maximum intensity.
B e (J + + 1)
v + = v vibib + 2B (3.102)
v − = v vib − 2B e J − (3.103)
Thus, Δvmax = v+ − v− = 2Be (J+ + J− + l) (3.104)

where Δvmax is the distance between maxima.
Combining Eqs (3.301) and (3.104)
2 T 8B e T
Δv max = 2B e = (3.105)
B e hc hc
h
Since Be = (From Chapter 2)
8π 2I e c
1 κT
Therefore, Δv max = (3.106)
πc I e
Using Eq. (3.106), we can determine the moment of inertia of a molecule (Ie) from the distance between the
maxima of the bands.
Bjerrum also proposed an identical formula to determine moment of inertia from Δvmax.
κT
Ie =
(π c Δv )2
Due to this reason, the bands with a double maximum in rotation–vibration spectra of diatomics are called
Bjerrum doublets. Since Δvmax cannot be measured very accurately, the accuracy of determination of moment of
inertia by this method is very low. However, we can have a tentative value of moment of inertia by this method
without analysing the fine structure of the bands.
Problem 3.18: The observed distance between the maxima in the P- and R-branches of the rotation-vibration
spectrum for the l = 3.5m. (2857 cm−1) band of HCl is 125.19 cm−1 (0.15 m). Calculate the moment of inertia of
HC1 at 273 K and compare it with the literature value.
Solution Using Eq. (3.106)
Ie =
8.3142 ×10
10 23( A ) 273K
631 = 2.71 10 −46 kg 2
= 1.631au
125.19 ( )×3 (cm / s)⎤⎦
2
⎡3.14 ×125 10 10
⎣
This value of moment of inertia is in agreement with the literature value, i.e. 2.71×10−46 kg2.
Problem 3.19: (a) Predict the wave numbers of the lines with maximum intensities at 500 K in the P- and
R-branches of the rotation-vibration spectrum of HI. we = 2309.01 cm−1, xe = 17.2 × 10−3 and Be = 65.11 cm−1.
(b) Using the data from Part (a), determine the moment of inertia of HI and compare it with the literature value.
s’
Solution (a) In order to determine the wave numbers of the lines with maximum intensities, J max for the P- and
R-branches need to be defined.
To define J+ and J− from Eqs (3.99) and (3.100) respectively, let us first calculate the value
κT 1.3805 × 10 −23 × 500
= = 5.33
B e hc 65.11 × 6.6252 × 10 −34 × 3 × 1010
Substituting the value of κT /B

B e hc in Eqs (3.99) and (3.100), we get
2 J +2 3J + − 4 33 0
2 J −2 J − 5.33 = 0
Solution of these quadratics results in

J+ = 1
J− = 1
Consequently, from Eqs (3.102) and (3.103)
v + = 2309.01 + 2 × 65.11(1 + 1) = 2569.44 cm −1
v − = 2309.01 − 2 × 65.11 × 1 = 2188.79 cm −1
∴ Δv max = ( 2569.44 − 2188.79) = 380.65 cm −1
(b) The tentative value of

8.314 × 10 23 × 500
Ie = 2
= 3.230 au
⎡⎣3.14 3 × 1010 × 380.65⎤⎦
The value of Ie from Be = 2.59 au.

Molecular constants of some diatomic molecules are given in Appendix 3.4A and the reciprocal table for the
conversion of microns (m) to wave numbers (cm−1) is given in Appendix 3.5A.
3.14 NATURE AND NUMBER OF VIBRATIONAL MOTIONS

IN POLYATOMIC MOLECULES
Atoms in a molecule are bound together through chemical bonds and oscillate about their mean positions which
results in the change of molecular shape. Depending upon these type of motions of atoms in molecules relative to
each other, different-natured vibrations result, e.g. in a diatomic molecule (say HC1), only one type of vibrational
motion exists while in linear and bent triatomic molecules (say CO2 and H2O) additional vibrations are possible
as shown in Fig. 3.18. Thus, it is evident from Fig. 3.18 that in polyatomic molecules there are several ways in
which the molecules may vibrate. However, the number of vibrations, in which each atom vibrates about its equi-
librium position with the same frequency but with different amplitude and motions of all atoms are in phase called
O O O
H H H H H H
Symmetric Stretching, ν1 Symmetric Bending, ν2 Asymmetric Stretching, ν3

(3651.7 cm , ) −1
(1595.0 cm−1, ) (3755.8 cm−1, ⊥)
O C O O C O
Symmetric Stretching, ν1 Asymmetric Stretching, ν3

(1337 cm−1, ,R) (2349 cm−1, , IR )
− + −
O C O O C O
Bending, ν2 (667 cm−1, ⊥, IR)
Fig. 3.18 Normal modes of vibration of (a) water molecule, and (b) the carbon dioxide molecule; (+) and (-) signify displacements in and out
of the plane of the paper.
fundamental or normal modes of vibration, are fixed. When two atoms move in opposite directions, a stretching
or valence vibration results while their movement at an angle results in bending vibration. The various types of
stretching and bending vibrational modes are listed in Table 3.6.
Any molecule thus may have different set of frequencies each of which may cause IR absorption. If a nonlinear
molecule has n atoms, it has 3n total degrees of freedom of which 3 are translations and 3 are rotations and the
remaining 3n − 6 are vibrational degrees of freedom. However, in case of linear molecules the number of vibra-
tional modes will be 3n − 5 since the rotation about molecular axis does not displace any nuclei which results in
only 2 rotational degrees of freedom. Thus n − 1 and 2n − 5 will be the number of stretching and bending vibra-
tions respectively in nonlinear molecules and n − 1 and 2n − 4 in linear molecules. The three translations along
the three coordinate axes or rotations about two or three suitable axes are also termed as null or non−genuine
vibrations because there is no restoring force for these motions, the vibrational frequency is zero.
Table 3.6 Various types of stretching and bending vibrational modes in polyatomic molecules.
Fundamental Stretching Vibration vst A X Y
X
Symmetric Stretching Vibration vsym B A
Y
X
Asymmetric Stretching Vibration vasym B A
Y
+ −
In-Plane Bending Vibration δb A X Y
− +
Out-of-Plane Bending Vibration γb A X Y
X
Symmetric Bending Vibration bs B A
Y
X
Asymmetric Bending Vibration ba B A
Y
X
In-Plane Rocking Vibration rr B A
Y
X +−
Out-of-Plane Wagging Vibration χw B A
−+ Y +−
X +−
Twisting Vibration τt B A
Y −+
Ring Deformation in Plane d
+ +
Ring Deformation Out-of-Plane d
− −
+
It should be noted that in symmetric molecules such as CO2, C2H2, CH4, CCl4, etc., certain frequencies
coincide due to which the total number of different frequencies observed is less than the expected from the for-
mula 3n − 5 or 3n − 6. The frequencies which coincide with one another vibrate in mutually perpendicular planes
and the vibration corresponding to the observed frequency is said to be doubly degenerate, e.g. in case of CO2, the
coincidence frequencies are the two identical frequencies that correspond to the vibrating ν2 of the molecule in
two mutually perpendicular planes. Triply degenerate vibrations can also occur in some molecules.
Further, in polyatomic except diatomic molecules, the fundamental vibrations are slightly anharmonic due to
their delocalisation effect but their components are fundamental and consisting of simple integral multiples of
the fundamental vibrational frequency called overtones are harmonic. Thus, overtones corresponding to ΔV = ±2,
±3, . . . are possible for each normal mode. The two or more vibrational normal modes may also combine to give
combination (sum or difference) tones, e.g. the normal frequencies v1 and v2 may give v1 + v2, v1+ 2v2 and v1 − v2,
etc., combination bands. Therefore, a molecule which vibrates anharmonically causes a dipole moment to oscil-
late with harmonic frequencies and their overtones and combination tones. It is, therefore, possible for molecules
to absorb radiation in the over- and combination-tones region also. This results in the appearance of additional
bands in the IR spectra. The intensity of these bands is weak in comparison to fundamental bands and depends
upon the amount of anharmonicity in the molecule, e.g. the intensity ratio of carbonyl normal mode to its first
overtone is of the order of 100/1.
The vibrational transitions in polyatomic molecules are generally abbreviated as follows: For fundamental
bands all Vi = 0 except one which is equal to unity, e.g. V (0, 0, 0, ...) → V(1, 0, 0, ...). In case of overtone transi-
tions all Vi = 0 except one which is greater than unity, e.g. V(0, 0, 0, …) → V(0, 2, 0, ...), while for combination
bands, various combinations of Vi exists, e.g. V(0, 0, 0, ...) → V(0, 1, 1, ...), etc.
3.14.1 Fermi Resonance

Sometimes it so happens that the energies of the different normal vibrational modes or combinations of vibrations
of polyatomic molecule coincides closely with one another. This characteristic feature of the vibrational modes of
molecules is known as accidental degeneracy. Depending upon the type of symmetry, they may interact, if they
have the same type of symmetry, resulting in the mutual perturbation of each other and hence displacement of
frequencies in opposite directions. This type of interaction is due to the fact that two close molecular vibrational
frequencies resonate and exchange energy and this phenomenon is known as Fermi resonance. This is similar
to vibrational perturbations of diatomic molecules. The only essential difference is that for diatomic molecules
only vibrational levels of different electronic states can have nearly the same energy and thus perturb one another
whereas in Fermi resonance, two vibrational levels of the same electronic level can have the same energy and per-
turb each other, e.g. in case of CO2 the levels V1 = 0, V2 = 2, V3 = 0 has almost the same energy as the levels V1 = 1,
V2 = 0, V3 = 0, since v1 = 1337 and 2v2 = 667 × 2 = 1334 cm−1. In diatomics, the perturbation is because of the inter-
action of vibration and electronic motions whereas in polyatomics, the perturbation is due to interaction between
different vibrations when two levels happen to lie very close together. Further, in the case of complex molecules
having so many fundamentals and combination tones, chances of accidental degeneracy becomes bright. How-
ever, it should be borne in mind that not all such degeneracies lead to resonance. In these types of interactions,
the overtone or combination band gains intensity at the expense of the fundamental and the higher frequency
shifts in the higher direction while the lower one gets depressed. Fermi resonance is also detected in COS, CS2,
COSe, ICN, C2H2 but not in H2O, even though two of its vibrations have close frequencies at v1 = 3651.7 cm−1 and
v2 = 3755.8 cm−1, since these vibrations belong to different types of symmetry.
Though we have already described the effect of Fermi resonance on the intensities and energies of the two
interacting transitions, it is nice to elaborate it quantum mechanically. The intensity of overtone or combination
transitions is generally less than one tenth, the intensity of fundamental transitions. If vi, vj and vk are the three
fundamental vibrational frequencies of a molecule such that (vi + vj) ≈ vk then the magnitude of interactions is
given by
ij ,k = ∫ ψ iij Ê ykdt
The interaction operator Ê is nonzero because the vibrations are anharmonic. It is also totally symmetric and for
this reason the two states yij and yk must belong to the same symmetry otherwise the integral will not have finite value.
This interaction displaces the two expected frequencies away from each other. Moreover, the interaction between yij
and yk which affects the actual energies, also gives rise to new wave functions which are linear combinations of these
two. Thus, actual excited state whose energy is very close to (vi + v j) will be described by a wave function
Ψ′ij = C(
Ψ C Ψijj x Ψk )
where C is a normalisation factor and x ≤ 1. the actual excited state whose
energy is very close to vk is expressed by the wave function.
(a)
Ψk′ = C(
C Ψk x Ψijj )
Thus a transition from the ground state y0 to yij will have low intensity and
is proportional to the small integral
(b)
∫y0 (x, y, z ) yi j dτ
and transition to the actual excited state, y ′ij, has an intensity proportional to
∫y0 (x, y, z ) yij dτ + x ∫y0 (x, y, z ) yk dτ

νi + νjνk
where the second integral must be large, because y0 → yk transition is intense. The Fig. 3.19 The schematic diagram depic-
weak overtone transitions ‘borrow’intensity from the strong fundamental transition ting Fermi resonance. (a) The spectrum
(infrared or Raman) expected for the
because it is close to it and has an excited state with the same symmetry. The closer
combination, (vi + vj) and vk in the absence
the energies (vi + vj) and vk, the larger the sharing of total intensity equally between of Fermi resonance. (b) The actual spec-
(vi + vj) and vk. The schematic diagram depicting Fermi resonance is shown in trum where two bands diverge and share
Fig. 3.19. intensity because of Fermi interaction.
3.14.2 Coriolis Force

Further, in addition to Fermi resonance, Coriolis force and the phenomenon of inversion also influence the vibra-
tional and rotational spectra of molecules. In the nonrigid rotator model of the diatomic molecule, a constant term,
called the centrifugal distortion constant, is introduced in order to narrow down the discrepancies between the
observable experimental and theoretical data. The constant term appears because of the centrifugal force given by
Fcentrifugal = mrw2 (3.107)
The centrifugal force is exerted on particles which are stationary in the rotating system. The magnitude of
centrifugal force depends upon the distance of the particle from the axis of rotation, Cm. However, if the particle
is moving with certain velocity because of the oscillation of a particle in a rotating system, a new force called
Coriolis force becomes operative and is given by
F coriolis = 2m vx w sin f (3.108)
Here in Eqs (3.107) and (3.108), m is the mass of the nucleus, vx is the velocity of the particle due to the vibra-
tion of the particle, r is the distance from the axis of rotation, w is the angular velocity and f is the angle between
the axis of rotation and the direction of vx.This force is independent of the position of the particle but depends
upon the velocity vx, i.e. not only by its magnitude but also by its direction with respect to the axis of rotation.
When the particle moves along the axis of rotation, coriolis force is zero. The greater the value of f, the greater
will be the Coriolis force, and has maximum value when motion of the particle is at right angles to the direction
of rotation.
The Coriolis force leads to an additional interaction between rotation and vibration which in general is much
larger than the effect of centrifugal force since velocity due to vibration (vx) is much larger than that because of
rotation (rw). Such type of interaction occurs only in the vibrating molecule and is measured in terms of a con-
stant called the Coriolis coupling constant. It is denoted by the symbol ς. It is the magnitude of the vibrational
angular momentum in units of h/2p for the particular vibrational state and is to be taken positive or negative
depending on whether or not the direction of rotation of the electric dipole moment of the molecule coincides
with the direction of the angular momentum of vibration. It is different for different vibrations and its value lies
between +1 and −1. Thus, in the case of diatomic molecules, the variation a of the rotation constant B with the
vibrational quantum number is because of the following three reasons: (i) for harmonic vibration, the mean value
of (l/I )average ≠ 1/Ie although (r)average = re, (ii) for anharomonic oscillation, (r)average > re, and (iii) the Coriolis coupling
constant. In other words, we can say that,
ai = a i (harm.) + ai (anharm) + ai ( Cor.) (3.109)

ai for different types of molecules have been expressed in terms of potential constants and internuclear distances.
The formulae for the linear XY2 molecules are
−6 B e2
a .) =
α1 ( harm
w1
+B e2
a .) =
α 2 ( harm (3.110)
w2
+2B e2
a .) =
α 3 ( harm
w3
24 B e3I e3 / 2α111
a arm .) =
α1 (anha
w 13 hc
−8B e3I e3 / 2α122
a arm .) =
α 2 (anha (3.111)
w 12 w 2 hc
−8B e3I e3 / 2α133
a arm .) =
α 3 (anha
w 12 w 3 hc
a1 (Cor.) = 0
4 B e2w 2
α 2 (Cor.) = (3.112)
( 2
3 − 2
2 )
−8B e2w 3
α3 ( )=
( 2
3 − 2
2 )
Here, α111, α122 and α133 are the anharmonic potential constants and w1 (symmetric stretch), w2 (bend) and
w3(asymmetric stretch) are the fundamental modes of vibration of linear XY2 type of molecules. Knowing the
values of the various constants, w1 , w2 and w3; the values of different types of ai s can be estimated.
Further, since Coriolis coupling constant is proportional to the speed of rotation, a splitting that grows with J is
expected and indeed l-type splitting in ∏, Δ-vibrational levels of linear polyatomic molecules has been observed.
The l-type splitting in a vibrational state ∏ is given by
Δv = qJ(J + 1) (3.113)
where q = ai (harm.) + ai (Cor.) and the subscript i stands for the ⊥ vibration that is excited in the Π state. The sum
of the ζi values for all vibrations of the same species is independent of the potential constants and hence can be
expressed in terms of the moments of inertia. The sum rule is valid provided no resonance occurs and harmonicity
is ignored. For planar or pyramidal molecules:
IA
ζ ζ4 = −1 (3.114)
2I B
For planar molecules, since
IA = 2IB
Therefore, ζ3 = −ζ4
For the two F2 vibrations v3 and v4 of XY4 molecules, ζ3 + ζ4 = 1/2 since here B = A. ζ2 = 0 for the E vibration
of XY4.
The general rule specifying the vibrational states between which a Coriolis interaction can take place is due
to Jahn, i.e. two vibrations will interact in consequence of Coriolis force in the rotating molecule only when the
product of their species contains the species of a rotation. In case of symmetric top molecules, the Coriolis forces
that occur in the rotating molecule may produce an interaction between mutually degenerate vibrations, which in
turn results in splitting of the degeneracy. There are three species of degenerate vibrational levels in tetrahedral
molecules, i.e. E, F1 and F2. The fundamentals of Y4 and XY4 have only species E and F2. No Coriolis splitting
arises for the doubly degenerate vibrational states since E × E = A1 + A2 + E, i.e. does not contain the species of
rotation which is F1 in this case, The rotational energy levels are F+(J) J
the same as for the non-degenerate vibrational states. However, F0(J) 6
for the triply degenerate vibrational states, the Coriolis interaction J F–(J)
5
causes a splitting. The rotational energy values for the three levels 6
which are similar to that for symmetric top molecules (Fig. 3.20) 5 4
are given by the formulae: F2 4 3
F + (J) = BV J (J + 1) +2 BV ζi (J + 1) (ΔJ = −1) (3.115) 3 2
2 1
1 0
0
F 0 (J) = B V J (J + 1) (ΔJ = 0) (3.116)
O R Q P S
F −(J) = BV J (J + 1) − 2 B v ζi J (ΔJ = +1) (3.117) 6
5
The formulae hold good for vibrational states of species F1 as A1
well as of F2,, the former occurring only as upper states of certain 4
overtone and combination bands of the XY4 molecules. The fine 3
structure of rotation-vibration bands of tetrahedral molecules for the 2
1
F2 – A1 transitions is well understood. The spacing in different bands 0
is different. The reason for this is, as for symmetric top molecules, Fig. 3.20 Rotational energy levels of a spherical top
the Coriolis coupling between rotation and vibration in the upper molecule in a triply degenerate (F2) and totally sym-
degenerate vibration state. Consequently, the upper states of corre- metric (A1) vibrational states. The broken lines do not
occur.
sponding lines of the P-, Q- and R-branches are somewhat different.
Using Eqs (3.115), (3.116) and (3.117) and applying the selection
rules, for the F+, F0 and F− substates, we obtain for the R-branch
R (J ) = v 0 2BV″ − 2BV″ ζ i 3BV″ − BV′ 2BV″ ζ i )J ( BV″ − BV′ )J 2 (3.118)
for the Q-branch,

Q (J ) = v 0 ( ″
V V )J + ( ″
V − V )J 2 (3.119)
and for the P-branch,
P (J ) = v 0 (BV″ + BV′ 2BV″ ζ i )J ( BV″ − BV′ )J 2 (3.120)
The P- and R-branches form one series of lines represented by the formulae:
v = v 0 + (BV″ BV′ − 2BV″ ζ i )m + ( BV″ BV′ )m 2 (3.121)
where m = J + l for R, m = −J for P-branch and where one line m = 0 is missing. For fundamentals BV′′ − BV′ = 0,
therefore, the separation of successive lines in this series is nearly constant and is equal to 2B (1 − ςi ), and all lines
of the Q-branch coincide at the zero line v0. Since ςi, is different for different vibrations, the spacing of the lines
in different infrared bands may differ widely. The moment of inertia of tetrahedral molecule, therefore, cannot
be determined from the fine structure of one infrared band alone. However, because of the sum rule for ςi, it is
possible to determine the moment of inertia if the spacing in all infrared-active fundamentals is known. For XY4
molecules there are two such fundamentals, v3 and v4 for which ς3 + ς4 = 1/2. Therefore, the sum of spacings in v3
and v4 = 3B. Thus, B, IB and the bond distance X-Y in XY4 can be determined. The fundamental bands of CH4 are
observed at v1 (A1) = 2914.2, v2 (E) = 1520, v3(F2) = 3020.3 and v4(F2) = 1306.2 cm−1.
Recall from Chapter 2 that inversion in case of nonplanar XY3 type molecules manifests itself in the vibrational
spectrum of such molecules. The doubling of all the vibrational energy levels occurs because of two potential
minima corresponding to inversion. In the infrared only sublevels of opposite parity combine with one another
(+ ↔ −) while in the Raman effect only sublevels of the same parity combine with one another (+ ↔ +, − ↔ −).
When the inversion doubling of the vibrational levels is unobservably small, the selection rules have no observ-
able effect on the spectrum. Only the selection rules for one potential minimum are considered. However when the
inversion doubling is observable as in the case of NH3, the selection rules for sublevels are also taken into account.
All the infrared and Raman transitions are then double as shown in Figs 3.21(a) and (b). The doublet splitting
of the bands in the infrared and Raman spectra is different because of different selection rules in the respective
Table 3.7 The energy states (0, V2 , 0, 0) of NH3, and ND3,.

State (V2 ) NH3(cm−1) ND3 (cm−1)
+
0 0 0
0 −
0.66 ∼0
+
1 932.24 745.6
1− 968.08 749.0
+
2 1597.4 1359
−
2 1910 1429
+
3 2380 1830
−
3 2861 2113
4+ 2495
−
4 2868
spectra. The doublet splitting of the upper and lower levels is the
sum in the infrared [Fig. 3.21(a)], whereas it is the difference
(a) (b)
in the Raman spectrum [Fig. 3.21(b)]. Totally symmetric v1 V
and v2 bands of NH3 are double, i.e. 1IR ( 1
),
− −
v 2 IR ( ); v 2R ( 1
) . In case of ND3, +
2
+
v2IR (748.6, 749.0 cm−1).
The doublet splitting of v2 band in case of ND3 is small in com-
parison to that of NH3. The energy states (0, V2, 0, 0) of NH3 and −
−
ND3 are recorded in Table 3.7. + 1 +
Problem 3.20: Write shorthand notation for the following

bands: (i) 2v2; 3n − 6 = 3 (ii) v2 + v3 + v6; 3n− 6 = 6 and (iii) 2v2 +
− −
v3 + 2v4; 3n−5 = 4. + 0 +
Infrared Raman
Solution The shorthand notation for (i) 2v2 band is (0, 0, 0) Fig. 3.21 Effect of inversion doubling on (a) infrared,
→ (0, 2, 0), (ii) v2 + v3 + v6 band is (0, 0, 0, 0, 0, 0) → (0, 1, 1, 0, 0, and (b) Raman spectra.
1), and (iii) 2v2 + v3 + 2v4 band is (0, 0, 0, 0) → (0, 2, 1, 2). This also
suggests that molecule in linear.
Problem 3.21: State the meaning of the following vibrational transitions in a polyatomic molecule: (i) (0, 0, 0)
→ (0, 0, 2); (ii) (0, 2, 0, 0) → (0, 1, 1, 1); (iii) (0, 0, 0, 0) → (0, 2, 1, 1); and (iv) (0, 1, 1, 1, 2) → (0, 2, 1, 2, 0).
Solution (i) The transition V(0, 0, 0) → V(0, 0, 2) means that the vibration n3 is excited from V3 = 0 → vibra-
tion V3 = 2 and this is the first overtone with frequency 2v3. In total there are three modes of vibrations. (ii) By
transition V(0,2,0, 0) → V(0, 1,1,1), we mean that the vibration n2 is de-excited from V2 = 2 → V2 = 1 while the
vibrations n3 and n4 are excited from V3 = 0 →V3 = 1 and V4 = 0 → V4 = l, respectively. The band corresponding
to this transition is a combination with frequency v2 + v3 +v4. In all, there are four modes of vibrations. (iii) The
transition (0, 0, 0, 0) → (0, 2 1, 1) is a combination bandwith frequency, 2v2 + v3 + v4. Overall there are four vibra-
tional degrees of freedom. (iv) The transition (0, 1, 1, 1, 2) → (0, 2, 1, 2, 0) is a combination bandwith frequency
v2 + v4 + 2v5. In all, there are five vibrational degrees of freedom.
Problem 3.22: The frequencies of three fundamental modes of vibration of CO2 have been observed at 667,
1337 and 2349 cm−1, Calculate the frequencies of the absorption bands for the following arbitrary transitions: (i)
V(0, 0, 0) → V(2, 0, 0); (ii) V(0, 0, 0) → V(0, 1, 2); (iii) V(0, 0, 2) →V(l, 1, 0); and (iv) V(0, 0, 2) → V(2,1,0).
Solution The frequencies for (i) V(0, 0, 0) → V(2, 0, 0) transition = 667 × 2 = 1334 cm−1; (ii) V(0, 0, 0) →
V(0, 1, 2) transition = 1337 + 2 × 2349 = 6035 cm−1; (iii) V(0, 0, 2) → V(1. 1, 0) = 667 + 1337 + 2 × 2349 = 6702 cm−1;
and (iv) V(0, 0, 2) → (2, 1, 0) = 2 × 667 + 1337 + 2 × 2349 = 7369 cm−1.
Problem 3.23: Determine the value of ai(Cor.) in case of CO2 if w1 = 1337, w2 = 667 and w3 = 2349 cm−1,
Be = 0.3906 cm−1, and Ie = 18.816 × 10−40 g cm2.
Solution The value of ai (Cor.) can be calculated by substituting the values of different given constants into
Eq. (3.112).
4 (0.3906 ) × 667
2
(Cor.)
α1 0; α ( )
= . × 10 −55 cm
= 8.02 1
2
(2349 − 667) (2349 + 667)
and
−8 (0.3906 ) × 2349
2
(Cor.)
α3 = = −5 65 10 −4 cm
c −1
(2349 − 667) (2349 + 667 )
Problem 3.24: For CH4, the spacing of lines in the band v3 at 3020 cm−1 near the origin is 9.93 cm−1 and in the
band v4 at 1306 cm−1 is 5.74 cm−1. Determine the values of Coriolis constants for the two F2 vibrations ν3 and v4.
Solution For XY4 molecules,
1
ζ ζ4 = and sum of spacings in v3 and v4 = 3B
2
Therefore, 3B = 9.93 + 5.74 = 15.67 cm−1, and B = 5.223 cm−1.
This is an average of the B values of the upper and lower states. For fundamentals, the separation of successive
lines in the series = 2B(l − ζi)
Therefore for n3,
2B(1 − ζ3) = 9.93
2 × 5.223(1−ζ3) = 9.93
10.446 − 9.93 0.516

ς3 = = = 0 05
10.446 10.446
and, for n4; ζ4 = 0.45.
3.15 ROTATION–VIBRATION SPECTRA OF POLYATOMIC MOLECULES

When the molecule has an angular momentum about the internuclear axis, the vibrational change is then
calculated theoretically by rotational transitions. However, due to large difference between the moments of inertia
in the initial and final levels, very small changes actually occur in the rotational energy due to which a series of
very closely spaced lines, called the Q-branch appears at the centre of the branch, e.g. nitric oxide or polyatomic
(n > 2) molecules. This is shown in Figs 3.22 and 3.23.
Further, it is inferred that the rotational selection rules depend upon whether the vibration is accompanied by
a change in dipole moment perpendicular or parallel to the symmetry axis. The selection rules for parallel vibra-
tions of linear molecules such as HCN, N2O, CO2, C2H2, etc., are the same as for diatomic molecules, i.e.
ΔJ = ±1 and ΔV = ±1, (±2, ±3,...)
for anharmonic motion. The selection rules for perpendicular vibrations
ΔJ = 0, ±1 and ΔV = ±1, (±2, ±3,...)
reveal that a vibrational change can occur without simultaneous rotational transition. Consequently, the Q-branch
is observed along with the P- and R-branches in perpendicular bending mode of HCN. Further, the fine structure
of nonlinear polyatomic molecules depends upon whether the molecule has symmetric or asymmetric rotation. In
spite of the fact that the fine structure is complex in pure rotational spectra, the Q-branch will always be observed
in parallel and perpendicular bands of the molecules since in all cases ΔJ = 0, ±1.
Origin
1220 1240 1260 1280 850 900 950

−1
n(cm ) n (cm−1)
(a) (b)
Fig. 3.22 (a) Parallel, and (b) perpendicular modes of methyl iodide.
3220 3240 3260 3280 3300 3320 3340 3360 3380 3390 620 640 660 680 700 720 740 760 780 800
n(cm−1) n (cm−1)
(a) (b)
Fig. 3.23 The spectrum of (a) parallel symmetric stretch, and (b) perpendicular bending modes of HCN. Note the
P-, R-branches in (a) P-, Q-, R-branches in (b).
3.15.1 Structure and Characteristics of Rotation–Vibration Bands

in Polyatomic Molecules
For a vast majority of molecules, the rotational constants are so small that the infrared spectra taken at even the
highest resolution do not show individual rotational lines, the expected line spacings being smaller than Dop-
pler half widths. However, envelops of their vibration–rotation bands possess distinctive features, which depend
uniquely on the rotational constants of the molecule and the direction of change of dipole moment (parallel,
perpendicular or hybrid bands). The observables involved are (i) the separation of P- and R-branch peaks (ΔvPR);
(ii) the relative intensity of the Q-branch as compared to the integrated intensity of the whole band (IQ/ITotal)
(Fig. 3.24); and (iii) the actual shape of the band contour. If explicit relations could be established between these
observables and the molecular parameters, one could use band contour analysis to deduce these parameters. The
situation is, however, quite complex both from theoretical and experimental standpoints.
3.15.2 The Band Types

Recalling the definitions of the rotational constants (A, B and C) and the principal moments of inertia (IA, IB and
Ic ) as given in Chapter 2, one can easily define the momental ellipsoid (ME ). The ME is an ellipsoid with the
principal axes a, b and c such that
a = 2(IA)−1/2, b = 2(IB)−1/2 and c = 2(IC)−1/2
Here, a is the ‘major’, b the ‘intermediate’, and, c the ‘minor’ axis of inertia.
For linear and prolate symmetric top molecules, a band is referred to as parallel (||) or perpendicular ( ⊥) when
it corresponds to a dipole change parallel or normal to the unique (major) a-axis of inertia For oblate symmetric
top molecules, a band is referred to as parallel or perpendicular depending on whether it corresponds to electri-
cal oscillations parallel or perpendicular to the unique (minor) c-axis of inertia. However, for asymmetrical top
molecules, the letters A-type, B-type and C-type are used to qualify the bands, denoting that the transition dipole
moments are parallel to the a-, b- and c-axis respectively. Type A bands exhibit a fairly strong central maximum,
corresponding to the Q-lines, with two accompanying maxima on either side, corresponding to the P- and R-lines,
for any value of r(r = B/A → 1 or C/A → 1), the Zumbalt parameter. The contour of such type of band is shown
in Fig. 3.25(a). Type C bands have similar envelops, except for small r values when the central branch no longer
stands out, so that only one broad maximum results. The type C band contour is shown in Fig. 3.25(b). The P and
R maxima in type C bands are broad and are not as pronounced as in the case of type A bands. Type B bands have
no central maximum but, the Q lines form two maxima, i.e. QQ or QQ′ on either side and fairly close to zero line,
in addition to the P and R maxima. A contour of type B band is presented in Fig. 3.25(c). These Q maxima are
prominent particularly for the larger r values. For small moments of inertia the intensity distribution is not quite
symmetrical about the band origin. Sometimes instead of QQ′ maxima only a single maxima is observed with
P- and R-maxima on either side of Q as shown in Fig. 3.25(d). For pure B-type bands the Q positions are obtained
from the positions of minimum between the P- and R-branches. IQ/ITotal may also be fixed as a criterion to charac-
terise the type of bands in the spectra of molecules. The range of intensity ratio of different types of virgin bands
is different. It is to be noted that in the spectrum of molecules, it is not always possible to examine all the bands for
ΔvPR and IQ/ITotal values due to either the complex structure resulting from overtones, binary combinations, Fermi
resonance, hot bands, etc., or their very feeble intensities, especially in isotopically labelled molecules. For all
practical purposes, only those bands are chosen which exhibit no overlapping with other bands.
PQR thus implies that the band consists of three branches with the P(ΔJ = −1) on the low-frequency side and
the R-branch (ΔJ = +1) on the high-frequency side of the Q branch (ΔJ = 0). PQQR or PQQ′R means that the
band structure with four maxima is seen. By PR separation we mean the interval (in cm−1) between the intensity
maxima of P- and R-branches.
lQ
= 0.172
lTOTAL
Q
P Q
R (a) (b)
R
P A Type C Type
Absorbance
lQ lQ
= 0.176 = 0.237
lTOTAL lTOTAL
P R
Absorbance
Q
(c) B Type (d) B Type
lQ lQ
Q Q = 0.362 P R = 0.203
lTOTAL lTOTAL
P R
−1
Frequency(cm )
Frequency (cm−1)
Fig. 3.24 Representative curve for-calculating
IQ/ITOTAL in an A-type band. Shaded portion indi- Fig. 3.25 Nature of typical (a) A -, (b) C -, and (c, d) B-type infrared
cates the area attributed to Q-branch. bands.
3.15.3 Theoretical Aspects of PR-separation and IQ /ITotal

in Different Types of Molecules
(a) Linear Molecules For a linear molecule the PR separation, δ in a vibration–rotation vapour phase band
contour at TK is given by
1
⎛ 8B T ⎞ 2
Δv PR ≡δ =⎜ = 2.358 TB (cm −1 ) (3.122)
⎝ hc ⎟⎠
where B is the rotational constant and other symbols have their usual meanings. In the derivation of this formula,
the change of rotational constants in different vibrational states, the centrifugal distortion effects and the compli-
cations resulting from Coriolis interaction have been neglected; and it has also been assumed that the intensity
of rotational transition is primarily governed by the product of the statistical weight and the Maxwell-Boltzmann
distribution function for the particular energy level from which the absorption is taking place.
(b) Symmetric-Top Molecules For symmetric-top molecules, the expression for PR separation in parallel
bands can be represented as
ΔvPR (||) = S(b) . δ (3.123)
where parameter b is defined as
b = (A/B) −1 or (C/B) – 1 (3.124)
for prolate or oblate molecules respectively and the function S(b) is given by an empirical relation
0.721
log (β ) = (3.125)
(β )1.13
which is found to hold good within 0.5 per cent for b in the range −0.5 to +100. The expression for the ratio
IQ/ITotal in the parallel-type bands is given by
IQ ln[β1// + ( + β ) / 2 ] − [β /( + β ))]1/ 2
= (3.126)
I Total β[β /( + β )]1/ 2
for b > 0, a case in which molecules are all near prolate symmetric tops. Thus, IQ/ITotal is a function of parameter
‘b ’ alone. It tends to ~0.5 for disc-shaped molecules (b = −1/2), ~0.33 for a spherical-top molecule (b = 0), and
falls to 0 for a linear molecule (b = ∞).
For the perpendicular bands in symmetric-top molecules, there are no general formulae for ΔvPR or for IQ /ITotal.
For (⊥) bands, (i) the Q-branch gets stronger with b increasing from −0.5 to ~1, beyond which the bands rapidly
cease to show PQR structure, and (ii) ΔvPR decreases monotonously with increasing b.
(c) Asymmetric-Top Molecules For the asymmetric-top molecules, the energy level scheme becomes more
complex, being expressed (for a near prolate molecule) by
F (J , τ ) = BJ (J ) (A − B ) Eτ (κ a ) (3.127)
B is defined by
1
B (B + C ) (3.128)
2
The asymmetry parameter κa is defined by
1
B
2
(A + C )
κa = (3.129)
1
(A C)
2
and the parameter l is defined as
l = (A − C)/B (3.130)
Since large J values are involved in heavy molecules, one can obtain approximate solutions for the (2J + 1) values
of F(J, τ) in terms of κa and l to obtain band contours. Eτ(κa) represents the (2 J + 1) solutions.
For both parallel and perpendicular-type bands of the prolate and oblate asymmetric-top molecules, general
sets of empirical expressions for PR separations applicable in various ranges of κ and l are available. For parallel
bands of near prolate symmetric tops, the expression used for determination of PR separation is in the form
Δv PR A (||) = S (b ).d (3.131)
where δ and S (β ) are obtained on replacing B by β in the expressions for δ and S(b) respectively.
For B(⊥) and C(⊥) bands in the range 3/4 < l < 3.
Δv PR B ( ⊥) = δ (3.132)
3
and Δv PR C ( ⊥) = [S ( b ).d ] (3.133)
2
Expressions are not available for the ratio IQ/ITotal for bands of asymmetric top molecules. However, we can
extend the application of relation (Eq. 3.126) to asymmetric top molecules just by replacing b with β .
(d ) Hybrid Bands PR separation in hybrid bands can be obtained from the empirical formula
Δv ( ) cosα Δv (B ) β
Δv PR (αβ ) = (3.134)
cosα + cosβ
where a and b are the angles between the oscillating dipole and the a and b axes of the molecule, and Δν(A) and
Δν(B) are the PR separations for pure A-type and pure B-type bands, respectively. The uncertainties in measure-
ment of ΔvPR values (in hybrid bands in particular) are so large that relation (Eq. 3.134) can at best be used to
check hybrid character but not for deducing the direction of dipole oscillation within a narrow range. In fact, a
simple formula like
αΔv ( ) βΔv (B )
Δv PR (αβ ) = (3.135)
α β
gives as good a fit as the formula (3.134) in almost all cases.
Problem 3.25: The vapour-phase spectrum of methylthiolformate shows resolved rotational contours. The
PR separations as determined from the C = Ο stretching (1708, 1698, 1688 cm−1) and CH3 symmetric stretching
(2954, 2945, 2937 cm−1) bands, both of type A have been found to be 20 and 19 cm−1 respectively. Determine (a)
the moment of inertia of the molecule, (b) the values of the parameters l and κa and (c) the configuration of the
molecule if the rotational constants of the cis- and trans- configurations are
Rotational Constants (MHz) cis trans

A 11149.9 19392.8
B 5009.7 3577.9
C 3533.2 3078.9
Solution (a) Moments of inertia are determined from the rotational constants by making use of the conversion
factor.
0537 105 (amu ) ( A ) 2

h −1 °
= 5.0537
8π 2
h
Moment of inertia ( MI ) =
8π × Rotational constant
2
MI (au) cis trans

IA 45.3258 26.0601
IB 100.8800 141.2501
IC 143.0371 164.1427
(b) Using expressions (3.129) and (3.130), we obtain

1
5009.7 .9 + 3533.2)
(11149.9
κ a ( )= 2 = −0.6123
1
(11149.9 − 3533.2)
2
1
3577.9 − (19392.8 + 3078.9)
κa( )= 2 = −0.99388
3
1
2
(19392.8 − 3078.9)
( .9 − 3533. )
I cis = = 1.5203
5009.9
I trans =
(19392.8 − 3078.9) = 4.5596
3577.9
(c) In order to determine the PR separations of Type-A, -B and -C band contours, the parameters, B , β and
S (β ) need to be defined.
Using Eqs (3.128), (3.124) and (3.125) we get
cis trans
1 1
B cis = ( 5009.7 + 3533.2) Btrans = ( 3577.9 + 3078.9)
2 2
= 4271.45 MHz = 3328.4 MHz
11149.9 19392.8
β cis = −1 β trans = −1
4271.45 3328.4
= 1.6103 = 4.8264
0.721 0.721
log ( β ) cis = 1.13 log ( β ) trans = 1.13
( .6103 + 4 ) ( .8264 + 4 )
= 0.102 = 0.0615
S ( β ) cis = 1.2667 S ( β ) trans = 1.1521
Substituting the values of Bcis and Btrans into Eq. (3.122) we get δcis = 16.6 cm−1 and δtrans = 14.0 cm−1
respectively.
Now the PR separations for the cis- and trans-configurations are determined from Eqs (3.131), (3.132) and
(3.133) and are as follows.
Band Type cis trans

A 21.0 16.1
B 16.6 14.0
C 31.5 24.1
The theoretical value of PR separation for the type A band for the cis configuration conforms to the PR separa-
tions observed experimentally for the C = O (19 cm−1) and CH3 (20 cm−1) stretching bands both of type A. Thus,
in methylthiolformate, the methyl group is cis to the carbonyl group.
Problem 3.26: The absorption band at 920 cm−1 in the vapour-phase infrared spectra of methylthiolformate
has a type C contour with a strong Q-branch and is assigned to CH (formyl) out-of-plane wagging. Using the
data from Problem (3.25), find the spacing of the Q-branch fine structure on the P and R shoulders of a type C
contour.
Under the operating condition of the instrument, the rotational fine structure of the lines with a separation of
1 cm−1 or greater is observable.
Solution The presence of such a C-type contour suggests that the molecule has either planar or near planar
molecular geometry. The spacing of such a Q-branch fine structure of a type C band is determined from the
expression
2⎡ ( B + C ) ⎤ −1
Δv = ⎢ A− cm (3.136)
c⎣ 2 ⎥⎦
The values of Δvcis and Δvtrans are found to be of the order of 0.46 and 1.07 cm−1 respectively.
Under the operating conditions of the spectrometer, the absence of rotational fine structure lines suggests that
the molecule most probably exists in cis-configuration as shown below.
O
H C CH3
S
3.16 APPLICATIONS OF IR SPECTROSCOPY

The applications of infrared spectroscopy have been discussed under three parts, viz. qualitative, quantitative and
biological applications.
3.16.1 Qualitative Applications

By qualitative applications in infrared spectroscopy, we mean functional group identification, skeleton identifica-
tion and structural elucidation on the basis of spectral observations and any other additional information if avail-
able. Infrared spectroscopy is of great help, if information on elemental analysis is also available. This method
enjoys advantage over all other methods, in that it supplies information of whole molecule, composite and whole,
but not about protons as in nuclear magnetic resonance, electronic transitions as in ultraviolet/visible and about a
free radical as in electronic spin resonance spectroscopy. Thus, IR spectroscopy is used in almost every possible
way to study chemical systems. It is generally an easy matter to place a sample in a spectrometer and examine its
absorption spectrum and see whether it has changed due to some change in conditions or not. Some of the typical
qualitative applications of IR spectroscopy are discussed here.
(a) Structural Elucidation The spectra of compounds is divided mainly into two parts: (i) above 1500 cm−1,
and (ii) below 1500 cm−1. The region above 1500 cm−1 is mainly due to the normal vibrations of functional groups
while below 1500 cm−1, called the fingerprint region contains mainly skeletal motions due to normal bending vi-
brations and fundamental stretching vibrations of bonds like C—X where X = C, N, O, Cl, Br, I, F. The fingerprint
bands are widely used as means of identifying structural units within polyatomic molecules. Further, isotopic sub-
stitution plays a major role in structural elucidation since it helps in sorting out the skeletal and functional group
vibrations in complex molecules. These days, different detailed and comprehensive correlation tables in which
the region and exact positions of the frequencies due to different modes of vibrations are available. The infrared
spectra-structure correlations by regions given by NB Colthup and the characteristic infrared frequencies of some
functional groups are given in Table 3.8. Thus, simply by matching the spectral data of unknown compounds with
the standard ones reported in the tables, one can propose a tentative structure to the compound. However, it should
be borne in mind that the changes in the group frequencies in different molecules may be better understood if
the interaction between the groups and the environment is specifically considered. Some of the important factors
which affect the frequencies are explained here.
(i) Mass Distribution The vibrational frequency of any vibrator is usually governed by its mass and force
constant. A change in the mass distribution in a group of molecules alters the form as well as the frequency of
vibration; an increase in mass usually causes a decrease in frequency.
The effect of nature of mass on the C = O vibrational band is shown in Fig. 3.26. It is estimated that these
mechanical changes in the nature of C = O vibration lowers its frequency by about 15 wave numbers, when heavier
subsituent on the carbonyl carbon [Fig. 3.26(a)] is replaced by the lighter atom, i.e. hydrogen [Fig. 3.26(b)]. Thus,
the lowering in frequency of C = O is mainly controlled by mass effect while an increase of 15 cm−1 in going from
ketones to aldehydes in mainly due to the fact that aldehydes have larger C = O force constant than ketones.
Table 3.8 Infrared spectra-structure correlations (in cm−1).
Correlations by regions
3700−3100 OH, NH, ≡ C—H
3100−3000 olefinic, aromatic, three membered ring CH
3000−2700 aliphatic CH
2300−1900 X ≡ Y, X = Y = Z
1900−1550 C=O
1700−1550 C = C, C = N
1600 and 1500 aromatic ring
1460 CH2, CH3
1375 C—CH3
1300−1000 C—O
1000−600 olefinic, aromatic CH wag
Aliphatic groups
CH3 2960, 2870 and 1460 (±10)
C—CH3 1375
C(CH3)2 1385 m and 1370 m
C(CH3)3 1394 w and 1370 s
O = C—CH3 1370 s
CH2 2925, 2850 and 1460
CH2—CH2—CH2—CH2 724
O = C—CH2, N ≡ C—CH2 1420
Olefins
C=C 1680–1630
Conjugated C = C—C = C 1650 and 1600
R—CH = CH2 3080 w 1820 w 1645 m 1415 m 990 s, 910 s
R2C = CH2 3080w 1780 w 1645 m 1415 m 890 s
R—CH = CH—R cis 3030 w 1645 m 1415 730–650 m
R—CH = CH—R trans 3030 w 1670 w 970 s
R2C = CHR 3030 w 1670 w 840–790
(R-hydrocarbon group)
Aromatics
aromatic CH 3100−3000
1600,1580,1500.and
aromatic ring frequencies
1450
5 adjacent, H780–730 mono 710–680
4 adj. H770–735
3 adj. H 795–770 meta 710–680
2 adj. H850–795
1 lone H890–835
Overtones and combinations 2000–1700
in plane CH bend 1200–1000
Triple and cumulated double bonds
C—C ≡ C—H 2140–2100 ≡ C–H 3340–3267
C = C = CH2 2000–1900
CH2—C ≡ N 2260–2240 CH2—CN 1420
conj—C ≡ N 2235–2220
(Continued)
Table 3.8 Infrared spectra-structure correlations (in cm−1) —Cont'd.

—S—C ≡ N 2170–2135
—N = C = S 2150–2050
—N = C = O 2275–2263
Carbonyls
1695–1630 NH2 3350,3180 and 1630 NH;3300 and
R—CO—N
1550
R—CO—OH 1720–1680 OH ca 3000 broad
1725–1705, singly conjugated 1700–1670; doubly
R—CO—R
conj. 1680-1640
1740–1720 conj. 1710–1685 CH 2900–2800 and 2775
R—CO—H
– 2695
R—CO—O—R 1750–1735 conj. 1735–1715
strained ring C = O 1850–1750
R—O—CO—CI 1785–1775 C—O 1172–1134
R—CO—Cl 1810–1795 aryl–CO—Cl 1785–1765 and 1750–1735
R—CO—O—CO—R 1825–1815 s and 1755–1745 m
cyclic anhydride 1870–1845 w and 1800–1775 s
C—O
CH2—O—CH2 1140–1085
CH2 = CH—O—CH2 1225–1200
Aryl—O—CH2 1310–1210 and 1050–1010
epoxy 950–815, 880–805 ca 3050
alcohol ca 3300 broad
CH2—OH 1075–1000
R2(CH—OH) 1125–1090
R3(C—OH) 1210–1100
Aryl-OH 1260–1180
Nitrogen compounds
NH2 3370 3300 3200 (w) 1600
NH 3350
CH2—NH2 1080
CH2—NH—CH2 1130
Aryl-NH2 1300
N−R3 2800
C=N 1690–1630
C = N—O 1000–900
O = N—O—C 1650 trans, 1610 cis and 814–751
O2N—O—C 1640, 1280 and 870–840
CH2—NO2 1550 and 1375
CH3—C—NO2 1550, 1390 and 1360
Cl—C—NO2 1570 1350
arom. NO2 1515 1350
R—C ≡ N 2240–2260 (m)
Halogens
C—Cl 830–560
C—CH2—CH2—Cl 725 and 650
CH2—Cl 1300–1240

cyclic C—Cl 742 eq, 688 ax
aryl C—Cl 1094–1035 p→ m → o
C−Br 700–515
C—CH2—CH2—Br 644 and 563
CH2—Br ca. 1240
cyclic C—Br 685 eq. 658 ax
aryl C-Br 1075–1025 p→ m → o
C—F 1000–1400 (s)
C—I ∼500 (s)
Sulphur
SH 2590–2540w
C—SO2—C 1320–1140
C—SO2—N 1340–1160
C—SO2—OH 1350–1160
C—SO2—O—C 1350–1180
C—SO2—Cl 1375–1180
C—SO2—F 1400–1205
C SO3− H 3O + 1200–1050
C SO3− Na + 1200–1050
C—S 600–800 (w)
C=S 1050–1200 (s)
Boron
BH 2640–2350
B .. H .. B 2220–1540
B—O 1380–1310
B—N 1465–1330
B—OH 3300–3200
Phosphorus
PH 2440–2275
P=O 1300–1140
P—OH 2700–2200 very broad
P—O—C 1050–970
P—O—C2H5 2990, 1485, 1450, 1395, 1375, 1160
P—O—CH3 2960, 1460, 1180
Silicon
Si—-H 2250–2100
Si—-O 1100–1000
SiCH3 1265–765
Si(CH3)2 1265–860 and 800
Si(CH3) 1265–840 and 765
3
Inorganic ions
2−
3 1410–1450(v.s.)
860–880(m)
SO 2−
4 1080–1130 (v.s.)
610–680 (m)
NO3− 1350–1380 (v.s.)

815–846 (m)
(Continued)
NO 2− 1320–1380
1230–1250 (v.s.)
800–840(w)
NO 2 ⎫ 1618, 1320, 750

⎪⎪
NO 2− ⎬ Fund. frequencies 1328, 1261, 828
⎪
NO 22 − ⎪⎭ 1298, 1238, 880
NH+4 3030–3300 (v.s.)

1390–1430 (w)
3− 2−
PO , HPO , H 2 PO 4
4 4 1000–1100 (s)
CN − , SCN , OCN − 2000–2200 (s)
Silicates 900–1100 (s)
v.s. = very strong; s = strong; m = medium; w = weak; b = broad.
(ii) Electronegativity The intrinsic power of an atom to attract the O

R
electrons within the molecule is called its electronegativity. A covalent (a) H C C OR C O
3
bond between C and F develops an ionic character, because a partial R
negative charge at F and equivalent positive charge on C are produced
due to higher electronegativity of F than C. The coulombic attraction 1740 cm−1 1715 cm−1
between opposite charges makes the bond stronger than a purely cova-
lent one. Thus, the position and intensity of the corresponding band in O
R
the spectrum will be affected due to the ionic character of the bond. (b) H C OR C O
For example, the C—H stretching frequency in HCN is observed at H
3312 cm−1 while that of F—C in FCN is observed at 1250 cm−1. This
lowering of frequency is not mainly due to mass effect but due to higher
1720 cm−1 1730 cm−1
ionic character of F—C bond (∼43 per cent). The relatively higher ionic
character makes F—C bond stronger and hence large value of F—C Fig. 3.26 The C = O mass effect: (a) in acetates/
ketones, and (b) in formates/ aldehydes.
force constant (9.2 millidynes/Å in comparison to that of C−H bond
where the value of force constant has been estimated to be of the order
of 5.8 millidynes/Å). Percentage of ionic character = l00[1−exp{-0.25(XA − XB)2}] where XA and XB are the elec-
tronegativities of atoms A and B respectively on the Pauling scale.
(iii) Inductive Effect The action of one group of a molecule to affect electrostatistically the electron distribu-
tion in the other groups is described as the inductive effect. It may act across space as well as through bond. Due to
the change in the charge distribution in the molecule, the frequencies of many bonds may be altered. The inductive
effects on C = O stretching frequencies are shown in Fig. 3.27. It is inferred from Fig. 3.27 that electron attracting
groups tend to raise while electron repelling groups tend to lower the frequency of the C = O stretching band.
(iv) Resonance Splitting As a rule the vibrations of different symmetries, e.g. the in-plane and out-of-plane
vibrations in a planar molecule do not interact with each other. However, when each of the two bonds oscillates
with a common atom having nearly the same individual frequency, they interact very strongly. The resulting
+ −
C O C O C O Cl O H O
C C C C
R − CO − R = 1715 cm−1
H H H Cl
Cl − CO − R = 1800 cm−1 (b)
Cl − CO − CI = 1828 cm−1
(a)
Fig. 3.27 Inductive effects on C = O stretching frequencies (a) through bond, and (b) across the space (field effect).
frequencies are displaced from their original positions, depending upon the relative phasing and coupling of two
oscillators, e.g. in allene (CH2 = C = CH2), because of the common atom, the coupling between the two C = C
stretching vibrations is so strong that the in- and out-of-phase C = C stretching vibrations will be in the single
(1071 cm−1) and triple (1980 cm−1) bond regions respectively.
(v) Mesomeric Effect This is a resonance effect and may be affected by electron-loving and hating groups
or due to conjugation. Consequently, the group frequencies may shift due to this effect. It can be shown math-
ematically from wave mechanics that when an electron exchange is possible and does take place, the energy
of resonating structures is less than that of the molecule without exchange. Thus, resonance energy leads to an
increase in stability of the resulting structures. Resonance may arise from oscillations of electrons from one posi-
tion in the molecule to another. This is what happens in the case of C6H5CHO, or C6H5OH and we get resonance
structures. In C6H5CHO, the exchange of p-electrons takes place between C = O bond and benzene ring and we
call it by the name of conjugation. For effective conjugation, the two parts should be planar or nearly planar.
The influence of this effect on the stretching vibration of the carbonyl group in some typical molecular systems
is shown in Figs 3.28(a) to (c). The decrease of carbonyl double bond character, i.e. carbonyl stretching fre-
quency in amides, esters and acid chlorides due to this effect follows the order: amides > esters > acid chlorides
[Fig. 3.28(a)]. It conforms to the expected trend on the basis of mesomeric electron donation of heteroatoms, i.e.
nitrogen, oxygen and chlorine to the C = O group, i.e. N > O > Cl. Further, when electron withdrawing group is
present on the ester oxygen or amide nitrogen, the mesomeric electron donation decreases, thereby the carbonyl
frequency for esters or amides having such groups increases, [Fig. 3.28(b)]. Figure 3.28(c) exhibits another type
of mesomeric effect, i.e. conjugation. When C = O is conjugated with C = C or aromatic ring, the delocalisation
of electrons results in lowering of the C = O frequency.
(vi) Hydrogen Bonding When a proton is attached to a more electronegative atom, say A (= Cl, O, F, etc.),
forms a partial bond with a neighbouring electronegative atom, say B (= Cl, O, F, etc.) or a p-bond, e.g. C = O,
C = C, aromatic ring, etc., it is known as hydrogen bonding. It may be intramolecular or intermolecular; the
former being stronger than the latter.
The strongest hydrogen bonds are formed when the atomic centres A, H and B are collinear. The energy of a
hydrogen bond is of the order of a few kilocalories (10 kcal mole−1); weaker by a factor of 10 than the covalent
bond between atoms and stronger by almost the same factor than the Van der Waals bond.
The vibrations of the H-bonded complex A—H …... B useful in H-bonding studies are; (i) the fundamen-
tal A—H stretching frequencies, v(A—H); (ii) A-H in-plane deformation (bending) frequencies, δ(A—H);
O C Cl O C O O C N
(a) −
O C Cl+ O− C O+ O− C N+
−1 cm−1 −1
1800 cm 1740 1680 cm
O O
(b) CH3 C O CH2 CH2 CH3 C O CH CH2
1740 cm−1 1770 cm−1
(c) CH2 C CH2 1715 cm−1
CH2 C R 1685 cm−1
R C R 1655 cm−1
R C C or
Fig. 3.28 (a)-(c) Mesomeric effect on carbonyl stretching frequencies.

(iii) A—H out-of-plane deformation (bending) frequencies, g (A—H); (iv) AH ….

B fundamental stretching frequencies, ν(H... B); (v) AH... B deformation (bending)
frequencies, δ(H…B) and g (H….B). These are represented in Fig. 3.29. (a) A H B
The main effects of hydrogen bonding on the vibrations of the participants are as
follows:
(b) A H B
• The absorption bands due to the A—H stretching vibrations (fundamentals and
overtones) are shifted to lower frequencies. These shifts range from about 30 cm−1
to several hundred cm−1 or more. This effect is due to the weakening of the force (c)
A H B
constant for the A—H stretching mode caused by the formation of the H bond.
• The shifted absorption bands due to the H bonded A—H stretching vibration are
much broader than the corresponding bands of the non-H bonded A—H groups. (d)
A H B
The change in band width (Δv1/2) varies from ~30 cm−1 to 100 cm−1 or more. The
breadth and structure of H bonded A—H stretching vibrations are affected very
little by change in phase or temperature. (e) A H B
• In addition to the broadening, the integrated intensity of fundamental A—H
stretching bands grows sometimes by factors of up to ten or more. However the
corresponding overtones decrease slightly in integrated intensity. The reasons for
Fig. 3.29 Vibrations of hydro-
these intensity effects are probably related to the fact that H bonds have sub- gen bonded complex A—H …. B.
stantial electrostatic character and the intensity of absorption due to IR active
vibrations is directly proportional to the rate of change of the electrostatic dipole moment with internuclear
distance.
• The A—H deformation modes are shifted to higher frequencies. These shifts are appreciably smaller as com-
pared to that of A—H stretching vibrations. Formation of H bonds constraints the deformation vibrations and
therefore increases the force constants for these modes.
• The A—H deformation modes do not show any substantial band broadening or intensity change when H bond-
ing occurs.
• New vibrational modes corresponding to H….B stretching and deformation are found at low frequencies in the
far IR region.
• The vibrational modes of the H bond acceptor, B are shifted by H bonding. These shifts may be either to longer
or shorter wavelengths, and are generally much smaller than those found for the donor A—H vibrations.
(vii) Dielectric Constant The stretching frequencies of polar groups of molecules in nonpolar solvents depend
on the dielectric constant D of the solvent. The frequency vs in the solution phase is related to dielectric constant of
the solvent by the Kirkwood–Bauer–Magat equation, i.e (v0 − vs)/v0 = K(D − 1)/(2D +1). Here v0 is the frequency
in the vapour phase and K is a proportionality factor.
(viii) Bond Angle Ring Strain The observed C = C frequency shift with bond angle ring strain in alicyclic
systems is caused by a change in interaction between C = C and C—C stretching vibrations rather than by a
large change in the force constant. In cycloalkenes, (CH2)n—2H where n = 3, 4, 5, 6, when C = C—C angle
is 90°, the interaction between C—C and C = C stretching vibrations is nearly zero. As a result, the C = C fre-
quency is expected to be minimum when C = C—C angle is 90° and progressively grows as the angle gets larger
or smaller. On the other hand in alkylidene substituted cycloalkanes, [(CH2)n—2H] = C, the exocyclic C = C
stretching frequency progressively increases with the bond angle strain, i.e decreases as the cyclic C = C—C
angle decreases. In l, 2-dialkyl-cycloalkenes; C—[(CH2)n—4H]—C; a decrease in cyclic C—C = C angle (which
lowers the frequency) is compensated by increase in exocyclic C—C = C angle (which raises frequency). The
C = C stretching frequency, therefore, remains unchanged until the ring angle becomes smaller than 90°. A further
decrease in the cyclic C—C = C angle then raises the frequency.
In lactones, the C = O and C—O frequency shifts with ring size are primarily caused by the mechanical
effects rather than by large changes in force constants. As the ring size decreases, the C = O stretching frequency
increases whereas C—O decreases, The divergence in the C = O and C—O stretching frequencies with ring size
is caused by the change in interaction between C—O and C = O stretching vibrations.
Since all these factors affect the characteristic group frequencies to different extent, one must be very careful
while assigning a particular band unequivocally in an unknown spectrum. In large molecules, the assignment of
normal modes, their overtones, combinations (sum or difference), bands becomes fairly complicated. However,
if the point group for the molecule in vapour phase is known by any other method, the vibrations can be divided
into various symmetry species of the group. The selection rules from the character table will provide information
about the allowed and forbidden transitions which will ultimately help
the vibrational assignments, e.g. the data on the IR vibrational frequen-
cies (in cm−1); 3380 m, 3020 m, 2970 m, 2585 vw, 1735 vst, 1425 m,
Optical density
1365 s, 1287 m, 1225 w, 1128 vst, 1072 w, 1000 s, 860 w, 830 m and 1970
628 vst (in Raman only) and elemental analysis, C = 31.57, H = 5.26 and
S = 42.10 per cent correspond to the structure CH3COSH. The absence 2070
of any band in the OH stretching and bending regions rules out the pos-
sibility of structure CH3CSOH. Deuteration effect also conforms to
these conclusions.
Polarised infrared spectra of Re2(CO)10 when recorded in a suit-
2100 2050 2000 1950 1900
able liquid crystal solvent show three bands at 2070 m, 2010 vs and −1
Frequency (cm )
1970 w cm−1 as shown in Fig. 3.30. Fig. 3.30 Polarised infrared spectra of
Re2(CO)10 molecules orient with their long molecular axis which is Re2 (CO)10 in a liquid crystal in the 1900-2100
parallel to the direction of infrared incident beam. Thus, the two bands cm−1region.
at 2070 and 1970 cm−1 are polarised parallel (rir > l) to the long molecu-
lar axis whereas the band at 2010 cm−1 is polarised perpendicular (rir < 1) to the long molecular axis as expected.
Me2(CO)10 molecule belongs to D4d symmetry and hence have three allowed C = O stretching normal modes of
vibrations, i.e. 2B1 + E1. According to group theory, E1 vibration is polarised perpendicular and B2 vibrations are
polarised parallel to the molecular axis.
(b) Identification of Molecular Species Since IR spectra are differentiable for many different species, one
can easily make use of this property for the identification of molecular species in a mixture and also molecular
species of the same compound in the solution. The presence of three bands at 3638, 3632 and 3627 cm−−1 in the
OH stretching region in the high resolution spectrum of 1-propanol in dilute CC14 indicate the presence of three
OH groups in conformationally distinct environments. The spectrum of 1-propanol in the OH stretching region
and equilibria between its rotamers are shown in Figs 3.31 and 3.32 respectively.
Recently, IR spectroscopy, low and high resolution in conjunction with other spectroscopic and non-spectro-
scopic techniques has proved to be a boon to the space chemists for the detection of various molecular species in
space, e.g. spectral analysis has revealed the presence of CO2 (97 per cent), N2 (< 2 per cent), H2O vapours (~1 per
cent) and O2 (~0.1 per cent) in the atmosphere of Venus. The absence of amide bands in Venus material due to the
peptide linkages in proteins rules out the possibility of life on this planet as expected earlier.
(c) Study of Environmental Effects of Molecular Systems Since molecules are generally affected by their
environments, the IR spectral shifts can give valuable information about the inter- as well as intramolecular inter-
actions. This is demonstrated with the aid of the following examples. The IR spectroscopic studies on acetic acid
reveal that it exists mainly as dimer due to intermolecular H-bonding at room temperature while at 150°C it is
predominantly monomeric. The broad OH band at 3000 cm−1 observed at room temperature appears at 3580 cm−1
as a narrow band when temperature is raised to 150°C. The in-plane and out of plane bending vibrations of
3638
Ia Ib
I 3632 H H
H′ H′
H3C H H3C
II 3627 C C C C
H O −ΔH0 = 0 H H O
H′′ H′′
III
−ΔH0 = 0.55 −ΔH0 = 0.16
H
H H
O O
H 3C H H3C
C C C C
H′′ −ΔH0 = 0.38 H′′
H H
H′ H′
3650 3620 III II
−1)
Frequency (cm Fig. 3.32 Equilibria between the rotamers of 1-propanol
Fig. 3.31 The ν01(O—H) band of 1-propanol in CCI4. (ΔH0 in kcal mole–1).
the OH group at 1400 and 1000 cm−1 shift to higher frequencies at low temperatures where acetic acid is more
associated. These results are supported by deuteration effect also.
Spectra of ethyl acetate in ethanol and ethanol + water show three bands at 1740, 1725 and 1705 cm−1. They
may be due to (a) the free C = O group of ethyl acetate, (b) 1 : 1 bond complex C = O ..... H—O—C2H5, and (c) 1:2
H bonded complex.
C O H O C2H5
H O C2H5
The IR spectra of thiolacetic acid and thiolbenzioc acids in both the fundamental and first overtone regions of
the S—H stretching vibration in a mixture of CC13F and C2F4Br2 in the temperature range 20 to −190°C reveal
that S—H…. O type bonds are formed at low temperature or at high concentration of solute. The existence of two
associated bands (2570 cm−1, 2530 cm−1) in the S—H stretching region and the virtual absence of the free S—H
and C = O bands at low temperature seems to show that only the cyclic structure (I) remains at the lowest tem-
perature but the possibility of having open structure (II) is likely at intermediate temperatures. Chain and cyclic
O
R C
2530 cm−1 1686 cm−1
S H O
O H S 2570 cm−1 C R
H S
R C C R
S H O
1686 cm−1 2588 cm−1
(I) (II)
Fig. 3.33 The cyclic and open-chain structures of aliphatic and aromatic thiolacids.
structures are shown in Fig. 3.33.

If the molecule is physically adsorbed it is subjected only to weak intermolecular forces of the van der Waals
type and thus its symmetry is only slightly perturbed from that of the gas phase. Accordingly, the infrared spec-
trum is altered only slightly, and small frequency shifts, usually less than 1 per cent are observed (some Raman
vibrations inactive in IR become IR active because of adsorption). During the chemisorption process, however,
the symmetry of the adsorbed species is completely different from that of the gaseous molecule, the surface bond
is very strong, and the adsorption may be dissociative in nature. In this case, a completely new infrared spectrum
is observed and band shifts and intensities are far removed from those of the gaseous adsorbate.
C—O stretching frequency observed in gaseous carbon monoxide at 2143 cm−1, is shifted to 2000−2100 cm−1
after monodentate ligand formation with transition-metal compounds of Ni, Fe, Co, Mn and Re. Bidentate ligand
formation causes a greater shift in frequency, the C—O stretch usually is observed below 2000 cm−1. Thus, the
bands above 2000 cm−1 are attributed to a ‘linear’ CO group on the surface, Nis—C = O (type A), and those
below 2000 cm−1 to a ‘bridged’ CO structure, (Nis)2 C = O (type B). CO chemisorbed on silica supported Fe,
i.e. Fe/SiO2 gives rise to two bands at 1950 and 850 cm−1. They are attributed to C—O and Fe—C stretching vibra-
tions respectively. The force constants for the C—O and Fe—C bonds are of the order of 14 and 4 millidynes/Å,
respectively. The Fe—C bond distance as computed from the application of Pauling electronegativity data has
been found to be 1.67 Å and the bond is estimated to have 12 percent ionic character. The bond order of both of
the C—O and Fe—C bond is ∼2. In view of this, the surface species is written as Fes = C = O.
Adsorption of ammonia on well-dried silica–alumina surfaces reveals that coordinately bonded ammonia gives
bands at 3335, 3280 and 1610 cm−1 and the adsorbed NH4+ ions exhibit bands at 3270 and 1440 cm−1.
Adsorption of the isotopic species 14N2, 14N15N and 15N2 on Ni-SiO2 surface gives rise to separate distinctive
bands in the spectrum at 2195, 2160 and 2123 cm−1 respectively, thereby suggesting that dissociation of the
nitrogen molecule does not take place during the adsorption process. The force constants determined from the
isotopic shifts are found to be of the order of 19.1 × 10−5 and 3 × 10−5 dynes/cm for the N—N and Ni—N bonds
respectively.
Reaction product obtained on oxidation of ethylene depends upon the nature of catalyst used and there is a cor-
relation between ethylene frequency and catalyst frequency. NO2 exhibits three vibrational modes v1, v2 and v3 at
1321, 648 and 1621 cm−1. In the presence of NO2, ethylene is oxidised by molecular oxygen to formaldehyde, i.e.
C2 H 4 ⎯O ; NO 2
→ CH 2 O
The loss of oxygen from the NO2 molecule proceeds via the v3 vibration (1621 cm−1) since C2H4 also exhibits
an identical frequency at 1625 cm−1. Thus, only the vibrations of catalyst and reactant having identical frequencies
will cause fragmentation. Hydrolysis of ethylene in presence of Al2O3 yields ethanol. The ethylene frequency at
950 cm−1 corresponds to the AlO frequency of the catalyst at 964 cm−1. Ethylene oxide is obtained as an oxidised
product when AgO is used as a catalyst. The first overtone of AgO at 962 cm−1 corresponds to the ethylene fre-
quency at 950 cm−1.
The stretching frequency of NO is observed at 1843 cm−1 whereas those of NO+ and NO− falls in the range
1575−1940 cm−1 and 1040−1200 cm−1 respectively. The adsorption of NO on nickel surface exhibits bands at
1840 (N—O), 650 and 625 (Ni—N) cm−1. The force constants for the respective bands are 13.0 and 5.7 millidynes/Å.
On both the Fe and Ni surface, NO is partially decomposed to give N2(g) and O, i.e. 2NO(g) → N2(g) + 2Oads. An
absorption band due to nickel oxide formed by this process has also been observed and at this stage the surface
potential is found to be negative, due to adsorbed oxygen. Adsorption of more NO imparts positive potential to
+
the surface which may be due to the presence of NOads . At this stage the reaction NO(g) + NOads → N2O(g) + Oads,
further complicates the interpretation of the spectra. Neither hydrogen nor oxygen at room temperature reacts
with adsorbed species.
CO2 chemisorbed on alumina at room temperature exhibits strong bands at 1770, 1640, 1480 and 1232 cm−1.
The physically adsorbed CO2 on silica is observed at 2350 and 1800−1870 cm−1.
The normal and combination modes of vibration of C2H2 are v1[vsym(C—H), R, 3374 cm−1], v2 [v(C ≡ C), R,
1974 cm−1], v3[vasym(C—H), R, IR, 3287 cm−1], ν4[δa(C—H), R, 612 cm−1], v5[δd(C—H), IR, 729 cm−1] and v4 +
ν5 (1341 cm−1). Adsorption of C2H2 on alumina gives a spectrum in which two species can be identified. One spe-
cies can be completely removed from the surface by a 5-second evacuation at room temperature, is observed only
at pressures above 0.5 cm and is characterised by bands at 3220 and 1950 cm−1. The former band is attributed to
the (v3) C−H stretching vibration and the latter to the (v2) C ≡ C stretching mode. The strongly adsorbed acetylene
is stable up to 300°C and shows a ≡ C—H stretching mode at 3300 cm−1 and a —C ≡ C— stretching vibration at
2007 cm−1. The v4 + v5 combination band does not appear in the spectrum. The value of the v3 vibration (3300 cm−1)
lies between the value for gaseous C2H2 (3287 cm−1) and methylacetylene (3334 cm−1, gas; 3305 cm−1, liquid). This
in conjunction with the asymmetry implied by the presence of —C ≡ C— vibration and the lack of v4 + v5 combina-
tion band suggests that C2H2 is adsorbed in an end-to end position: Xs….H—C ≡ C—H or Xs—C ≡ C—H. Further,
the addition of C2D2 to a hydroxylated surface shows that the intensity of the OH band at the highest frequency
diminished and a new single band at 2785 cm−1 due to OD appears. This exchange may be represented as
Xs—O—H + Xs... D−C ≡ C−D → Xs—O—D + Xs….H−C ≡ C−D
The surface O—D groups can be removed by evacuation at 900°C. This leads to infer that the sites responsible
for C2H2 adsorption are not hydroxyl groups.
The vibrational spectra of molecules undergoing rotational isomerism and keto-enol tautomerism are most
likely to be affected on adsorption, because the free rotation of molecules is restricted upon adsorption. The cis/
trans ratio as determined from spectral data on dichloroethane (C2H4Cl2) adsorbed on silica gel is 1.9 as com-
pared to 1 in chloroform solution and 1.4 in the neat liquid. The keto/enol ratio in acetylacetone adsorbed on
silica gel has been found to be of the order of 8.6. This ratio in chloroform solution and for pure liquid is 1 and
2.4 respectively.
The rotational motion of surface functional groups can also be studied. In the IR adsorption studies of H2O/
D2O on silica surface, P (very weak), Q (strong), R (very weak) branches of surface silanol, i.e. Si OH/OD has
been observed. The separation of the band maxima of P- and R-branches are 200/140 cm−1. The relation between
the separation of the P- and R-branches, i.e. Δν due to Peri and Herzberg is given by the expression:
8κTB
Δν = = 2.358 TB cm −1 ,
hc
where the terms and symbols have their usual meanings.
If the Si − O − H bond angle is q and if f is the angle between the axis of rotation and the O − H axis, then
B μm r 2 2
(ϕ − 90°)
Thus if rOH = 0.958 (from water), the angle j can be calculated. The true Si − O − H bond angle q is slightly
less than j and is calculated to be 113°.
In view of the foregoing examples on the adsorption of different type of chemical species on different adsorbing
surfaces, it is concluded that the spectra of adsorbed species is pressure, temperature and nature of the adsorb-
ing surface dependent.
(d) Study of Charge-Transfer Complexes Molecular complex of charge-transfer type is formed between
an electron donor with low ionisation potential, i.e. an electron-rich molecule and an electron acceptor with high
electron affinity, i.e. an electron-deficient molecule. Charge transfer occurs primarily in the excited state and is
stabilised relative to the ground state by electrostatic interaction between the separated charges on the donor and
acceptor whereas in H-bonded adducts, the electrostatic contribution is predominantly to the ground state of the
complex. When the infrared spectra of charge transfer complexes are compared with those of isolated molecules
which form the complexes, four types of changes may occur:
(i) The vibrational frequencies in donor or acceptor (or both) may be shifted. The spectra of complexes clearly
resemble the summation of spectra of both the donor and acceptor portions if there is weak interaction between
the components. Some of the differences which are observed in the solid state may be because of crystal pack-
ing. Sometimes, the vibrational frequencies are shifted as much as 150 cm−1. In such a case, the initial bond-
ing within the constituents is disturbed by complex formation, as to be expected, but only enough to change
a certain fraction of bond order. The decrease in energy of a particular bond also provides information about
the particular site of a charge transfer interaction, for instance, the interaction of sulphur dioxide with N,N-
dimethylaniline is through the nitrogen atom since the C−N stretching frequency of free amine decreases on
complex formation. The decrease in C = O stretching energy of N, N-dimethylacetamide on complexation
with iodine is an evidence for the interaction of iodine with oxygen atom of the amide. The band shift on
complex formation may be fixed as a criterion for the classification of donor-acceptor complexes, e.g. the
infrared spectra of the solid complex of o-phenylenediamine with 7, 7, 8, 8−tetracyano-quinodimethane with
molecular composition 1 : 1 are similar to the spectra of components and hence the complexes are character-
ised as of nonbonded type. However, in the spectra of 2:1 composition, the C = N and the ring C = C stretching
bands get shifted with variation in intensity and the complexes are classified as of ionic type. The general lack
of change in the infrared spectrum on complex formation is used to distinguish between weak charge transfer
complex and the products of electron transfer or proton transfer reactions. In case of complete electron trans-
fer from donor to the acceptor in the ground state, the spectrum corresponds closely to the sum of the spectra
of the two ions. On the other hand, in case of proton transfer, a new band due to the formation of a protonated
group may also appear, e.g. when the infrared spectrum of an addition compound of 2,6−dimethylpyridine
with picric acid with molecular ratio 1 : 1 is recorded in KBr, the band due to OH group of picric acid at
3110 cm−1 is not observed. Instead, a new band at 3080 cm−1 characteristic for the —NH+ group is observed.
The expected range for the —NH+ group vibrational band is 2900−2500 cm−1. Thus, one can safely assume
that the bonding in the complex occurs essentially through the proton transfer and consequently electrostatic
attraction between positive and negative ions as shown in the following scheme.
−
OH O
O2N NO2 O2N NO2
+ +
CH3 N CH3 CH3 + CH3
N
NO2 NO2
NH
(ii) The intensities of the bands may be influenced considerably. The changes in energy and intensity of the
vibrational bands on charge transfer complexation are generally larger in complexes involving σ-donors and
σ-acceptors than in the p-p interactions.
(iii) New low-energy bands may appear due to the fundamental stretching vibrations of the intermolecular bonds
formed between the isolated molecules involved in the complexation, e.g. two bands at 145 and 184 cm−l in the
infrared spectra of iodine-trimethylamine solid complex in nujol mull have been assigned to intermolecular
vibration mode of N…I and intramolecular I—I stretching vibration respectively, N …. I distance in the complex
is 2.27 Å. The I—I bond distance in the free acceptor is 2.67 while 2.84 Å in the complex. The N …. I vibra-
tion modes for iodine-pyridine, iodine-3, methylpyridine, iodine-3, chloropyridine and iodine-3, bromopyridine
complexes in cyclohexane solution appear at 94, 89, 73 and 65 cm−1 respectively.
(iv) The infrared vibration may become active with decrease in energy, e.g. the infrared inactive stretching
vibrations of the homopolar diatomic halogen acceptors such as I2, Br2 or Cl2 become infrared active but
with decrease in energy on complexation with an electron donor, e.g. the Cl−Cl fundamental vibration
band of liquid 35C137C1 which appears at 541 cm−1 in the Raman spectrum has been observed at 513 cm−1
in the infrared spectrum of Cl35 CI37 in benzene. The Raman spectra of bromine and iodine gases exhibit
bands at 317 and 213 cm−1 respectively. However, in toluene solutions, the respective bands appear at 300
and 204 cm−1 in the infrared spectra.
(e) Study of Polymers Infrared spectroscopy has been employed to characterise chain structure of polymer
and has led to the explanations of the reactions of multifunctional monomers including rearrangement and isom-
erisation. End group, cross links, branches and many other structural characteristics are studied by this technique.
This method is also used as a tool to measure end groups, isomeric composition, stereoregularity, copolymer
composition, additives, fillers, plastisisers.
Infrared spectra of pure polymers show only a few bands, e.g. only five infrared bands are observed in spec-
trum of polythene. Two types of distinct relations have been observed between spectra of monomer and polymer:
(i) spectrum of a polymer that has a complicated monomer structure is similar to the spectrum of monomer,
e.g. cellulose; (ii) spectrum of a polymer that has a simple monomer structure is different from the spectrum of
monomer. Monomer spectrum in such a case is more complex than polymer spectrum, e.g. polythene.
Spectra of polymers could best be understood on the basis of coupled vibrators. Let us consider a chain of n
identical vibrators coupled together and having identical frequency. Then the single identical frequency is split
into n components given by
v m2 = v 02 + w c2 (1 + cos )
mπ
where θ = and m 1, 2, 3,.., m are the modified and v 0 is the unmodified
m frequencies and wc is the coupling
n +1
parameter. Thus one frequency should split into n frequencies closely spaced since n is always small. The charac-
teristic group frequency of a monomer is retained in the polymer, e.g. C—H frequency in polythene. When wc is
large, a wide range of frequencies can occur and gives rise to skeletal vibrations. Square of observed frequency
when plotted as a function of (1 + cos q) gives rise to straight line. When n becomes infinite, only one band is
active in infrared while all others are IR inactive, so one shifted frequency is observed, e.g. C = O in polyvinyl
formate is observed at 1740 cm−1 while at 1765 cm−1 in the monomer. Large shift in frequencies of some of the
modes of vibrations has been observed for large value of wc, e.g. methylene rocking mode in polythene moves
from 716 to 1107 cm−1. It is to be noted that structure frequency correlations are valid for few bands. When the
chemical repeat unit is complex, wc is small and the spectrum of polymer is very similar to monomer where no
characteristic frequency exists, the extrapolation is done from relationship in frequency shift and units in chain.
It is thus apparent that the identification of a polymer is done by comparison of an unknown spectrum. In case
of copolymers, it is the superposition of the spectra of homopolymers. Since functional-group frequencies depend
on environmental factors, a random copolymer may not have same spectrum as block copolymer or superimposed
homopolymers.
One can also determine quantitatively: (a) copolymer ratios, (b) functional groups, (c) monomer impurities,
(d) end groups, and (e) additives in polymers by this technique.
End-group determination is done by studying the polymer of different degree of polymerisation. The band that
shows maximum variation of intensity is the one arising from the number of end groups in the polymer. The struc-
ture of the end group determines the nature of polymerisation reaction. These are least influenced by the nature
of the rest of the polymer.
The analysis is done by dissolving the polymer or by using film-by-absorbance ratio method. Sometimes inter-
nal standard is added, and it may be disperse dye before casting the film. The visible absorption gives the thickness
and acts as an internal standard for the infrared region.
Let us now see how methyl groups of polyethylene are estimated. A band which is independent of crystallinity
is chosen, i.e. C—H band of CH3 group at 7.25m. But it is interfered by 7.31 and 7.39m bands which are compen-
sated by polymethylene wedge. Calibrate with n-hexadecane or other standards, compensate by polymethylene
and determine absorption coefficient (specific absorbance)
A
ε=
d ×t
where d is the density in g/mL and t is the thickness.
A graph is then plotted between specific absorbance and number of methyl groups per 1000 carbon atoms. Now
place the analytical film and measure the number of CH3 groups per 1000 carbon atoms.
Isomerisation may give rise to (i) new bands, (ii) shifting of frequencies, and (iii) band broadening. Any of
these can be used to determine composition of stereoisomers.
In PVA, the bands at 916 and 1141 cm−1 are characteristic of syndiotactic structure. In PVC a ratio of
635–692 cm−1 bands has been found to be proportional to isotactic syndiotactic ratio. The crystallinity in polymers
induces some changes in the infrared spectra, viz., (i) some new bands may appear, (ii) bands may split into two
with same polarisation, (iii) split bands may have mutually perpendicular dichroism. The last alternative is usually
seen in bands from helical conformation. Double helix exhibits three bands for a single bond. In polystyrene, the
1070 cm−1 band is split into two components with same polarisation on crystallisation.
The film of nylon-6 exhibits variation of crystallinity on various heat treatments. Crystallisation is followed by
growth of bands at 935, 970, 1030 cm−1 whereas the amorphous band at 990 cm−1 diminishes on heat treatment.
One can deduce extinction coefficient if the material is highly crystalline or a single crystal.
Polarised infrared radiation is also helpful in the study of polymer spectra. When an amorphous solid is exposed
to unpolarised infrared radiation, the probability of radiation interaction with all the possible modes of vibration
is equal. When a beam of plane polarised radiation is passed through an oriented film, the absorption band whose
dipole moment change is parallel to the vibrating electric vector of radiation will be absorbed most and if it is
perpendicular, the absorption will be minimum. The dichroic ratio given by As /Ad may be correlated with the
orientation function or degree of orientation of a given sample.
Stabilisers and fillers can be separated from resinous material after soxhlet extraction, by dissolving in a suitable
solvent, say tetrachloroethane, and washing with tetrahydropyran. The solid fillers, etc., are separated by centrifuga-
tion. Only inorganic stabilisers and fillers will separate. The usual stabilisers are lead carbonate, sulphate, phthalate,
etc. The bands recommended for estimation are at 1410 cm−1 for lead carbonate, 877 cm−1 for calcium carbonate,
741 cm−1 for antimony oxide and 1535 cm−1 for phthalate. It is necessary to first calibrate for a given component.
(f) Study of Bridged Inorganic Complexes The infrared spectra aid in providing information about the
presence of bridge in inorganic molecules. There are three types of partial assignments concerned with the bridge
frequencies.
(i) ν asym
s is about 50−100 cm−1 greater than ν sym , e.g.
System ν asym
s ν sym d
P2 O74− 940 909 201
H2S2O7 806 732 —

S2O5Cl2 773 716 486
P2O3Cl4 806 713 —
(ii) ν asym
s is about 200 cm−1 greater than ν sym and the bending frequency is 1/3 to 1/2 of ν sym , e.g.
System vasym vsym d
Mn2P2O7 964 743 −
978 727 −
P2 O74- (calc)
K2 Cr2O7 797, 764 566, 556 220
Na4As2O7 735 550 245
Cl2O7 695 495 −
(iii) ν asym
s is about twice as great as ν sym which in turn is about twice as great as the bending frequency, e.g.
System ν asym
s ν sym d
Si 2 O6−
7 1029 503 169
H2S2O7 809 328 155

Ge−O−Ge 880–830 458 −
S2O5 Cl2 760 298 147

Cl2O7 695 280 −
P2O3Cl4 962 425 159

3.16.2 Quantitative Applications

The quantitative applications of IR spectroscopy are based mainly on the exploitation of frequency, intensity
and width of the band of the vibrational mode to determine the quantities such as concentration, thermodynamic
functions, solvation number, bond strengths and distances, etc. In the present part, the theory of methodology is
discussed prior to its application for estimating the quantities. Some of the quantitative aspects of IR spectroscopy
have been discussed here in two sections: (a) problems based on intensity data; and (b) problems based on vibra-
tional frequency and band width data.
(a) Problems Based on Intensity Data The earlier spectroscopic studies were mainly concerned with the
use of vibrational frequency bands but these days intensity of vibrational bands has become a very useful param-
eter in spectroscopic studies. It gives fundamental information regarding the nature of bonds between the atoms
since absolute intensity is related to the change in the dipole moment of the bond between, the two atoms during
the vibrating process.
(i) Absorption Intensity The intrinsic absorption corresponding to a transition from the ground state, m to an
excited state, n may be expressed in terms of transition probability coefficient, Bmn by the expression
ΔI = –hc Wmn Bmn lmn Nm ΔX (3.137)
where clmn = I gives the intensity of the incident beam. ΔI is the amount absorbed in traversing a layer of thickness ΔX,
Nm is the number of molecules per unit volume in state, m and the population, Nn is considered negligibly small.
The Einstein coefficient, Bmn is related with the matrix element, Rmn of the electric moment, MQ between the
states m and n. We have
R mn ∫
ψ *M Q ψ m d τ
n
(3.138)
and then Bmn is given by

8π 3 2
B mn = 2 R mn
(3.139)
3ch
The matrix element Rmn is also called the transition moment between the given states m and n. For an allowed
transition Rmn is nonzero.
Formula (3.138) may be used to determine the interaction of electromagnetic wave with (a) electric dipole
moment, (b) magnetic dipole moment, (c) quadrupole or higher moments, (d) induced dipole moments (in case
of Raman scattering), etc., by substituting the appropriate moment in place of MQ. In particular, Rmn turns out to
be zero for all cases where the product of the two wave functions involved does not have the symmetry of one of
the components of MQ.
In general, the instantaneous dipole moment, MQ for an oscillating molecule may be expressed by a Maclaurin
series about Q = 0.
⎛ dM ⎞ 1 ⎛ d 2M ⎞
M Q ( M )Q = 0 + ⎜ Q + Q 2 + .... (3.140)
⎝ dQ ⎟⎠ Q = 0 2! ⎜⎝ d 2Q ⎟⎠ Q = 0
As has already been stated that the first term in the expression gives the permanent dipole moment which is
responsible only for the pure rotational infrared spectrum, the second term is responsible for the appearance of
vibrational infrared spectrum while the quadratic and higher terms are effective in providing intensity to overtones
and combination bands. When MQ given by Eq. (3.140) is substituted in Eq. (3.138), the integral corresponding
to the permanent dipole moment (M)Q = 0 becomes zero in view of the orthogonality of functions and Rmn comes
out to be
1 ⎛ dM ⎞
R mn = ⎜ ⎟ (3.141)
2α ⎝ dQ ⎠ Q = 0
with the well-known relations
2π ( μ )1/ 2
α= (3.142)
h
1
and ( k / μ m )1/ 2 W mn (3.143)
2π c
The extinction coefficient A′ is given by
1 ΔI
A′ = − = hW mn B mn N m (3.144)
ΔX I
and using relations (3.137), (3.139) and (3.144), we get

2
π N ⎛ dM ⎞
A′ = 2 ⎜ (3.145)
3c μ m ⎝ dQ ⎟⎠
where Nm is put equal to N as a close approximation. I It
Sample
• Relation with Experimental Measurements In actual From
Detector
experiments, if L cm path of the solution containing C g moles source
I I0
of absorbing sample per liter transmits intensity, It, out of the Reference
incident intensity, I0 (Fig. 3.34), we get Fig. 3.34 Intensity relations in a double beam spectrom-
1 I CN A eter. I, the intensity from the source, /0 is passed by the
A′ = ln 0 and N = (3.146) reference, It by the sample solution.
LC I t 1000
where NA is Avogadro’s number. Substitution in Eq. (3.145) gives the molar extinction coefficient
2
1 I πC N A ⎛ dM ⎞
A′ = ln 0 = 2 (3.147)
LC I t 3c μ m 1000 ⎜⎝ dQ ⎟⎠
The relation (3.147) is one of the forms of Beer–Lambert–Bouger equation.
I
T = t = e − εCCl
I0
Recall from Chapter 1 that, although the transmittance is neither directly nor inversely proportional to concen-
tration, ln (1/T) is related to concentration as shown by the more familiar version of exponential relation
I ⎛ 1⎞
A = ln 0 = ln = kC
It ⎝T ⎠
where k = el and A is the absorbance.
Absorbance is the greatest where transmittance is least. The Beer–Lambert–Bouger equation predicts that if
the concentration of the absorbing species is varied, the transmittance and absorbance will vary as shown by the
dotted and dark lines in Fig. 3.35 respectively. Routinely it is easier to work with straight line behaviour of A when
making quantitative determination from spectra.
At high concentration of solute, deviations from Beer’s law have been
Percent radiation
noticed. These can be removed by calibration with a series of solutions of Transmission

known concentration. The plot of absorbance versus concentration may not
only be concave downward (negative deviation) or upward (positive devia- Absorbance
tion), it may have intercept on the axis of zero concentration, at a finite value,
corresponding to background absorption. This quantity can be found by
extrapolation and subtracted from all values of apparent absorbance, if a true Concentration
value for the absorptivity is desired. Fig. 3.35 The behaviour of transmittance
Further, it is to be noted that actually the transition mn is not a sharp and absorbance as concentration of the
transition. Apart from natural width in gases, the absorption band spreads absorbing species is varied.
due to Doppler and pressure broadening. In liquids various other physiochemical effects make the bands very
broad. Added to these is the effect of finite slit width of the spectrometer. Thus, extinction coefficient values have
not proved very useful, since these depend very much on the instrumental factors. Hence, in using the relation
(3.147) to deduce (dM/dQ), one must obtain integrated absorption coefficient A′ or Iabs over the complete band
corresponding to the given transition mn. If Av, is the molecular extinction coefficient of frequency in an absorp-
tion band, we define A or Iabs by
1 ⎛I ⎞
A ∫
entire band
Av d = ∫ ln ⎜ 0 ⎟ dv
Cl entire band ⎝ I t ⎠
(3.148)
In practice, Eq. (3.148) is integrated between two arbitrary values v1 and v2 on either side of v0. If the curve is
symmetrical,
1
v0 (v 1 + v 2 )
2
Consequently,
v
1 2
ln (I 0 / I t ) dv
Cl v∫1
A= (3.149)
Experimentally, a graph between An against n is plotted. Subsequently, either the squares on the graph paper are
counted or the paper profile is weighed or the band areas are evaluated by using a planimeter to obtain the quan-
tity A. Its unit is cm mole−1 or cm2/moles. Some data on the integrated absorption intensities of series of com-
pounds that could not be distinguished on the basis of their frequencies are listed in Table 3.9.
Table 3.9 Absolute absorption Intensities for several group vibrations.
Group Position of Bands Integral Absorption

(cm−1) (cm2/ mole-s ¥ 107)
C=O 1720
Esters 13
Alkylketones 8
Ketosteroids 9-20
C—O 1200
Esters 15
Alkylketones 2.5
Acetoxy steroids 21
N—H 3400
Dialkylamines 0.05
Alkylanilines 1.7
Diarylamines 2.2
Pyrroles, carbazoles 5.4
Indoles 6.6
C≡N 2250
Alkylcyanides 0.25
CH3 2900
Hydrocarbons 4.4
—COCH3 0.74
—COOCH3 2.54
CH3 1460
Hydrocarbons 0.54
—COCH3 1.3
—COOCH3 2.0
CH2 2900
Hydrocarbons 3.8
—COCH2 0.5
CH2 1460
Hydrocarbons 0.23
—COCH2— 1.17
—COOCH2 0.65
It is quite evident from Table 3.9 that there is a wide variation in absolute intensities of the same group present
in different molecules. This variation is due to the effect of structural environments of the molecule on the group.
Hence, absolute intensity can be considered an additional parameter for the identification of molecules. There-
fore, in times to come, the characteristic intensity tables similar to group frequencies tables will be available for
identification of molecules.
• Effect of Various Factors on Intensity The intensity of spectral band depends mainly on: (i) transition
probability, and (ii) the difference in the population in states m and n, provided the density of radiation of fre-
quency is kept constant.
Theoretically, intensity is directly related to change in dipole moment of the bond between the two atoms
during the vibratory process. The absolute intensity of the absorption band is given by Iabs = a(dM/dr)2, where a is
a constant. Therefore, for a more certain and better characterisation of an absorption band, in addition to its posi-
tion, the bandwidth and intensity should also be recorded. The measurement of intensities in the solid phase is not
reliable due to scattering or reflection losses. Intensities of solid samples are mostly restricted to the determination
of emax which is markedly dependent on the particle size. In spite of these limitations and anomalies in solid state
spectra, quantitative analysis of components in mixtures can still be carried out.
Infrared band intensities in gas phase will be different from those in the liquid phase because of perturbations
due to neighbouring molecules in the latter case. The absolute intensity of a group vibration can be determined
only in gas phase but it is more convenient to measure intensities in solution. The difference between the band
intensity in solution and in gas phase may be ascribed to be due to several factors. In solution phase, specific inter-
action between solute and the solvent molecules like hydrogen bonding and dipole-dipole interactions will occur
if solvents are polar. A solute molecule resides in a cavity associated with a solute dielectric constant which in turn
is surrounded by the solvent medium of a particular dielectric constant. The packing of solvent molecules around
the solute molecules varies with the molecular shape of the solute. In addition to these effects, several other effects
of smaller magnitude like short range intermolecular forces, anisotropic character of polarisabilities and shapes of
solvent molecules also contribute to the change in band intensity in solution as compared to the gas phase.
The intensity of the band is also affected by temperature. As the temperature increases, the population of rota-
tional energy level changes. The population of vibrationally excited levels is very small at room temperature (say
around 25°C) and intensity is not significantly affected by the redistribution among rotational levels. Temperature
is not an important factor in gas phase measurements. In the case of solutions, however, the absorption at the band
maximum decreases with increase in temperature, accompanied by the increase in bandwidth. This is due to the
change in band shape resulting from an increased rate of collision of the solute molecules with the walls of the sol-
vent cage. Integrated band intensity also seems to decrease with increase in temperature. An alternate explanation
of intensity variation with temperature has been offered in terms of the coupling of vibrational modes in the liquid
state. It may be concluded that wide variations in temperature must be avoided in band intensity measurements.
(ii) Concentration Determination The most important advantage of IR quantitative analysis is its specificity.
This enables one to perform the analysis with practically very little pretreatment of the sample. This technique
is particularly useful for the estimation of chemical species which are closely related structurally, e.g. o, m and
p-cresol and isomers of similar type which is not possible in ultraviolet/visible because of close similarity of the
spectra. Secondly, since there are many bands characteristic of a particular isomer, one can choose proper band
which is free from the interference and is capable of maximum sensitivity. The basic theory of quantitative analy-
sis is based on Lambert–Beer–Bouger law.
The absorbance in IR spectroscopy is measured from the 100
band under consideration by the baseline method. In this
Baseline
method a tangent, i.e. baseline is drawn from the shoulders
on either side of the peak as shown for a hypothetical band

in Fig 3.36. Baseline should be chosen carefully because a
second component absorbing near the analytical wavelength
may raise or lower the baseline if its concentration also varies. I0
In this situation, certainty of results could be confirmed by (a) (b)
analysing several absorption bands in the same spectrum. I t
The consistency in results would prove the validity of the

method. λ
Further, Fig. 3.36 reveals that at the band maximum Fig. 3.36 Hypothetical infrared absorption spectra (a) illus-
the difference between the levels I0 and It, where these are trating the use of the baseline measurement technique; and
the percentage transmissions, will give the contribution (b) showing uncertainty as to where to draw the baseline.
of the solute to the absorption. Since A = log(I0/It), one can determine the concentration of chemical species
either by the direct application of Beer’s law in case it is followed or by plotting working curves. In addition to
Lambert–Beer–Bouger law, the other methods employed for concentration determination are explained here:
• Slope Ratio Method or Method of Variable Path Length This method eliminates the effect due to other
interfering absorption or the correctness of I0. The equation of Beer’s law log (I0/It) = eCl can be expressed in the
form
I
log 0 = εC
Cl + K (3.150)
It
where K is absorbance due to cell and other absorbers in the path. Equation (3.150) can also be written as
log It = −eCl − K + log I0 (3.151)
Thus, if we measure the transmitted intensity for different path lengths, and plot log It versus l, a straight line
results with slope = −eC and intercept = log I0 − K. The experiment is then repeated on sample solution for similar
value of path lengths. Then again we will have
log It = −elC′−K + log I0 (3.152)
This will give a straight line with slope = −e C′ and intercept = log I0 − K, Therefore, by simply taking the ratio of
slope, i.e. eC/eC ′ = m1/m2 we can get the concentration of C ′ = C × m2/ m1 of the unknown sample.
• Absorbance Ratio Method This method is based on the use of some internal standard and is used for solid
samples in KBr disc. In this method, the ratio of absorbances, i.e. A1/A2 is plotted against concentration. This tech-
nique is used for salts of organic acids, e.g. sodium valerate, disodiumadipate, etc. One can also use absorbance
ratio method in case of films or pellets. Suppose we have mixtures of two compounds x and y. Since the thickness
is the same then,
A εC C
R= x = x x =m x (3.153)
A y εy Cy Cy
A standard curve is first drawn for synthetic mixtures of R against Cx/Cy and then unknown is determined from
that. This method finds wide applications in copolymer analysis. Some quantitative analysis has also been done in
aqueous medium. BaF2 windows are used and work is restricted to 6.5 to 9m region, since other region is strongly
absorbed by water. This is useful in biological systems and especially direct analysis of water phase of an organic
water two layers system.
The additivity of absorbance is important in multiple analysis, the simultaneous determination of two or more
absorbing species in the same solution. We can write
A i Ai = l i i Ci
The relation can be understood by reference to Fig. 3.37 for two components analysis. Curves 1 and 2 are the
absorption spectra of the pure components. Curve 3 is the spectrum of a mixture. The maxima in the mixture
do not exactly match with the respective maxima of the pure components. It is assumed that no other substance
present absorbs in this region. It will be seen that both
substances contribute to absorption at both wavelengths
lu and lv. Let C1 and C2 be the concentrations of the two
components in the mixture and let A1v represents absor-
bance of substance 1 at wavelength lv, etc. Then the A1u
absorbance of Curve 3 at lu will be given by,
A2v
A3u = A1u + A2u = elu lC1 + e2u lC2
and at lv by 3
2
A3v = A1v + A2v = elv lC1 + e2ν lC2 A2u
A1v
Since A3u and A3v can be determined from the experi-
mental observations on the mixture and es are obtained in 1
advance from the formula A = e Cl, then the above equa-
lu lv
tions can be solved simultaneously for C1 and C2. For
accurate results, e1v and e2u should be as low as possible Fig. 3.37 Hypothetical figure exhibiting two-component analysis.
while elu and e2v should be high. When e2u and e1v are negligibly small as compared to elu and e2v respectively, the
above expressions may be written as
A3u = elu lC1
and A3v = e2v lC2
However, in spite of all these facts, the quantitative aspects of IR spectroscopy in relation to concentration
determination of chemical species have diminished in importance partly because of their inherent difficulties
and largely due to the emergence of quicker and convenient methods of quantisation like gas chromatography for
volatile substances, etc. The role of IR spectroscopy after quantitative separation by gas chromatography is only
to identify each component qualitatively.
Problem 3.27: The following observations have been reported for the apparent absorbance of cyclohexanone
in cyclohexane solution at 5.83m.
Concentration (g/lit.) Absorbance
5.00 0.190
10.00 0.244
15.00 0.293
20.00 0.345
25.00 0.390
30.00 0.444
35.00 0.487
40.00 0.532
45.00 0.562
50.00 0.585
Plot these data: (a) Over what range is Beer’s law followed after necessary background correction is made?
(b) Compute the value of e, the molar absorptivity.
Solution The plot between apparent absorbance and concentration is shown in Fig. 3.38. From Fig. 3.38, the
background correction is found to be of the order of 0.14. After applying the background correction, the real
values of absorbance at different concentrations are given in the following tabular form.
Concentration g/lit. True Absorbance
5.00 0.050
10.00 0.104
15.00 0.153
20.00 0.205
25.00 0.250
30.00 0.304
35.00 0.347
40.00 0.392
45.00 0.422
50.00 0.445
0.600
0.550
0.500
0.450 0.500
0.450
0.400
Apparent Absorbance
Apparent Absorbance
0.400
0.350 0.350
0.300 0.300
0.250 0.250
0.200 0−39.0 g/lit.
0.200
0.150
0.150 0.100
0.100 0.050
0.050 0
10 20 30 40 50
0 Concentration (g/lit.)
0 5 10 15 20 25 30 35 40 45 45
Concentration (g/lit) Fig. 3.39 The behaviour of true absorbance as
the concentration of cyclohexanone is varied.
Fig. 3.38 The behaviour of apparent absorbance as the concentration
of cyclohexanone is varied.
The graph based on concentration and true absorbance is presented in Fig. 3.39 from which it is evident that
Beer’s law holds good in the concentration range 0−39.0 g/lit beyond which negative deviation is observed.
The value of e, the molar absorptivity is then determined from the relation
AM
ε= .
Cb
From A = 0.05, C = 5g/lit. and b = 0.096 mm. we have
0 05
ε= = 1.02 × 102 lit mole−1 cm−1
⎛ 5⎞
0.0096(cm) ⎜ ⎟ / lit
⎝ 98 ⎠
Problem 3.28: A sample of ethyl bromide suspected of containing trace amounts of water, ethanol and ben-
zene was examined in an infrared spectrophotometer. The following relative absorbances were obtained by a
baseline method: at 2.65m, A = 0.110, at 2.75 m, A = 0.220, at 14.7 m, A = 0.008. Calculate the amounts of water
and ethanol in parts per million and of benzene in 100 mL.
Name of Compound A Conc. (%), Length (mm) Wavelength (m)

Water 5.240 100,3.0 2.65
Ethyl alcohol 2.670 100,3.0 2.75
Benzene 0.830 100,0.1 14.70
Solution In order to determine the concentration of benzenc, water and ethyl alcohol in a given sample of ethyl
bromide, we must know the molar absorptivity of the respective impurities in ethyl bromide. Thus, using the refer-
ence data and the formula
AM
ε= .
bC
A 5.24
we will have ε H 2O = = = 17.46 mole −11cm 1
b (C /M ) 0.. (cm) × 1( mole )

Similarly, for C2H5OH and C6H6, we get
εC 2 H 5OH = 8.90 mole −11 cm 1 and εC6 H 6 = 83.00 mole −11 cm 1

Now by using values of absorptivities, the concentration of H2O, C2H5OH and benzene are evaluated as fol-
lows. We know that
AM
C= .
b
0.110 × 18
Therefore, C H 2O = g/lit = 37.8 10 −2 g/lit
0.3 × 17.46
Hence, the amount of water in the sample of ethyl bromide
= 37.8 × 10−2 × 103 mg/lit = 378 ppm
0.220
Similarly, C C 2 H 5OH = × 46 g /lit
/ 378.99 × 10 −2 g / lit
8.9 × 0.3
Thus, the amount the ethyl alcohol in ethyl bromide
= 378.99 × 10−2 × 103 mg/lit = 3790 ppm

Similarly,
0.008 × 78
CC6 H 6 = g/ = 7.518 × 10 −1 g / = 0.7518 g /lit
0 01 83
= 0.7518×103 mg/lit
= 751.8 mg/lit
Therefore, the amount of benzene in ethyl bromide
751.8 × 100
= = 75.18 g / 100 mL
1000
(iii) Detection of a Vibrational Band The vibrational bands due to different groups present in the molecule
will be recorded by the spectrometer up to a value of It /I0 = 0.99. However, at these high values of It /I0, i.e. close to
0.99, the sensitivity of the instrument is much reduced and weaker bands might not be observed in the spectrum.
Thus, simply from the value of It /I0, one can infer whether a band due to a particular vibrational mode would be
observed in the spectrum or not.
Problem 3.29: Carbon monoxide gas (effective area 17Å2/CO molecule) was adsorbed on a silica gel disc of
0.01 cm thickness and 1 cm2 area. If the internal surface area and bulk density of silica is 100 m2 g−1 and 2 gcm−3
respectively and if the extinction coefficient of a C O vibration is 3.3×105 cm2 mole−1, show that the absorption
band due to C O absorption for 0.1 monolayer coverage on silica gel can easily be resolved by an IR spectrom-
eter. Each gram of porous adsorbent has ≈100 × 1020 Å2 surface available for adsorption.
Solution 1 g of silica gel has 100 × 1020 Å2 of the surface available for adsorption. Since the effective area for
CO molecule is 17 Å2, it follows that 100 × 1020 Å2/17Å2 = 6 × 1020 CO molecules can be adsorbed on 1 g of silica.
Each gram of porous adsorbent has ≈100 × 1020 Å2 surface available for adsorbent.
Now, 6 × 1023 CO molecules (Avogadro’s number) occupy 22400 mLs at NTP.
Therefore, 6 × 1020 CO molecules occupy (22400/6 × 1023) × 6 × 1020 = 22.4 mLs at NTP or 2.24 mLs for
0.1 monolayer coverage.
But 2.24 mLs of CO molecules at NTP contains
6 10 23 2.24
= 6 1019 CO molecules.
22400
Since l = 0.01 cm.
Therefore, the number of molecules of CO absorbing/unit area of the IR beam = 6 × 1019 × 0.01 = 6 × 1017, also
ε per CO molecule is,
3 3 × 105
ε= = 0 55 × 10 −18 cm2/CO group
6 × 10 23
Therefore, εCl = 6 × 1017 × 0.55 × 10 −18 = 0.33
Since I t /I 0 e − ε CCl
I t − εC
Cl −0.33
Therefore, log10 = = = − 0.1432 = 1.8568
I 0 2.303 2.303
or transmittance = It /I0 = 0.72 which can easily be resolved by an IR spectrometer. It is well-known that carbonyl
groups are very strongly absorbing and easily detected in the spectrum.
Problem 3.30: Ethylene was adsorbed on a silica gel disc of thickness 0.01 cm and 1 cm2 in area. If for —CH2
group the absorptivity is of the order of 4000 cm2/g, show that the absorption band due to —CH2 group for 0.1
monolayer coverage on silica gel can be recognised by an IR spectrometer.
Solution The extinction coefficient of a —CH2 vibration can be written,

4 × 103 × 14
ε= = cm 2/CH 2 group
6 × 10 23
= 9.333 × 10−20 cm2/CH2 group
Since l = 0.01 cm.
Therefore, as in Problem 3.29, the number of CH2 groups/cm2 of the infrared beam = 6 × 1019 × 0.01 = 6 × 1017
for a monolayer coverage of 0.1.
εCl = 9.333 × 10 −20 × 6 × 10 −17 = 0.0559
I t −0.0559
log10 = = − 0.0242 = 1.9758
I0 2.303
It
or = 0 94
I0
Thus —CH2 group can be resolved by a spectrometer, but to a lesser extent than the carbonyl group. It is also
evident from the absolute absorption intensity data of these groups that the integral absorption of the CH2 group
is small than that of the C O group (Table 3.9).
(iv) Ratio of Z and E Isomers in a Given Molecule The IR spectra of mixture of conformations will contain
vibrational bands characteristic of these states with intensities proportional to the concentration of each of the
isomers present. It follows that by choosing two corresponding bands one belonging to each isomer
IZ Z CZ l (3.154)
IE E CEl (3.155)
where Is′ denote the band areas proportional to the band intensities of isomers Z and E, as′ represent the inte-
grated absorption coefficient and l is the cell path length. For all practical purposes we assume aZ = aE. = 1.
Therefore, from Eqs (3.154) and (3.155)
I Z CZ
= (3.156)
I E CE
Thus, it is evident from Eq. (3.156) that simply by measuring the areas under the bands corresponding to
Z and E isomers, one can find the percentage of the respective conformers.
Problem 3.31: The spectrum of secondary bands at 3425/3388 and 3417/3381 cm−1 were assigned to the N−H
stretching vibrations of Z and E conformers of N-isobutylthioformamide respectively. From the IR spectrum of
N-isobutyl-thioformamide in Fig. 3.40, determine the percentage of Z and E isomers.
Solution The ratio of areas under the vibrational bands at 3425 and 3417 cm−1 is 3.5:1 (Fig. 3.40).
100
I Z CZ
We know that =
I E CE 80
3381
C Z 3.5 C C E 3.5 + 1 100
Therefore = or Z = or C E = = 22 per cent 3388
CE 1 CE 1 45 60
and CZ = 100 − 22 = 78 per cent 40

3417
3425
(v) Evaluation of the Energy Difference between Isomeric States We know
20
from the preceding problem,
IA = aA CA l (3.154)
0
3500 3400 3300
IB = aB CB l (3.155)
n (cm−1)
where terms and symbols have their usual meanings. The equilibrium constant for Fig. 3.40 The spectrum of N-
the preceding problem, isobutyl-thioformamide in CCI4 in the
NH stretching region.
K (3.157)
can be expressed as
CB I BαA
K= = (3.158)
CA I A αB
But, according to Van’t Hoff isochore
− ΔH 0 ΔS 0
ln K = + (3.159)
RT R
Consequently,
⎛I α ⎞ ΔH 0 ΔS 0
ln ⎜ B A ⎟ = − +
⎝ I A αB ⎠ RT R
⎛ I ⎞ ΔH 0 ΔS 0 ⎛α ⎞
ln ⎜ A ⎟ = − + ln ⎜ A ⎟ (3.160)
⎝ I B ⎠ RT R ⎝ αB ⎠
Accordingly if the intensities of the corresponding bands at different temperatures are known, ΔH
Δ 0 the energy
difference between the isomeric states A and B can be determined from the slope of the plot ln ( IA/IB) against 1/T
This treatment assumes that ΔH0 , ΔS0 and ln (aA /aB) are independent of temperature which is usually assumed to
be valid for the process of rotational isomerism.
Problem 3.32: The thermal equilibrium of equatorial and axial conformations of methylcyclohexane by
means of infrared absorption spectra have been studied. The fundamental frequencies of the IR bands at 547
and 607 cm−1 are assigned to the equatorial and axial conformations. If the ratio of the respective band areas for
these two conformations at 25, 50 and 110 °C are 8.0, 5.0 and 4.25 respectively, calculate ΔH0 and ΔS0 for these
isomeric species.
Solution Now ln (IA /IB) at different temperatures are
ln(IA /IB) T, K 1/T × 103

2.0790 298.15 3.35
1.6090 323. 15 3.09
1.4469 383.15 2.61
The slope and intercept of the plot of the ratio of the axial and equatorial band areas for methylcyclohexane
against 1/T can be determined from Fig. 3.41.
2.00
Slope = 0.781; Intercept = − 0.875
But slope according to Eq. (3.160) is ΔH0/R 1.75
Therefore, ΔH0 = +1552 cal/mole 1.50

In K
ΔS 0 1.25
Also intercept =− = −0.875
R
1.00
So, ΔS0 = 1.738 cal/mole deg (it is being assumed that aB = aA = 1). 2.00 3.00 4.00 5.00
1 × 103
−1.25 T
(vi) Solvation Number Most of the surfactants are hardly soluble in nonpo-
lar solvents, slightly soluble in nonhydrogen-bonding polar solvents and highly −1.50
soluble in hydrogen-bonding polar solvents. The high solubility of surfactants
in hydrogen-bonding polar solvents has been attributed to the existence of −1.75
some intermolecular interactions between the solute and the solvent mole- −0.875
cules. Actually, the infrared spectrum of the solvent shows remarkable changes −2.00
in the fundamental bands of the solvent and bands of a particular stretching
Fig. 3.41 Variation of equilibrium con-
vibration corresponding to the free and the bonded states are observed sepa- stant with temperature.
rately. Since the ratio of the intensities of the bonded states to that of the free
state can be regarded as a measure of the strength of the solute–solvent interaction, quantitative analysis of the
absorption intensities can provide a valuable information regarding the process of dissolution of the surfactants.
It can also provide a convenient method to determine the number of solvent molecules that combine with one
molecule of the solute. In actual practice, sample solutions containing various amounts of the solute (surfactant)
are prepared by weighing the solute and the solvent in the sample flask and recording their infrared spectra. This
procedure is, however, modified for those surfactants that are insoluble in a given solvent. This is based on the
observation that a surfactant which is insoluble in a given solvent (A) can be rendered soluble in this solvent by
the addition of a small quantity of another hydrogen bonding polar solvent (B). In this case binary solutions of
A−B, in which the molecular concentrations of A and B are C A0 C B0 , are first made. The ratio of C A0 C B0 is
called a dilution ratio and for a given dilution ratio, various amounts of surfactants are added and the change in
intensity of a particular stretching band is observed. The procedure is repeated for various dilution ratios. The
molar concentrations of A, B and the surfactants (S) in the solution are determined from the observed weight
concentration and the density of the solution while the number of solvent molecules that combine with one mol-
ecule of the surfactant (S) can be calculated in the following manner.
The relative intensities in cm−1 of the free, If and bonded, Ib bands may be defined as
1 ⎛I ⎞
If = ∫ ln ⎜ 0 ⎟ d ( v) (3.161)
l band ⎝ I t ⎠
1 ⎛ I0 ⎞
and Ib = ∫ ln ⎜⎝ I t ⎟⎠ d ( v) (3.162)
l band
where l is the thickness of the sample, I0 the energy of the incident radiation, It the energy of the transmitted radia-
tion and v the frequency of the radiation.
Normally the relative intensities defined by Eqs (3.161) and (3.162) are first corrected for local field effect and
then reduced to the values at the molar concentration of the pure solvent C B0 by means of the expressions
I*f = fd. fc. If (3.163)
I*b = fd. fc. Ib (3.164)
9n D
where fd =
( n + 2) 2
2
D
C B0
fc =
CB
and nD = refractive index of the solution. The solvent in such a solution exists in bonded and free states so that the
molar concentration, C B0 of the solvent in solution can be expressed as
CB0 = Cb + Cf (3.165)
where Cb and Cf are the molar concentrations of the solvent in the bonded and free states respectively. The abso-
lute intensities τ of the bonded and free states are related to the corresponding I* by means of
C*b tb = I*b (3.166)
C*f tf = I*f (3.167)
where C*b = fcCb (3.168)
and C*f = fcCf (3.169)
I b* I f*
so that + = C b* + C f* = f (C
Cb Cf f c ⋅ C B0 (3.170)
τ b τf
which on rearrangement yields
⎡ τb ⎤ *
I b* ⎢ τ ⎥ I f + τ bC f c
0
(3.171)
⎣ f⎦
Once τ b and τ f are known, the relative intensity for the bonded state (Ib ) is calculated using Eq, (3.164) and
concentration Cb so that solvent in the bonded state is evaluated from expressions (3.164), (3.166) and (3.168), i.e.
Ib
C fd. (3.172)
τb
The solvation number is then calculated from
CB
n= (3.173)
CS
Problem 3.33: The relative intensities of the C-D stretching vibration for the chloroform-d-dodecyl-
dimethylamine oxide (C12AO) system and the relevant data are
C S C 12 A O nD d I *b I *f
(wt percent) (cm-1) (cm-1)
0 1.4470 1.480 0 1.032

0.2569 1.4471 1.478 0.186 1.279
1.0106 1.4472 1.472 0.629 1.262
1.4062 1.4473 1.469 0.886 1.252
1.7816 1.4474 1.465 1.064 1.216
2.8574 1.4476 1.457 1.709 1.209
2.8608 1.4476 1.457 1.647 1.189
5.2304 1.4481 1.438 2.943 1.120
6.4981 1.4486 1.428 3.610 1.068

8.7316 1.4486 1.411 5.029 1.010
Calculate the solvation number for C12AO at various concentrations of C12AO provided it is soluble in chloroform−d
°
and C CDCI 3
= 12.3 M.
Solution In, order to calculate τ b and τ f according to Eq. (3.171), I b* is first plotted against I f* for different
values of CS. Such a plot for this system is shown in Fig. 3.42 and it yields (I*f )I* = 1.30 cm−1.
b=0
Further, since the plot of I b* against I f* is linear, the slope of the line yields τ b / τ f . But
(I * )
f
I *b =0
C B0 τ f = C CD
0
3Cl (
M )τ f
13 cm −1
Therefore, τf = = 106 cm2/mole 4.0
12.3 × 10 −3 mole/cm3
I∗b (cm−1)
which when combined with τb/τf = 16 yields τb = 1696 cm2/mole.
Now for CS = 0.2569 wt per cent CCDCl3 = 99.7431 wt. per cent, I b* = 2.0
0.186 cm−1, and nD = 1.4471.
Consequently, from Eq. (3.164)
9 1.4471 12.3MI b 0
0.186 cm −1 = 1.0 1.1 1.2 1.3 1.4
⎡(1.4471) + 2⎤ ⎛ 997.431⎞ M
2
2
lf0(cm−1)
⎣ ⎦ ⎜⎝ 120.5 ⎟⎠ ∗
∗
Fig. 3.42 b versus I f plot for the chloroform-
d−C12 AO system.
or I b = 0.1584 cm−1
Further, from expression (3.172)
9 × 1.4471 0.1584 0.1584

Cb = 2
= 0.7773 × = 0.0726 × 10−3 mole/cm3
⎡(1.4471) + 2⎤
2 1696 1696
⎣ ⎦
and so from Eq. (3.173)

0.0726 × 10 −3
(
n for CS = wt per cent mole cm ) =
0.017 × 10 −3
= 4.2
Problem 3.34: The relative intensities of the C-D stretching vibration for the chloroform-d-cetyl trimethylam-
monium chloride (CTACl)-carbon-tetrachloride system have been studied. There relative intensity data at dilution
0 0
ratios CCCl4
/ CCDCl3
of 0 and 7.49 are given below.
CS × 103 nD d I*b I*f

0 0
C CCl 4 C CDCl 3 (mole/cm3) (cm−1) (cm−1)
0 0 1.4470 1.480 0 1.311
0.048 1.4480 1.465 0.428 1.278
0.096 1.4489 1.463 0.810 1.269
0.127 1.4493 1.458 1.070 1.255
0.157 1.4499 1.453 1.331 1.234
0.195 1.4507 1.446 1.594 1.223
0.283 1.4514 1.431 2.475 1.169
0.366 1.4529 1.416 3.126 1.141
7.49 0 1.4591 1.565 0 0.0769
0.010 1.4591 1.563 0.033 0.0760
(Continued)
C 0CCl 4 / C 0CDCl 3 CS × 103 I*b I*f

nD d
(mole/cm3) (cm−1) (cm−1)
0.016 1.4591 1.562 0.053 0.0758
0.025 1.4593 1.560 0.080 0.0746
0.028 1.4592 1.560 0.095 0.0748
Calculate the solvation number of CTACl at its various concentrations provided it is insoluble in CDCl3 but ren-
dered soluble in CDCl3 by the addition of CC14 and C°CDCl3 = 12.3M.
Solution We first consider the data for the dilution ratio 0.0. From the plot of I*b against I*f (Fig. 3.43) we get
0
I f* )I b =0 .311c 1
ut (I f* )I b 0 3
( = 12.3M)τ f , which yields
1.311 −11
τf = ( /mole / 3
)
12.3 × 10 −3 3.0
= 106.58 cm2/mole ∼107cm2/mole
Further, the slope of the I*b against I*f (Fig. 3.43) is −18 and so according 2.0
b (cm )
−1
to Eq. (3.171), − τb/τf = −18 which gives τb = 1930 cm2/mole.
°
Now for CS = 0.048 × 10−3 mole/cm3, I*b = 0.428 cm−1, = 12.3 M,
I∗
CDCl3
nD = 1.448. 1.0
Consequently, from Eq. (3.164)
9 1.448 12.3I b
0.428 cm −1 = 2
⎡(1.448)2 + 2⎤ 12.3 0
⎣ ⎦ 1.0 1.1 1.2 1.3 1.4
I∗(cm−1)
f
Ib = 0.55119 cm−1
Fig. 3.43 Ib* versus I*f plot for the
−1 Chloroform-d-CTACl-carbon tetrachloride
Ib 0.5512 cm
Further, C fd . = 0.7765 × system
τb 1930 cm 2 / mole
= 0.0002218 mole/cm3
and so
C b 0.2218 × 10 −3
n= = = 4 62
CS 0.048 × 10 −3
Similarly, for the dilution ratio 7.49, the relevant data for CS = 0.010 × 10−3 mole/cm3 are
* = 0.033 cm−1
b
nD = 1.4591
9 × 1.4591
fd = 2
= 0.7702
⎡(1.4591)2 + 2⎤
⎣ ⎦
C BO 12.3 M
fc = = =1 0
C B 12.3 M
0.033
Ib = = 0.04285 cm −1
0.7702
Similarly, Cb = fd⋅Ib/τb
0.7702 × 0.04285 ( 1
=
2709 ( 2 / mole)
= 0.0122 × 10−3 mole/cm3
C b 0.0122 × 10 −3 / cm3
and so, n= = = 1.2
CS 0.010 × 10 −3 / cm3
(vii) Equilibrium Constants and Energetics of Hydrogen Bonding Hydrogen bonding between a donar (A−H)
and an acceptor (B) may result in the formation of series of compounds with the general formula (A−H)m... Bn
where m and n are integral numbers. Once the equilibrium constant for the complexation reaction represented by
m (A H ) nB (A H )m Bn (3.174)
is known, free energy can easily be determined and enthalpy and entropy value can be obtained from the tempera-
ture variation of the equilibrium constants. We illustrate this for two different cases.
Case 1: When a Donor (A-H) and an Acceptor (B) form a 1 : 1 complex The complexation reaction is rep-
resented by
a ciation
asso
A H +B di
dissociation
i ti
A −H B (3.175)
The equilibrium constant for the complexation reaction (3.175) is

C A H ... B
K= (3.176)
CA HCB
where C A H ...B C A − H and C B represent the equilibrium concentration of the H-bonded donor, free donor and
acceptor respectively in solution. (It is assumed that the concentration of the proton donor is sufficiently low to
avoid self-association.). If C A H and CB denote the total concentration of the donor and the acceptor respectively
in the solution, then the material balance equation yields
C A0 H CA H + CA H ...B (3.177)
and C B0 B CA H ...B (3.178)
The equilibrium concentrations of the donor (A-H) and the H-bonded A−H….B complex in the solution can
easily be determined for a solution of a given C A0 H and C0B from a measurement of its absorbance at the frequen-
cies characteristic of either the donor (A−H) or of the donor and the complex A−H…. B species in the following
manner.
The absorptivity (extinction coefficient) of the proton donor (A−H) is first determined from measurement of
the absorbance A ν ( ) of a very dilute solution of the donor in an inert solvent at a frequency characteristic of the
donor. The absorbance of the solution at other concentrations of the solution is next measured at frequencies of
the donor and the equilibrium concentration C A H evaluated from
Av ( ) aA0 H C A H l (3.179)
where l is the cell length in cm. The Eq. (3.179) is the well-known Beer–Larnbert law. The equilibrium concentra-
tions C A H …B d C B at a particular total concentrations C0A H and C0B of the donor and the acceptor respectively
are then calculated from Eqs (3.177) and (3.178). Alternatively, the equilibrium concentration of the complex
C A H ...B ’ can also be independently calculated by first determining its absorptivity aA0 H ...B in the manner described
above for the donor for a solution that contains very large excess of the acceptor. Once aA0 H ...B is known, the absor-
bance of the solution at any concentration of the donor is next measured at wave numbers that are characteristic
of the complex. The equilibrium concentration of the complex is evaluated from Eq. (3.177) while that of CB from
Eq. (3.178). Equilibrium constant is then determined from the expression (3.176).
Case 2: Hydrogen Bonding due to Self-association When a solute, say, A, undergoes self-association due to
hydrogen bonding to give molecular species of the general formula An, the complexation reaction is represented by
nA An (3.180)
The equilibrium constant is given by
CAn
Kn = (3.181)
(C A ) n
We illustrate this for the case of a simple reaction denoted by

2A A2 (3.182)
for which the dimerisation constant is
CA2
Kd = (3.183)
(C A ) 2
where C A 2 and CA represent the equilibrium concentrations of the dimer and monomer respectively. Since A in
solution now exists both as A and A2, the material balance equation yields
0
A CA + 2 CA2 (3.184)
0
where C is the total concentration of A in the solution. If it is now assumed that the dimer does not contribute to
A
the absorbance at the frequency characteristic of the monomer and that the Beer law is valid for the monomer, then
employing the procedure as outlined for the Case (1) above, the equilibrium CA of the monomer is evaluated from
A ν(A ) aA0 C A l (3.185)
0
where An(A) is the absorbance of a solution at a frequency characteristic of the monomer, a is its absorptivity and
A
l the cell length in cm. This gives rise to
1 0 1 ⎛ 0 A v (A ) ⎞
CA2 (C A − C A ) CA − 0 ⎟ (3.186)
2 2⎝ aA l ⎠
Substitution of CA and C A 2 values in Eq. (3.183) yields

2
1 ⎡ 0 A v (A ) ⎤ ⎡ A v (A ) ⎤
Kd CA − 0 ⎥ ⎢ a0 l ⎥ (3.187)
2⎣ aA l ⎦ ⎣ A ⎦
which on rearrangement yields
⎡ C A0 ⎤ a A l
2
a0 l 2
A v (A ) = A ⎢ ⎥− (3.188)
2K d ⎣ A v ( A ) ⎦ 2K d
Thus, a plot of A ν ( A ) versus C A0 /A
A v ( A ) yields straight line of
2
a0 l 2
Slope (m) = A (3.189)
2K d
l
and intercept ( ) aA0 (3.190)
2K d
m
so that Kd = (3.191)
2b 2
(viii) Evaluation of ΔG0, ΔH0 and ΔS0 from Equilibrium Constants The free energy change, ΔG, for the
H-bonded systems is given by
ΔG G products − G reactants
C products
ΔG G p0
((G −G0 ) RT ln (3.192)
C reactants
0 0
where G products and G reactants denote the standard free energies while C products and C reactant represent the equilibrium
concentrations. Further, as these reactions are at equilibrium, ΔG = 0, so that Eq. (3.192) becomes
ΔG 0 RT l K (3.193)
where K C products / C reactants
Furthermore, for an isothermal (at constant temperature) process
ΔG 0 ΔH
ΔH 0 − T ΔS 0 (3.194)
where ΔH 0 ΣH
Σ H f0 ( ) − ΣH f0 (reactants) (3.195)
is the difference between the enthalpies of the products and the reactants, similarly
ΔS ΣSS 0 (
Σ ) − ΣS 0 ( reactants) (3.196)
Combination of expressions (3.193) and (3.194) yields
− ΔH 0 ΔS 0
ln K = + (3.197)
RT R
Thus, a plot of ln K against 1/T would give a straight line of
ΔH 0
Δ
Slope (m) = −
R
ΔS 0
and intercept ( ) =
R
Hence ΔH0 and ΔS0 can be evaluated.
Problem 3.35: The concentrations of free, non-hydrogen bonded pyrrole in pyrrole-pyridine-acyl mixtures
from measurements of the intensity of the v01(N−H) band at 3498 cm−1 have been determined. The total concentra-
tion of pyrrole (Table 3.10) presents the initial concentration C A0 H and C B0 of pyrrole and pyridine respectively and
the measured equilibrium concentration CAH of pyrrole (a) assuming that only 1:1 H-bonded complex is formed,
calculate the equilibrium concentration CB and C A H ...B of pyridine and of the complex respectively and the equi-
librium constant K for the formation of the 1:1 complex (b). Calculate ΔH 0 , ΔS ΔS 0 and ΔG 0 at 298.15 K.
Solution (a) Consider the data at 12°C.
We know that C A0 H CA H + CA H ...B C B0 = C B CA H ... B
Table 3.10
0
C AH C B0 C AH 0
C AH C B0 C AH
A. 47° B.40°
0.00667 0.0835 0.006 0.00675 0.0787 0.00583
0.00667 0.0125 0.00566 0.00675 0.0984 0.00552
0.00667 0.167 0.00547 0.00675 0.148 0.00506
0.00667 0.209 0.00512 0.00675 0.197 0.00485
0.00675 0.246 0.00440
C. 330 D.300
0.00645 0.0426 0.00593 0.00678 0.0813 0.00565
0.00645 0.0532 0.00575 0.00678 0.102 0.00546
0.00645 0.0798 0.00555 0.00678 0.153 0.00498
0.00645 0.133 0.00505 0.00678 0.203 0.00467
0.00678 0.254 0.00424
E.20° F.12°
0.00672 0.0813 0.00554 0.00677 0.0632 0.00565
0.00672 0.102 0.00527 0.00677 0.0790 0.00536
0.00672 0.153 0.00477 0.00677 0.119 0.00493
0.00672 0.203 0.00435 0.00677 0.158 0.00457
0.00672 0.254 0.00401 0.00677 0.197 0.00424

so that for C A0 H , C B0 = 0 0632 and CA H = 0.00565 we get
CA H ...B C A0 H − C A H
= 0.00677 − 0.00565 = 0.00112
and CB C B0 − C A H ...B = 0.0632 − 0.00112 = 0.06208
Consequently, K is given by
C A -H ...B
K=
CAH CB
0.00112
= = 3 21
0.06208 × 0.00565
Similarly, K for different values of C A0 H , C B0 and CAH are evaluated and then averaged. The average value of K
determined in this manner for different temperatures are
T0C K
12 3.21
20 2.71
30 2.39
33 2.14
40 2.17
47 1.40
(b) In order to calculate ΔG0 at 298.15 K, ΔH0 and ΔS0 have first to be evaluated. This is achieved by plotting ln
K against l/T (see Fig. 3.44). The slope and intercept of this line are: slope = 3.375 × 103, intercept = −0.17.
But according to Eq. (3.197), the slope = − ΔH0/R and intercept = ΔS0/R.
Therefore, ΔH0 = −3.375 × 103 × 1.987 = −6.706 kcal/mole 1.20
and ΔS0 = − 0.17 × 1.987 = − 0.337 cal/deg/mole
Now ΔG0 at 298.15 K is given by 0.80
InK
ΔG0 = ΔH0 − 298.15 ΔS0 0.40
298.15 × 0.337
= −6.706 + 0.0
103 −0.17
−0.40
= − 6.606 kcal/mole 0 3.10 3.20 3.30 3.40 3.50
1 × 103
Problem 3.36: The self-association of 2-pyridone in CC14 has been T
studied by IR spectroscopy. From the data at 41.6ºC presented in Table 3.11. Fig. 3.44 Variation of equilibrium con-
(a) Calculate the equilibrium constant Kd for the cyclic dimerisation, (b) stant of 1:1 complex with temperature.
determine the absorptivity a, of the monomer at infinite dilution, (c) Calcu-
late ΔH0, ΔG0 and ΔS0 for 2-pyridone at 41.6ºC from the following data.
Temperature ºC Kd ¥ 10−3 (M−1)
16.2 10.43
30.3 6.27
35.3 4.49
41.6 2.71
45.9 2.80
Table 3.11
0
C A × 104 (mole/lit) A(3412 cm−1) (l = 1 cm)
1.999 0.118
3.191 0.162
4.423 0.193
5.305 0.217
6.638 0.243
6.643 0.247
9.040 0.296
9.947 0.315
12.847 0.379
15.402 0.419
16.347 0.418
19.714 0.484
24.163 0.555
Solution (a) The equilibrium constant, Kd for the cyclic dimerisation is calculated from the slope and intercept
of the C A0 / A v vs A plot (Fig. 3.45). 0.8
The slope and intercept of this line according to Eq. (3.188) are 0.6
2
aA0 l 2
slope ( ) = = 1.5 × 10 2 M 0.4
2K d A
0.2
a0 l
intercept (b) = − A = −0.16 M 0.0
2K d
Now according to (3.191), Kd is given by -0.2
0.0 20.0 40.0
C0 A/An × 104
Slope 1 5 × 10 2 M
Kd = = = 2 93 × 103 M 1
Fig. 3.45 Plot between C A0 / Aν
2 (intercept )
2
2(0.16) 2 M 2 and A.
(b) The absorptivity a, of the monomer of infinite dilution is determined from Eq. (3.190), i.e.
aA0 l
intercept (b) = −
2K d
Since K d = 2 93 × 103 M −1
l = 1 cm
b = – 0.16 M
16 M
2 0.16M 2.93 103 M 1
aA0 = = 0.9376 × 103 cm−1
l cm
(c) The values of ΔH0 and ΔS0 are first evaluated by plotting ln Kd vs 1/T (see Fig. 3.46). The slope and intercept
of this line are
slope = 4.16 × 103
intercept = 0.80
But according to Eq. (3.197),
−ΔH
Δ 0 ΔS 0 10.00
Slope = and intercept =
R R
9.00
Therefore, ΔH0 = −1.987 × 4.16 × 103 = − 8.265 kcal/mole In Kd
8.00
and ΔS0 = 0.80 × 1.987 = 1.589 cal/deg/mole
7.00
Consequently ΔG0 at 41.6 °C is given by 3.10 3.20 3.30 3.40 3.50
1
× 103
T
ΔG0 = ΔH0 − 314.75 × ΔS0 = − 8.265 − 0.500 = −8.765 kcal/mole
Fig. 3.46 Variation of equilibrium constant of
dimerisation with temperature.
(ix) Dipole Moments of Intramolecular Bonds Recall from the
foregoing chapter on microwave spectroscopy that the microwave spectroscopic and refractive index measure-
ments yield the dipole moments of the molecule as a whole. The dipole moments of intramolecular bonds can be
determined from the intensities of infrared absorption bands.
Quantum mechanically, the intensity of the infrared absorption band corresponding to the V′ → V″ transition
is related to the square of the quasimoment MV′,V″, by the expression.
3hc
[M V ′,V ″ ]
2
=
8π 2v ∫ A v dν (3.198)
entire band
3hc ∫ ψ * pψ dr
V″
[M V ′ ,V ″ ]
2
=
8π 2v ∫ A v dν = 2 2
(3.199)
entire band
∫ d ∫ ψ dr
dr
V′ V″
Here, the dimensions of charge are (Length)3/2 (Mass)1/2 (Time)−1, ν is the frequency, single and double
primed ψ V are the wave functions for lower and upper vibrational states of the molecules respectively, p is
the electric moment of the chemical bond whose valence vibrations give rise to the given absorption band.
It is thus fundamentally possible to determine the electric moment of the bonds from the area of the absorp-
tion band provided we know the wave functions ψ V ′ (r) and ψ V ″ (r) and the variation of the dipole transition
moment with the interatomic distance r, i.e. in the form of function p (r). The wave functions can be found
from the solution of the Schrodinger wave equation for normal vibrations of the molecule. The dependence
of p upon r can be solved only semiempiricalry, i.e. by correlating the computed values of dipole moment
on the assumption of a certain p-r relationship with directly measured values of the moments of the simplest
molecules. The dipole moments of Cl−H, O−H, C−H, C−F and Si—F bonds, etc., can be determined from
the intensities in the infrared spectra of HC1, H2O, CHCl3, CF4 and SiF4 respectively. The bond moments of
C−H and C−D bonds in C2H2 and in C2D2 and others can also be determined. The dipole moments of some
diatomic molecules are CO(0.11), HCl(1.04), OH(1.54), IC1(0.50), HI(0.38), HBr(0.79) and NO(0.16 Debyes;
1D = 10−18 absolute unit). The dipole moments of some typical linkages are: 2.5 (—C == O); 0.9 (—C—O), 3.3
(—C ≡≡ N) and 1.5 D(—N—H).
The absorption frequency and integrated intensity data on the —C == O, —C—O, —C ≡≡ N and —N—H
linkages (Table 3.9) yield very poor results on dipole moment of the respective linkages suggesting the limited
applicability of the method.
(b) Problems Based on Vibrational Frequency and Bandwidth Data In this part, use has been made of
the vibrational frequency and width of bands to determine the quantities such as force constant, anharmonicity
constant, thermodynamic functions, intermedions, etc.
(i) Strength of Chemical Bonds When a molecule is formed, atoms are bound together by a certain amount
of binding energy which is commonly called the bond strength. The binding energy arises from the electrostatic
interaction of electrons and nuclei. Thus, if it is assumed that the X—Y bond has a strength of k dynes/cm, it
means that there is a force of attraction between X and Y and it takes certain energy to tear them apart, which
is the dissociation energy. If such a system is imagined and if one can say X is displaced from its equilibrium
position and left, the system begins to vibrate with a frequency approximate to the force constant and masses. The
frequency and force constant in case of diatomics are related to each other by the expression (3.19).
k
w e = 1303.16
μm
Thus, we can estimate the force constant for a particular bond if the corresponding vibrational frequency
is known. The value of force constant for a specific bond in any molecule under the same environment is the
same, therefore, one can predict the vibrational frequency of similar vibrators present in different molecules. As
the vibrational frequency of normal modes depends upon a large number of factors, such as mass distribution,
inductive effect, mesomeric effect, etc., hence effect of these factors on force constant could also be examined.
In higher and complex molecules, in addition to stretching force constants, we have to fix the values of force
constants for bending vibrations also, which is not possible with Eq. (3.19). Moreover, in complex molecules the
normal modes of vibration are intermixed with each other due to delocalisation effects, and therefore, it is not
proper to apply Eq. (3.19) even to evaluate the stretching force constants from the vibrational frequencies. Pres-
ently, the mathematical tool of normal coordinate analysis in conjunction with group theory has proved fruitful
for fixing the force constants from vibrational frequencies of normal modes.
In this technique, the frequencies of the normal vibrations in terms of l = 4p2v2 are given by solving the secular
equation
det GF − E λ = 0 (3.200)
where G is the kinetic energy matrix, F the potential energy matrix and E the unitary matrix. The G matrix
elements are generated by the method due to Wilson et al. [Wilson, Decius and Ross, Molecular Vibrations,
McGraw-Hill, New York, 1955].
In actual practice, the kinetic energy matrix elements are not generated by this method since tables of elements
are available in the literature for all common cases: (i) for linear molecules by Ferigle and Meister [Ferigle and
Meister, Am J. Phys., 20 (1952), 421]; and (ii) for planar molecules by Shimanouchi [Shimanouchi, J, Chem Phys.,
25(1956), 660]. However, the most comprehensivε treatment is due to Decius and the expressions reported are
applicable even where the valence angles are 900; 109° 28′; and 120° [Decius, Chem, Phys., 16(1948), 1025].
The elements of F-matrix are obtained from different type of force fields available in the literature and the most
commonly potential fields are Central Force Field (CFF), Simple Valence Force Field (SVFF), Urey–Bradley
Force Field (UBFF), Modified Urey–Bradley Force Field (MUBFF), General Valence Force Field (GVFF), Orbital
Valence Force Field (OVFF), and Most General Harmonic Force Field (MGHFF) .The preferentiability of one
field over the other and assignment of vibrational bands on the basis of potential energy distribution has been dis-
cussed in detail by Ross [Ross, Inorganic Infrared and Raman spectra, McGraw-Hill, London, 1972]. Further, the
force constants have significance only with respect to the potential field in which they appear. In general, the field
in which the number of interaction constants is minimum and transferable in similar molecules is considered to be
the best and UBFF and MUBFF meet these requirements to the maximum extent. Although the potential energy
distribution is a better quantity than normalised amplitudes of vibration for band assignments, but they have their
own significance since they give additional information about force constants. The mean vibrational amplitudes
are generally determined by the method of Cyvin [Cyvin, Molecular Vibrations and Mean Square Amplitudes,
Elsevier, Amsterdam, 1968].
Problem 3.37: HCl gas is known to have a H—Cl stretching frequency at 2886 cm−1. Calculate (i) the stretching
force constant, (ii) the vibrational frequency due to DC1 molecule, and (iii) the ratio of amplitude of vibration of
the nuclei.
Solution We know that for diatomic molecule
v 303. 6( k /μ m )1/ 2
(i) Since v = 2886 cm−1, and
1.008 × 35.5 35.784

mm = = = 0.980 amu
1.008 + 35.5 36.508
( ) 2 × 0.98
Therefore, k= = 4.80 millidynes/ A
( .16) 2
(ii) Since deuteration does not affect the electronic configuration of the molecule, the force constant can safely
be transferred to DCl molecule. Here,
2.016 × 35.5 71.568
μm = = = 1.907 amu
2.106 + 35.5 37.516
k = 4.80 millidynes/A
1/ 2
⎛ 4 80 ⎞
Hence, v = 1303.16 ⎜ = 1303.16 × 1.568 = 2067 cm −1
⎝ 1.907 ⎟⎠
(iii) The ratio of vibrational amplitude of nuclei is

A H m CI 33.5
= = = 35.21
A CI m H 1.008
(ii) Effect of Adsorption on Force Constant The adsorption of a molecule on the surface of an adsorbent
while it severely restricts the rotational movements of the adsorbate, it affects its vibrational movements also.
This would consequently affect a particular vibrational frequency/(S) of the adsorbate and hence can be utilised
to get an estimate of the strength of the bond with the adsorbent that gives rise to that particular vibration. The
vibrational frequency for a diatomic molecule (assumed as simple harmonic oscillator) is given by
v 303. 6( k /μ m )1/ 2
The adsorption of this molecule on the surface of an adsorbent would result in a change in its vibrational fre-
quency v or its force constant. If the value of k decreases as compared to that in the gaseous state, it suggests a
weakening of the bond that holds the adsorbate on the surface of the adsorbent.
Problem 3.38: Gaseous CO is known to have a force constant of the order of 18.53 millidynes/A. When the
molecule is adsorbed on the surface of metal iridium, the C = O stretching bond is observed at 2050 cm−1, show
that the strength of C = O bond is weakened as a result of adsorption on metal surface.
Solution We know that
v 303. 6( k / μ m )1/ 2
v = 2050 cm-1
m1m 2 16 × 12 192
and μm = = = = 6.857 amu
m1 + m 2 28 28
Thus, the molar force constant, k, of CO after adsorption is
( ) 2 × 6.857 °
k ad = 2
= 16.95 millidynes/ A
( .16)
Since kad < k, i.e. 16.96 < 18.53, the strength of CO bond decreases as a result of adsorption of CO molecule the
surface of metal.
(iii) Molar Heat of Evaporation of Water from Infrared Frequency of the Intramolecular Vibration of the
Water Molecules Derived from Vapour Pressure Data The activity au, of non-H-bond water is proportional
to the vapour pressure. Thus, au = constant × p and aB au = C and
d au d aB d p λ
=− = = (3.201)
dT dT dT RT 2
The frequency of the intramolecular vibration of the water molecules, which are in immediate equilibrium with
the molecules without hydrogen bond is related to the molar heat of evaporation l, by the relation
l = NA hν (3.202)
Here, NA is Avogadro’s number, h is Planck’s constant and v the frequency of vibration.
In very dilute solutions of water, if a is the fraction of the H-bonded water molecules and Nu the number of
non-H-bonded water molecules, then the frequency of the band through which the hydrogen bonding is taking
place is given by Maxwell-Boltzmann law, i.e.
⎛ − hv ⎞
α N u = exp ⎜ (3.203)
⎝ κT ⎟⎠
Equation (3.203) can also be written as
− hv
ln α + ln
l Nu =
κT
Since a is constant, we get
d Nu hv
= (3.204)
dT κT 2
Since NA κ = R, the gas constant, therefore, Eq. (3.204) reduces to
d N u N A hv
= (3.205)
dT RT 2
d Nu d p λ N hv
= = 2
= A 2 (3.206)
dT dT RT RT
In not very dilute solutions of water
⎛ − hv ⎞
α au = exp ⎜ (3.207)
⎝ κT ⎟⎠
Taking logarithm and differentiating with respect to temperature, Eq. (3.207) yields
d au d p hv N hv λ
= = = A 2 = (3.208)
dT dT κT 2
RT RT 2
where λ = N A hv = N A hνc , c is the velocity of light and v the wave number. Integrating Eq. (3.208) and applying
the limits of pressure and temperature
p2 T
λ 2 dT
∫d
p1
p=
R T∫1 T 2
T2
λ ⎡ 1⎤
[ln p ]p1 =
p2
⎢− T ⎥
R ⎣ ⎦T1
p2 λ ⎡ 1 1 ⎤
ln = ⎢ − ⎥ (3.209)
p R ⎣T1 T 2 ⎦
p2
R ln
p1
λ=
⎡1 1⎤
⎢T − T ⎥
⎣ 1 2⎦
Since λ = N A hv = N A hcv
R l p 2 / p1
Hence, ν= (3.210)
⎡1 1⎤
N A hc ⎢ − ⎥
⎣T1 T 2 ⎦
Thus, by measuring the vapour pressure of dilute solutions of water at different temperatures one can determine
the frequency v of the intramolecular vibration of water molecules which are in immediate equilibrium with the
molecules without hydrogen bonding. Since v is related to molar heat of evaporation l, one can determine l also
from v , and hence from the vapour pressure data by the formula (3.210).
Problem 3.39: The vapour pressure data in the temperature range 0 −100°C is given in Table 3.12.
Table 3.12
Temperature (°C) P.atm

0 0.006028
10 0.012110
20 0.023070
25 0.031250
30 0.04186
40 0.07279
60 0.19660
70 0.30750
80 0.46740
90 0.69200
100 1.0000
Using the data in Table 3.12,(a) find the temperature intervals between 0 −100°C, in which only one of the known
frequencies of water is in a simple relation to the molar heat of evaporation in which there is no overlapping of
vibrations and hence (b) calculate the molar heat of evaporation from the frequencies thus obtained in the tem-
perature intervals between 0 −100°C. NA = 6.0225 × 1023/mole, h = 6.6256 × 10−27 erg s, c = 2.998 × 1010 cm/s,
R = 8.3143 × 107 erg/mole deg. and 1 cal = 4.186 × 107 erg.
Solution Substituting the values of vapour pressure of water at 0 and 10°C from Table 3.12 in the
Eq. (3.210),
R ln p / p1
v = =
N A hc ⎡ 1 1 ⎤
⎢T − T ⎥
⎣ 1 2⎦
Table 3.13
Temperature ( °C ) v (cm-1)
0 – 10 3746
10 – 20 3714
20
25 – 30 3654
30 – 40 3647
40 – 60 3599
60 – 70 3551
70 – 80 3524
80 – 90 3495
90 – 100 3465
8.3143 × 10 7
=
6.025 × 10 236.6256 × 10 −27 × 2.998 × 101 × ⎡⎣ 4.41472 − ( −5.11134 )⎤⎦ × 7725
= 0.6950115 × 0.69762 × 7225.9 = 3745.91∼3746 cm−1
Similarly, the value of v , in the different temperature intervals are determined and are given in Table 3.13.
From Table 3.13, only those frequencies, in which there is no overlapping of vibrations, are chosen by making
use of experimental frequencies of water and are tabulated in Table 3.14.
The molar heat of evaporation l can, thus be computed by using the data in Table 3.14 and equation, l = NAhcv
and are given in Table 3.15.
Table 3.14
Table 3.15
Temperature Interval ( C) 0
v ,cm from Vapour pressure
−1
Temperature (0C) kcal/mole
Data (from infrared and Raman
spectra) 0–10 10.70
0−10 3746(3755.8)
25–40 10.44
25−40 3654(3650,3651.7)
60–70 10.44
60−70 3551(3550)
90−100 3465(3450) 90–100 9.90
(iv) Thermodynamic Functions from Spectroscopic Data by Statistical Methods Although the need for
accurate thermodynamic quantities such as entropy S0, heat capacity C p0 , enthalpy H0 and Gibbs energy G° is actu-
ally felt in such problems as in calculating the temperature and composition of rocket exhausts and flames, etc., it
is not possible to determine the same at such high temperatures in laboratory. However, all these thermodynamic
functions for a chemical species can be easily and precisely determined from its relevant spectroscopic data
(which is available in most of the cases). Further, a comparison of the calculated and observed quantities provides
an additional evidence for the validity of the vibrational assignments.
Evaluation of the thermodynamic properties of any chemical species requires the knowledge of a quantity
called ‘Zumerstang, Z or partition function, Q. Literally speaking, this indicates how at equilibrium the systems
are partitioned among the available energy states. Mathematically, however, partition function is expressed as
⎛ −E ⎞
Z= ∑g
i
i exp ⎜ i ⎟
⎝κ T ⎠
(3.211)
where the summation is taken over all the energy levels and gi and Ei represent the degeneracy and energy respec-
tively of the ith energy level. Since chemical species possess translational, rotational, vibrational and electronic
energies, the total partition of the chemical species is expressed by
Z= ∑g ⎡⎣ r s + E rot + Evib + E elec ) /

(Etran T ⎤⎦
i
⎛ − Etran
r s⎞ ⎛ − E rot ⎞ ⎛ − Evib ⎞ ⎛ −E ⎞
= ∑ g eexp ⎜⎝
i κT ⎟⎠ ∑ g i exp ⎜⎝
i κT ⎟⎠
∑g
i
i
⎝ κT ⎠ i ∑ g i exp ⎜ elec ⎟
⎝ κT ⎠
= Z tran
r s Z rot Z vib Z elec (3.212)
In most cases at moderate temperatures only the ground electronic level is populated so that
Z = Ztrans Zrot Zvib (3.213)
3
⎛ 2p m κT ⎞ 2
Further, Z tran
r s V⎜ ⎟⎠ (3.214)
⎝ h2
1 3
Zrot =
(π I A I B I C ) 2 ⎛ 8π 2κT ⎞ 2
σ ⎜⎝ h2 ⎟⎠ (for asymmetric tops) (3.215)
− gi
⎡ ⎛ − hcw i ⎞ ⎤
and Zvib = ∏ ⎢1 − exp ⎜ (3.216)
i ⎣ ⎝ κT ⎟⎠ ⎥⎦
where V is the total volume of the system, m, IA, IB and Ic denote the molecular mass (in grams) and the three
principle moments of inertia respectively of the chemical species Further, s is the symmetry factor, wi is its ith
vibrational frequency (in cm−1).
The various thermodynamic quantities are related to Z by means of the well-known relations.
⎛Z⎞
G0 E 00 − RT ln ⎜ ⎟ (3.217)
⎝N⎠
⎡ d ⎤
H0 E 00 − RT 2 ⎢ (ln Z ) R (3.218)
⎣ dT ⎦
d ⎡ 2 d ⎤
C p0 = ⎢ RT dT (ln Z ) R (3.219)
dT ⎣ ⎦
and S0 =
(H 0
G0 ) (3.220)
T
where R is the gas constant, E 00 is the zero point energy and T is the absolute temperature.
For the purpose of actual calculations, however, these thermodynamic functions are expressed in the following
form.
G0 E 00
= −R l Z + R
T
⎡⎧ ( )3/ 2 RM 3/ 2T 5/ 2 p −1e ⎫
= − R ln ⎢ ⎨ 3 5/ 2
( = Z tran
r s )⎬
⎣⎩ h NA ⎭
⎪⎧ 8π ⎛ 2πκ ⎪⎫
3/ 2
2
κT ⎞
⎜⎝ 2 ⎟⎠ (I A I B I C ) ( = Z rot )⎬
1/ 2
×⎨ (3.221)
⎩ σ h ⎭
⎧3 −6 / 3 −5
⎫
− gi
⎤
×⎨ ∑ ⎡⎣1 − exp ( − x i )⎤⎦ ⎬ ⎥+ R
⎩ i ⎭ ⎥⎦
⎡⎣ M mN p = 1.0133 × 105 Pa⎤⎦
⎡ 4 + ∑ g xi ⎤
H0 E 00
= R ⎢⎢ i
2⎥
⎥ (3.222)
exp x i − 1)
⎢⎣ (
T
⎥⎦
⎡ g ( x i ) exp x i ⎤
2
C p R ⎢4 + ∑
0
⎥ (3.223)
⎢⎣ i (exp x i − 1)2 ⎥⎦
hcω i
where xi = = 1.4388ω i / T .
κT
Substituting the values of various constants, for an ideal gas at one atmospheric pressure Eq.(3.221) may be
simplified to
⎡ ⎤
⎢(3.45391log M 5.7565l 6 logT − 2.3026 log p + 8.8612)⎥
G 0 E 00 ⎢ ⎥
= − R ⎢ + (0.5 × 2.3026 log(I A I B I C ) 3.4539 logT ⎥+R (3.224)
T ⎢ ⎥
⎢ −2.3026 log σ − 4.2105) − g i ln ∑ ⎡1 − exp ( − x i )⎤ ⎥
⎢⎣ ⎣ ⎦ ⎥⎦
i
⎡9.2104 logT − 2 306 l g p + 1.1513 log(I A I B I C ) ⎤

= −R ⎢ ⎥
3 2 log σ + 4.6507
⎢⎣ +3.45391log M − 2.3026 ⎥⎦
+ Rg i l ∏[
i
− exp( − x i )] + R
where M is the mass of the molecule in amu and IA, IB and IC are the principal moments of inertia in amu Å2.
• Isotopic Species If the system contains isotopically substituted molecules, then the thermodynamic func-
tions for mixtures are calculated using the formula
K ∑N K
i
i i (3.225)
where K is any one of the thermodynamic functions for the mixture, Ni represents the mole fraction for the ith
species for which the value of the function is Ki.
• Entropy and Free Energy of Mixing The entropy and free energy changes due to mixing of non-interacting
ideal gases (for a total of 1 mole of the mixture of ideal gases) are given by
ΔS 0 = − R ∑ N i l N i (3.226)
i
ΔG 0
= R ∑ N i ln N i (3.227)
T i
where N i is the mole fraction of ith species.

• Heat, Free Energy and Equilibrium Constant of Formation
0
The heat Δ , free energy ( 0
(), )
entropy ( ) and equilibrium constant K p in terms of pressure, of formation of any molecule can be
computed at the desired temperatures using the following relation.
ΔHf T0 = ΔHf 298

0
.15
=( 0 0
) compound − ∑ ( 0 0
) elements (3.228)
ΔGf T0 = Δ
ΔHff T0 − T ΔS f T0 (3.229)
) − ∑ST (
0
where ΔSf T0 = S T0 ( ) (3.230)
−ΔGf T0
log10 K p = (3.231)
0.004575845T
Problem 3.40: The vapour-phase infrared spectra of chlorobenzene have been recorded and the following
frequencies (in cm−1) are assigned to its thirty normal modes of vibration, 416.8, 706.5, 1003.7, 1025.7, 1092.6,
1153, 1482.3, 1586.4, 3031, 3054, 3082, 294.7, 614.9, 1067.6, 1167.1, 1271.8, 1326.6, 1447.2, 1598.2, 3067,
3096, 197.5, 467.1, 683.9, 741.4, 902.5, 981.5, 403.4, 831.2, 961.7. The moments of inertia in amu Å2 along the a,
b and c axes are 89.113, 320.611 and 409.7775, and for 35ClC6H5 and for 37ClC6H5 89. 119, 329.811 and 418.9816
respectively. Calculate (a) C p0 ( H 0 E 00 ) / T , ( 0
) / , S0 for chlorobenzene at 298.15 K and p = 1.0133 × 105
0
Pa, (b) ( ) / , for chlorobenzene at 298.15 K if it contains the two isotopes of chlorine of 35 and 37 amu
in the ratio 3 :1 and molecular mass in amu for 35ClC6H5 is 112.07560 for 37ClC6H5 is 114.07265, (c) the entropy
(ΔS0) and free energy (ΔG0/T) of mixing of C6H5Cl35 and C6H5Cl37 and (d) The heat (ΔHf 0), free energy (ΔG0/T),
and the equilibrium constant in terms of pressure (log10 Kp), for the formation of chlorobenzene at 298.15 K.
(ΔHf 0298.15 = 12.39 kcal mole−1)
Solution The exponential terms used in the determination of thermodynamic functions from Eqs (3.222),
(3.223) and (3.224) have been computed and are recorded in Table 3.16.
Table 3.16 The exponential terms determined from the vibrational frequencies.
exp
x xi xi
wi(cm−1) xi Exp xi A= xi2 A 1 - exp ln[1-exp
xp x i − )
( exp
2
(exp xi − 1 )2
(1) (2) (3) (-xi ) (-xi )]
(4) (5) (6) (7) (8)
416.8 2.0113763 7.4735961 0.1783358 0.7214813 0.0479957 0.8661956 −0.1436445
706.5 3.4093986 3.0247049 × 101 0.0353605 0.4110309 0.0039858 0.9669389 −0.0336200
1003.7 4.8436142 1.2692725 × 102 0.0080042 0.1877822 0.0003154 0.9921215 −0.0079097
1025.7 4.9497808 1.4114403 × 102 0.0071864 0.1760692 0.0002520 0.9929150 −0.0071102
1092.6 5.2726241 1.9492680 × 102 0.0051832 0.1440952 0.0001402 0.9948699 −0.0051433
1153.0 5.5640999 2.6089028 × 102 0.0038626 0.1195825 0.0000823 0.9961670 −0.0038404
1482.3 7.1532223 1.2782182 × 103 0.0007836 0.0400939 0.0000043 0.9992177 −0.0007826
1586.4 7.6555838 2.1124081 × 103 0.0004738 0.0277709 0.0000017 0.9995266 −0.0004735
3031 14.6268751 2.2509774 × 106 0.0000004 0.0000950 2.8867534 × 10−12 0.999996 −0.0000004
3054 14.7378675 2.5152117 × 106 0.0000004 0.0000863 2.3296245 × 10−12 0.9999996 −0.0000004
3082 14.8729888 2.8791013 × 106 0.0000003 0.0000768 1.7942570 × 10−12 0.9999997 −0.0000003
294.7 1.4221511 4.1460295 0.4188966 0.8472241 0.1436879 0.7588054 −0.2760099
614.9 2.9673591 1.9440512 × 101 0.0571692 0.5033874 0.0087262 0.9485610 −0.0528092
1067.6 5.1519801 1.7277327 × 102 0.0058555 0.1554225 0.0001746 0.9942121 −0.0058047
1167.1 5.6321431 2.7925997 × 102 0.0036067 0.1144075 0.0000727 0.9964191 −0.0035873
1271.8 6.1374001 4.6284865 × 102 0.0021699 0.0817351 0.0000287 0.9978395 −0.0021629
1326.6 6.4018517 6.0296051 × 102 0.0016640 0.0681968 0.0000176 0.9983415 −0.0016599
1447.2 6.9838382 1.0790520 × 103 0.0009285 0.0452847 0.0000060 0.9990733 −0.0009272
1598.2 7.7125278 2.2361877 × 103 0.0004476 0.0266246 0.0000015 0.9995528 −0.0004473
3067 14.8006024 2.6780578 × 106 0.0000004 0.0000817 2.0636680 × 10−12 0.9999996 −0.0000004
3096 14.9405494 3.0803365 × 106 0.0000003 0.0000724 1.5746011 × 10−12 0.9999997 −0.0000003
197.5 0.9530874 2.5937051 1.0211856 0.9276201 0.3752466 0.6144511 −0.4870259
467.1 2.2541120 9.5268293 0.1310308 0.6657697 0.0310028 0.8950333 −0.1108944
683.9 3.3003365 2.7121763 × 101 0.0397478 0.4329417 0.0048367 0.9631292 −0.0375677
741.4 3.5778176 3.5795336 × 101 0.0295654 0.3784607 0.0029551 0.9720634 −0.0283343
902.5 4.3552474 7.7886089 × 101 0.0131754 0.2499137 0.0007367 0.9871607 −0.0129224

Table 3.16 The exponential terms determined from the vibrational frequencies—Cont'd.
exp
x xi xi
wi(cm−1) xi Exp xi A= xi2 A 1 - exp ln[1-exp
xp x i − )
( exp
2
(exp xi − 1 )2
(1) (2) (3) (-xi ) (-xi )]
(4) (5) (6) (7) (8)
981.5 4.7364823 1.1403236 × 102 0.0089253 0.2002324 0.0003707 0.9912306 −0.0088081
403.4 1.9467111 7.0056089 0.1942369 0.7360966 0.0539743 0.8572572 −0.1540172
831.2 4.0111708 5.5211472 × 101 0.0187865 0.3022660 0.0013649 0.9818878 −0.0182782
961.7 4.6409323 1.0364092 × 102 0.0098376 0.2118852 0.0004405 0.9903513 −0.0096955
7.7757872 0.6764115 −1.4134782
30 gi x i 30 xi
Now ∑ =∑ = 0.6764
i =1 (exp x i − 1) i =1 (exp x i − 1)
2 2
30 30
∑ gi i )] ∑ ln[1 − exp( − x i )]= − 1.4135
i =1 i =1
30 g i x i2 exp x i 30
and ∑ = ∑ x i2 A = 7.7757
i =1 (exp x i − 1)
2
i =1
Consequently,
H0 E 00 ⎡ 30 g xi ⎤
(a) = R ⎢4 + ∑ 2⎥
T i =1 (exp x i − 1)
⎣ ⎦
= 1.987 [4 + 0.6764]
= 9.29 cal deg−1 mole−1
⎡ 30 g x 2 exp x ⎤
C p0 R ⎢4 + ∑ i i
2⎥
i =1 (exp x i − 1)
⎣ ⎦
= 1.987 [4 + 7.77571] = 23.39 cal deg−1 mole−1
G0 E 00
= −1.987 [9.2104 log 298.15 − 2.306 log 1.0133 × 105
T
+ 1.1513 log (89.113 × 320.611 × 409.777)
+ 3.4539 log 112.07560 − 2.3026 log 2 + 4.65071]
+ 1.987(−1.4135) + 1.987 = −61.30 cal deg−1 mole−1.
H0 E 00 G0 E 00
and S0 = + = 61.30 + 9.29
30
T T

0 0
(b) In order to calculate ( ) /T chlorobenzene having isotopes of chlorine 35 and 37 amu, we must first
0 0
calculate the value of ( ) /T for 37C1C6H5. From Part (a), we can show that
G0 E 00
= − 61.40 cal deg−1 mole−1
T
0 0
Therefore ( ) /T for chlorobenzene containing isotopes 35Cl and 37Cl in the ratio 3 : 1
= ∑ N i K i (from 3.225)
i
⎛3 1 ⎞
= × −61.30 + × −61.40⎟
⎝4 4 ⎠
= −61.32 cal deg−1 mole−1
The values of energy parameters computed for 35ClC6H5 together with those of C(graphite), H2(gas), C12(gas)
and the experimental value of the heat of formation of chlorobenzene gas have been used to calculate ΔHf 0
T Δ
ΔGf
Gf 0
T
and log10Kp from relations (3.228) to (3.231) for the reaction
5 1
6 C(graphite) + H2(gas) + Cl2(gas) = C6H5Cl(gas)
2 2
Δ
ΔHf 0
298.15 = 12.39 kcal mole−1
The absolute values of entropy in cal deg−1 for elements involved in the reaction are
Cl2(g) = 53.286, H2(g) = 3 1.211, C(graphite) = 1.3609
Thus,
ΔSf 298
0
.15 = S 298.15 (
0
) − S 298
0
.15 ( )
⎛ 5 1 ⎞
= 70.59 − ⎜ 6C + H 2 + Cl 2 ⎟
⎝ 2 2 ⎠
⎛ 5 53.286 ⎞
= 70.59 − ⎜ 6 1.3609 + × 31
3 .211 + ⎟
⎝ 2 2 ⎠
= 70.59 − 112.84 = − 42.25 cal deg−1 mole−1
0
Gf 298.15 ΔHff 298
0
.15 − T ΔS f 298.15
0
= 12.39−298.15 × (−42.25 × 10−3)
= 24.98 kcal mole−1
−ΔGf 2298
0
and log10 K p = .15
0.004575845T
24.98
=− = −18.31
0.004575845 × 298.15
or Kp = 2.04 × 1018
(c) The entropy and free energy changes due to mixing of 35ClC6H5 and 37ClC6H5 in the ratio 3:1
ΔS 0 = − R ∑ N i l N i
i
= −R(0.75 In 0.75 + 0.25 In 0.25)
= −1.987 × −0.250

ΔG 0
and = R ∑ N i ln N i
T i
= 1.987 × −0.250
= −4.497 cal deg−1 mole−1

In order to facilitate calculations by computer, simplified expressions for contribution to the corresponding
thermodynamic quantity, owing to various types of degree of freedom are summarised below.
For diatomic and polyatomic molecules, the contribution to thermodynamic quantity due to translational
motion is
3 5
( 0
)
p trans R+R R = 4.968 (3.232)
2 2
3 5 5
0
S trans R l M + R lnT − R l p R − 7.282
2 2 2
= 6.8635 log M+ 11.4391 log T− 4.5756 log p− 2.3 13 (3.233)
= 6.8635 log M+ 11.4391 log T− 2.313 (p = 1 atm)
⎛ G 0 E 00 ⎞ 3 5
⎜⎝ T ⎟⎠ = − R l M − R l T + R l p + 7.282 (3.234)
trans
2 2
= −6.8635 log M− 11.4391 log T+ 7.282
⎛ H 0 E 00 ⎞ 3 5
⎜⎝ T ⎟⎠ = R + R = R = 4.968 (3.235)
trans
2 2
For C2H4(σ = 4) at 298.15 K and 1 atm pressure, we get from Eqs (3.232) to (3.235).
( 0
) trans = 4.968 cal mole −11K 1
0
Strans = 35.924 cal mole 1K −1
⎛ G 0 E 00 ⎞
⎜⎝ T ⎟⎠ = −30.960 cal mole −11 K 1
trans
⎛ H 0 E 00 ⎞ −11 1
⎜ ⎟ = 4.968 cal mole K
⎝ T ⎠ ttrans
For diatomic and linear polyatomic molecules, the contribution owing to rotational motion is
0
( ) rot R (3.236)
0
S rot = R ln T+ R ln I – R ln σ + 177.67 (3.237)
⎛ G 0 E 00 ⎞
⎜⎝ T ⎟⎠ = −R ln T − R ln I + R ln σ − 175.683 (3.238)
rot
⎛ H 0 E 00 ⎞
⎜⎝ T ⎟⎠ = R (3.239)
rot
For polyatomic nonlinear molecules,
0 3
( ) rot R (3.240)
2
3 1
0
S rot l T + R ln (I A I B I C ) R ln σ + 267.54
R ln (3.241)
2 2
⎛ G 0 E 00 ⎞ 3 1
⎜⎝ T ⎟⎠ = − 2 R l T − 2 R l (I A I B I C ) + R ln σ − 264.559 (3.242)
rot
⎛ H 0 E 00 ⎞ 3
⎜⎝ T ⎟⎠ = 2 R (3.243)
rot
For C2H4, IA = IB = IC = 17.33 × 10−40 g cm2 and at 298.15 K and 1 atm pressure; we obtain from Eqs (3.240) to
⎛ G 0 E 00 ⎞
(3.243): S 0
rot = 15.758 cal mole K , (
−1 −1 0
)
p rot = 2.981 cal mole K , ⎜⎝ T
−1 −1
⎟⎠ = −12.848 cal mole−1 K−1, and
rot
⎛ H 0 E 00 ⎞
⎟⎠ = 2.983 cal mole K .
−1 −1
⎜⎝ T
rot
The contribution due to vibrational degrees of freedom is
3n 6 / 3n 5 ⎡ x i2 exp.(
p.(( ) ⎤
eexp
( 0
) vib = R ∑ ⎢ 2⎥
(3.244)
i =1
⎣ (exp.( x i ) ) ⎦
3n 6 / 3n 5 ⎡ xi ⎤
0
S vib =R ∑ ⎢ exp.( x ) − 1 − ln(1 − exp( − x i )) ⎥ (3.245)
i =1
⎣ i ⎦
⎛ G 0 E 00 ⎞ 3n 6 / 3n 5
⎜⎝ T ⎟⎠ = R ∑ ln(1 exp( x i )) (3.246)

i =1
vib
⎛ H 0 E 00 ⎞ 3n 6 / 3n 5 ⎡ xi ⎤
⎜⎝ T ⎟⎠ = R ∑ ⎢ ⎥ (3.247)
vib
i =1
⎣ exp.( x i ) − 1⎦
For numerical calculation, xi, is expressed as
hcw i wi w
xi = = = 1.439 i
κT 0.6951T T
where wi is in cm−1.
Summation for polyatomic linear molecules covers 3n − 5 vibrations and for polyatomic nonlinear molecules
3n − 6 vibrations, where n is the number of atoms in the molecule.
For C2H4 the values of qi ( = hcwi /k) are 4344, 2335, 1931, 1187, 4424, 1581, 1365, 1357, 4469, 1432, 4302,
2078. Using this data, we obtain from Eqs (3.244) to 3.247).
0
p vib = 2.422 cal mole−1 K−1, S vib
0
= 0.623 cal mole−1 K−1,
⎛ G 0 − E 00 ⎞
−⎜ ⎟⎠ = 0.111 cal mole K
−1 −1
⎝ T vib
⎛ H 0 E 00 ⎞
⎟⎠ = 0.512 cal mole K .
−1 −1
and ⎜⎝ T
vib
Based on expressions (3.232) to (3.247), a computer programme in C+ + for the computation of thermodynamic
functions is given in Appendix 3.6A. The vibrational and other relevant data pertinent to determine the thermody-
namic functions manually and with the aid of computer programme are reported in Table 3.17. This data will be
beneficial to students for generating and solving the problems.
For more accurate results, the contribution due to other degrees of freedom, viz., internal rotation, nuclear
spin and electronspin must be added to the corresponding thermodynamic property whenever the situation
demands.
Table 3.17 Symmetry numbers, moments of Inertia and vibrational frequencies for some selected gases.
Gas σ I 10 40 ( g cm2 ) Frequency (cm−1)
Br2 2 342.5 325
Cl 2 2 113.9 560
F2 2 27.8 918
H2 2 0.463 4400
I2 2 741.5 214.55
N2 2 13.84 2358
O2 2 19.23 1580
CO 1 14.43 2170
HBr 1 3.26 2649
HCl 1 2.65 2991
HF I 1.35 4139
HI 1 4.31 2309
NO 1 16.51 1904
OH 1 1.50 3738
CO2 2 70.2 1337, 2349, 667 (doubly degenerate)
CS2 2 24.7 657, 1523, 397 (doubly degenerate)
HCN 1 18.7 2089, 3312, 712 (doubly degenerate)
H2O 2 1.784 3652, 1595, 3756
H2S 2 3.63 2611,1290,2684
N2O 1 66.1 589, 1285, 2224
SO2 2 42.5 1151, 524, 1361
3374, 1974, 3287, 612 (doubly degenerate) 729 (doubly

C2H2 2 23.51
degenerate)
3337, 950, 3414, (doubly degenerate), 1628 (doubly
NH3 3 3.22 degenerate)
2914, 1526 (doubly degenerate), 3020 (triply degenerate),
CH4 12 5.30
1306 (triply degenerate)
• Partition Function for Internal Rotation In complex molecules, the rotation of one group of atoms
relative to another (for example of the group CH3 about the bond C—C in ethane) must be taken into consid-
eration. In case of large barrier height, U, for instance for CH3—CH3, it is 3150, and for CH3OH, it is 2700
cal mole−1, the internal rotational motion is treated as torsional vibration and its contribution is added to the
thermodynamic functions. On the other hand, if U is small, the torsional frequency is dropped out and the
rotational partition function equals the product of Zrot. Zintrot. Zintrot is known as the partition function for free
rotation and is expressed as
( )
1/ 2
I T 2.7935 × 1019 (I redT )1/ 2
Z introt = = (3.248)
hσ introt σ introt
ln Z introt 1. log I redd − 2. log σ introt + 1. logT + 52.8356

logT
Here Ired = reduced moment of inertia about the axis of internal rotation. σ introt = symmetry number of internal
rotation. It is equal to the number of positions of maximum attraction between the atoms and rotating groups in
rotation through 360°, i.e. the number of potential maxima occurring in one revolution of the rotating group. For
C2H6, for example σ introt = 3, i.e. during one revolution the hydrogen atoms of one CH3 group occupy position
corresponding to the minimum distance from a hydrogen atom of the other CH3 group three times.
For two groups rotating relative to one of the axes of the molecules, I is formed by the equation
IAIB
I red = (3.249)
IA IB
where IA and IB are the moments of inertia of the groups rotating about this axis. If a molecule contains several
groups performing internal rotation, the calculation should be performed for each of them, and then the results
are summated.
If there is free internal rotation, the contribution owing to internal rotation to (H0 – E 00 ) and C p0 at not too low
temperatures is given by
1
( 0
)introt = R T (3.250)
2
0 1
( )
p introt R (3.251)
2
If there are several free internal rotations, a corresponding number of terms as in Eqs (3.250) and (3.251) have
to be added.
For torsional oscillation of 800 cm−1, the contribution to C p0 at T = 200 K as worked out from Eq. (3.244) for
vibrational contribution to C p0 has been found to be 0.21 cal mole−1 K−1 as compared to 0.99 cal mole−1 K−1 from
Eq. (3.251) for one free rotation.
The contribution to entropy and Gibbs free energy is given by
R ⎡ ⎛ 8π 3κ ⎞ ⎤
r = ⎢l T + l I red − 2 l + ln ⎜ 2 ⎟ + 1⎥
0
S introt introt (3.252)
2 ⎣ ⎝ h ⎠ ⎦
= 2.2868(log + llog
og I red − 2 l g ) + 89.932
⎛ G 0 − E 00 ⎞ R
−⎜ ⎟ = S introt
0
− (3.253)
⎝ T ⎠ introt 2
= S introt
0
− 0.993
For molecules like C2H6 or CH3—C≡C—CH3, σ introt = 3, and Ired = IA/4 = 2.759 10−40 g cm2, we get from
Eq. (3.252)
0
S introt = 2.2868 log T−2.714 (3.254)
•Partition Function for Nuclear Spin Owing to the very high energies of the excited nuclear states, the
nuclear partition function of any molecule or atom is the degeneracy of the ground state nuclear energy level. The
nuclear partition function is expressed as
⎛ − E 0 ( nuc ) ⎞ ⎛ − E l ( nuc ) ⎞ ⎛ − E 2( ) ⎞
Z g 0 ( nuc) exp ⎜ + g1( nuc ) exp ⎜ + g 2( exp ⎜ ⎟⎠ + ... (3.255)
⎝ κT ⎟⎠ ⎝ κT ⎟⎠
)
⎝ κT
Since higher states are rarely populated, Eq. (3.255) becomes

⎛ − E 0 ( nuc ) ⎞
Z g 0 ( nuc ) exp ⎜ (3.256)
⎝ κT ⎟⎠
If E 0 ( ) = 0, Eq. (3.256) reduces to

Z ( ) g 0( ) (3.257)
Further, the degeneracy of ortho state is given by
g ortho
( ) (I )( 2I
)( ) (3.258)
where I is the spin quantum number of the nucleus.

1
For hydrogen atom, I = ,
2
Z ortho
( ) g 0ortho
( ) =3
For deuterium atom, I = 1

Z ortho
( ) g 0ortho
( ) =6
When nuclear spin occurs, a factor taking into account its influence on the rotational level must be introduced
into the expression for rotational partition function. Assuming that Zrot and the nuclear partition function Znuc to be
independent of each other, we get
Z rotnuc Z rot Z nuc
= Z rot ∏ ( I i + ) (3.259)
i
= Zrot (2I1 + 1) (2I2 + 1) (2I3 + 1)……

where Ii , is the nuclear spin quantum number.
The number of multipliers of the form (2Ii + 1) in Eq. (3.259) equals the number of atoms in the molecule.
The spin of a nucleus equals the sum of the spins of the particles which form it. Therefore, if the sum of the
protons and neutrons is an even number, the spin equals an integer or zero, otherwise the spin is a half integer. The
values of spin quantum number for selected elements are given below.
Element C, He, O F,H,P D, N Cl
Spin 0 1/2 1 5/2
0 0 0
Spin does not influence ( )/ , but it does change the entropy and Gibbs energy. In this con-
p
nection, complete (absolute or spectroscopic) and practical (thermochemical) entropies/Gibbs energy are distin-
guished. The latter equals the total entropy/Gibbs, energy less the spin entropy, i.e.
b −R∑ ( 2I + 1)
0 0
S pract S abs (3.260)
⎛ G 0 E 00 ⎞ ⎛ G 0 E0 ⎞
Similarly ⎜⎝ T ⎟⎠ = − ⎜⎝ T ⎟⎠ − R ∑ l ( 2I 1) (3.261)
pract abs
We may consider that every atom has an identical spin regardless of whether it is in the isolated state or is a part
of a compound. Therefore, in calculating ΔS 0 and ΔG 0 and, consequently, in calculating equilibrium, the spin
components mutually eliminate each other. Nuclear spin must be taken into account only for very low tempera-
tures (for hydrogen below 0°C, for deuterium below 70°C, and for other substances at still lower temperatures).
The standard entropy of ethene with account taken of nuclear spin effect is 57.98 cal mole−1 K−1 We know that
0
S 298 S trans + S r0ot 0
S vib
For ethene, 0
S 298 = 36.02 + 15.80 + 0.65 52.47cal mole −11K 1
If the hydrogen spin nuclear effect is considered, we get from Eq. (3.260)
⎛ 1 ⎞
0
S 298 . R ln 2 × + 1 57.98 cal mole −11K 1
⎝ 2 ⎠
In case of molecules, consisting of same atoms, the rotational terms are divided into two intercombining
groups, having different statistical weights. The terms with greater statistical weight are called ortho terms while
with less statistical weight are called para terms. When nuclear spin is 0, the para terms are absent. The Z rot in
this case contains members either with even or odd J. In case of nuclear spin, however, we have
Z rot Z nuc = g ∑( J + ) exp [ −BhcJ

BhcJ J )/ T ]
ortho
+g p ∑( J + ) exp [ − + ] (3.262)
para
In case of H2, the ortho term (I = 1/2) have statistical weights, gortho = 3 (for odd values of J) and the para terms
(I = 0) gpara = 1 (for even values of J), Consequently, the partition function, Eq. (3.262), in such a case is expressed as
Z rot Z nuc = 3 ∑ (2J 1) p [ BhcJ

Bh
hcJ
cJ J 1 T]
J =1,3,5
+ ∑ (2J + 1) p [− + ] (3.263)
J = 0 ,2,4
For sufficiently high temperatures, (T > 100 K, Bhc << kT) each term of the sums of Eq. (3.263) can be approx-
imated to κT /2 B hc . Heat capacity, Cv, of both the modifications of hydrogen is equal to 5R/2, However, at low
temperatures, the heat capacities of both the modifications of hydrogen are different due to different dependence
of each of the two sums of Eq. (3.263) upon temperature. Since ordinary hydrogen is a mixture of three parts of
ortho and one part of para modifications, its total heat capacity, the rotational components of which are defined
by the formula
3C ν ,rot,ortho + C ν,rot,para
C rot = (3.264)
4
exhibits a peculiar temperature variation that differs from ordinary trend. This anomaly in the heat capacity of
hydrogen was the first experimental fact that led to the discovery of its ortho- and para-modifications.
• Partition Function for Electronic Excitation The degeneracy of each electronic level gelec = 2 j+ 1. There-
fore, the expression for electronic partition function may be expressed as
i =1, 2 ,3
⎛ − Ei ( elec ) ⎞
Z ∑(
i =0
ji + ) exp ⎜
⎝ κT ⎟⎠
⎛ − E1( ) ⎞
= ( j + ) + ( j + ) exp ⎜ ⎟⎠ + ... (3.265)
⎝ κT
where ji, is the quantum number of the electronic energy levels and equals the orbital quantum number plus the
total spin of the electrons in the molecule.
The electronic levels correspond to a very high energy, i.e. only a negligible fraction of the molecules is in an
excited state. The influence of second term is noticeable only when T > hv / 4κ . Other levels have to be taken
into account in addition to the ground level only at very high temperatures, when optical excitation of the mol-
ecules takes place. Even in this case, however, we may limit ourselves to two or three terms of Zelec. Consequently,
because of the absence of multiplet electronic levels in the majority of polyatomic gases except Cl2, O2, NO and
certain other gases at T > 2000 K, we can conclude that j = 0, gelec = 1 and Zelec = 1.
0
elec ln g 0 (elec ) (3.266)
⎛ G 0 E 00 ⎞
⎜⎝ T ⎟⎠ = − R l g 0 (elec ) (3.267)
elec
The electronic partition function of chlorine atom in the ground state (2P3/2) as computed from Eq. (3.265) is
⎛ 3 ⎞
Z l ) = ( 2 j 0 + 1) = 2 ×
g 0 ( elec + 1⎟ = 4
⎝ 2 ⎠
provided the energies of the higher excited states are very large. If the energy of the first excited state of chlorine
atom (2P1/2) is 1.8 × 10−20J, the electronic partition function at 1800 K as determined from Eq (3.265) is
⎛ −E ( )⎞
Zelec = ( j + ) + ( j + ) exp ⎜ 1 ⎟⎠
⎝ κT
⎡ −1 8 × 10 −20 J ⎤
= 4 + 2 exp ⎢ −23 −1 ⎥ = 4 + 2 exp (−0.724) = 4.969
⎣ (1.381 × 10 JK )(1800 K ) ⎦
(v) Frequency of Inactive Vibration Alternatively, the thermodynamic data, especially C p0 , can also be
employed to determine the frequency of an inactive vibration provided the frequencies of all its active modes are
determined spectroscopically. For this purpose C p0 of the chemical species is calculated by omitting the unknown
frequency, comparison of this value with the observed C p0 yields ΔCp. This is the contribution of the missing
frequency, v, to the calculated C p0.
The inactive frequency v is then calculated from the relation
g Rx 2e x
ΔC p = (3.268)
( )
2
−
hv 1.4388v
where x= =
κT T
Problem 3.41: The symmetry number of methane is 12. If its vibrational spectra contain the frequencies
2914.2, 1526.0, 1526.0, 3020.3, 3020.3 and 3020.3 cm−1, calculate the missing frequency (assuming that it exists
as a triplet) if its C P0 is 8.57 cal mole−1 deg−1.
Solution The IR spectra of methane should have 3n−6 (n = 5) = 9 normal modes of vibration and consequently
according to the Eq. (3.223). the observed
⎡ 9
g x i2 exp x i ⎤
C P0 = ⎢ 4 + ∑ 2⎥
⎣ i =1 (exp x i − 1) ⎦
But the calculated C0P for methane with the above frequencies would be
⎡ 9
g x i 2 exp x i ⎤
C P0 ( ) R ⎢4 + ∑ 2⎥
⎣ i =1 (exp x i − 1) ⎦
= 8 .08 cal deg−1mole−1

The difference between the observed and calculated C P0 for methane is contribution ΔC p of the missing fre-
quency to the calculated C P0 . Consequently,
ΔC p = 8.57 − 8.08 = 0.49 cal deg−1 mole−1
Rg i x 2e x
But ΔC p =
(e x − 1) 2
2 x
0 49 3( )
Therefore = x
1.987 ( 1) 2
x 2e x
or = 0.082
(e x 1) 2
This equation is solved graphically for various values of x and it yields x = 6.2.
v 1.4388v
But x = 1.4388 =
T 298.15
Therefore, ν = 1286.3 cm−1

This calculated missing frequency compares reasonably well with the observed frequency of 1306.2 cm−1.
(vi) Equilibrium Constants from Frequency Data According to the well-known thermodynamic relation for
Gibbs’ free energy of a gaseous substance
G = F + pV
⎡ ⎛ ∂ lnZ ⎞ ⎤
= − N A T ⎢lnZ − V ⎜ ⎟ ⎥ + E 00 (3.269)
⎣ ⎝ ∂V ⎠ T ⎦
Thus, the standard free energy change of the gaseous, chemical reaction can be determined from molecular
parameters and spectroscopic data of the reactants and products. Hence, from the equation
ΔG 0 = − R T l K (3.270)
we can get the equilibrium constants K. Assuming the gas to be ideal, Eq. (3.269) modifies to
G N A κT ( ) E 00 (3.271)
⎡ ∂ lnZ rot ⎤ ⎡ ∂ ln
l Z vib ⎤
since ⎢ ∂V ⎥ = ⎢ ∂V ⎥ = 0
⎣ ⎦T ⎣ ⎦T
⎡ ∂ lnZ trans ⎤ 1
and ⎢ ∂V ⎥ = V (3.272)
⎣ ⎦T
For one mole, Eq. (3.271) reduces to
G0 RT l trans rot Z vib + E 00 RT (3.273)
Let us consider a reaction, A + B → C .

Since the partition function of each species of the chemical reaction is evaluated relative to a different reference
point, thus while calculating ΔG 0, a common zero must be selected. This is achieved by carrying out the reaction
at 0 K, since at this temperature molecules will have only zero point energy as the translational and rotational
energy of the molecules will be zero. However, the potential energy curves of molecules may lie at different levels
on the absolute energy scale. The potential energy diagrams for three diatomic molecules is shown in Fig. 3.47,
Selecting the arbitrary zero as shown in Fig. 3.47, we may write
G A0 RT l A
trans Z rAot Z vib
A
a + RT
G B0 RT l B
Z rBot Z vib
B
b + RT (3.274)
U(r)
trans
G C0 RT l C
Z rCot Z vib
C
c + RT V=0
trans
V=0
V=0
But (c − a − b) is the total heat of reaction at absolute zero includ- a b
r
c
ing the zero point energies. Let us call this quantity ΔE

Δ 00. Hence, from
Eq. (3.274) Fig. 3.47 Schematic potential energy cur-
ves for species A, B and C.
C
Z trans Z rCot Z vib
C
ΔG 0 = − R T ln A
+ ΔE 00 RT (3.275)
Z trans Z rAot Z vib
A
Z tBrans Z rot
B
Z vAibb

0
C
Z trans Z rCot Z vib
C
e −ΔEE0 / RT
K= A (3.276)
Z trans Z rAot Z vib
A
Z tBrans Z rot
B
Z vBib e +Δn
where Δn is the difference between the number of product and reactant molecules, i.e. ( ). E 00 can
be determined for some reactions from spectroscopic data while for other reactions thermal data are required
to get E 00 This method of evaluating equilibrium constants is especially useful in gathering information on the
efficiency of isotopic substitution reactions.
Further, the equilibrium constant for the most general reaction,
aA + bB cC + dD
is expressed as
Δn
( Z C )c ( Z D )d ⎛ 1 ⎞ ⎛ −ΔE 00 ⎞
Kp = exp ⎜ (3.277)
( Z A )a ( Z B )b ⎜⎝ N A ⎟⎠ ⎝ RT ⎟⎠
0
where Z Z trans 0
Z rot Z vib Z elec Z nuc and Z trans is the translational partition function at 1 atm. Δ
ΔE 00 is the difference in
zero point energies of products and reactants, i.e.
ΔE 00 = cE
Δ cE(0C ) + dE(0D ) − aE
aE(0A ) − bE
bE(0B )
and Δn = (c + d ) − (a + b )
Problem 3.42: The vibrational frequencies (in cm−1) for HD, H2 and D2 are 3817, 4395 and 3118 respectively.
Calculate the equilibrium constant for the exchange reaction
H2 + D2 → 2HD
Solution The exchange reaction is
H2 + D2 → 2HD
The total energy change for this reaction at 0 K is the algebraic sum of the zero point energies
Δ 0
0 = 2E 00 ( HD ) − E 00 ( H 2 ) − E 00 ( D 2 )
hc
= ( 2v −v 2
−v 2
) ( )
2
Thus, 2
Z HD
K e
2
⎡ ⎤
⎢V 2 8
( ) ⎥ ×e
1 − 2 T
h2 HD h
=
2 −1 2π 2
V 2
×V 2
h H h h
⎛8 κ ⎞
( )
1
1− e hcv D T
D h
For each vibrational partition function hn >> T at room temperature and hence Zvib ∼ 1.
Also =2 1
3 2
M HD ⎞ 4I HD T
Thus, K= e
We know that
M = 1
vH cm 1
2 1
3 2
9 4 × 8 − hc (121) / 2 T
Hence K e
8 9
= 3.17 at T = 300 K
This value will not be accurate since we have used the integrated form of the rotational partition function,
which is not a good approximation for H2 at room temperature.
(vii) Specific Heat of Adsorption from Group Frequency Shift The physical adsorption of adsorbate on the
surface of silica causes a frequency shift of the surface silica OH groups. This frequency shift has been found to
vary linearly with the specific heat of adsorption, ΔQad and can be utilised to calculate the heat of adsorption of
adsorbate on a dehydroxylated silica surface Δ ad when the corresponding heat of adsorption on the hydroxy-
lated silica surface ΔH h is known, since
=Δ Δ (3.278)
Problem 3.43: During the physical adsorption of benzene, n-hexane, diethylether, etc., on the surface of
hydroxylated silica surface, it is observed that the frequency shift of the OH vibration for one half surface cover-
age and ΔQad can be expressed by
− −
Δ = 2 cm − (3.279)
If the frequency shift for adsorption on hydroxylated silica surface is 200 cm−1, calculate the specific heat of
adsorption of water on this silica surface
Solution Since Δv = ΔQad + 25
Therefore, for Δv OH = 200 cm −1

175
ΔQad = = 2.80 kcal mole −1
62.5
(viii) Intermedions from Vibrational Frequency of Some Adsorbed Chemical Species The adsorption of
an adsorbate on metal or metal oxide adsorbent surfaces gives rise to IR absorption bands higher or lower than that
of the free adsorbate molecule. This may be due to the difference in the electronic structure of the free adsorbate
and adsorbed molecules. The various species formed due to adsorption of single adsorbate, differ from each other
by an integral number of electrons, i.e.
X X adn ±1 X adn ± 2 X adn ± 3 , etc.
where n is a fraction less than 1.

The vibrational frequencies of some adsorbed CO species on surface of metals or metal oxides is of two types.
In the first type, the number of electrons contained in the adsorbed CO species varies by an integral number
whereas in the other no such relationship exists. The species of the first group are called intermedions and are
assumed to be responsible for chemical reactivity. The second type is known as carbonyls. They are more stable
whereas the intermedions are easily formed and removed. The position of a band due to intermedion is inde-
pendent of the nature of the substrate but dependent only on the cation involved. The fractional part of the non-
integral number of valence electrons associated with the intermedions is termed as polarisation fraction and is
dependent upon ionic charge.
The empirical relationship between the number of valence electrons E and the vibrational frequency v of the
adsorbed CO species on metal surfaces is
[2269.96 ± v (CO)] [12.1182 − E(CO)] = 268.31 cm−1 (3.280)
Problem 3.44: The adsorption of CO on copper (II) oxide and cobalt surfaces gives rise to vibrational bands
at 2173, 2121, 2127 and 2000 cm−1 in case of copper (II) oxide while at 2160, 2091 and 2179 cm−1 in case of
cobalt (Co). Calculate the number of valence electrons in different chemical species of CO which result due to
adsorption and based on these observation classify the species as inermedions and carbonyls.
Solution We know that, [2269.96 ± v (CO)] [12.1182 − E(CO)] = 268.31 cm−1. Thus, for v = 2173 cm−1,
[2269.96 ± 2173] [12.1182 − E(CO)] = 268.31 or E(CO) = 12.1182 − 2.7672 = 9.35.
Similar calculations for other frequencies yield.
E(2121) = 10.32; E(2127) = 10.24; E(2000) = 11.12; E(2160) = 9.68;
E(2091) = 10.62; E(2179) = 9.17
Applying the criteria for intermedions and carbonyls, the classification is given below.
Adsorbent Intermedions Carbonyls
Band Position Number of Band Position Number of

(cm-1) Electrons (cm-1) Electrons
Copper (II) oxide 2173 9.35 2127 10.24

2121 10.32 2000 11.12
Cobalt 2160 9.68 ⎯ ⎯
2179 9.17
2091 10.62 ⎯ ⎯
(ix) Barrier to Internal Rotation from Bandwidth The bandwidth ΓT in the spectra of liquids and solids can
be split into temperature dependent (ΓT − Γ0) and temperature independent (Γ0) components respectively. The
temperature dependent component (ΓT − Γ0) is directly proportional to e −U / RT , where U is the energy barrier to
reorientation by Brownian motion of the molecules of the type XH3 where X = C, N, P, Therefore, we can write
ΓT = Γ 0 + A 0e −U / RT (3.281)
where A0 is the constant.

This expression holds good for molecules having U >1 kcal where rotational modulation of the vibration is not
likely to contribute appreciably to bandwidth. It follows from Eq. (3.281) that
U
( )=− + ln A 0 (3.282)
RT
Therefore, simply by measuring the bandwidth at low temperature to get Γ0 where there is no rotation and
also at some higher temperature to get ΓT one can get U from a plot of ln (ΓT − Γ0) against 1/T. The plot will be a
straight line with
U
slope ( ) = − and intercept (b) ln A 0 .
R
The accuracy of this method depends upon the separation of the bands into rotational and non-rotational com-
ponents from Coriolis constants.
Problem 3.45: The spectra of solid Au2I2 (CH3)4 as CSI disk in the temperature range 90−298 K have been
recorded. The bandwidths of the asymmetric deformation band at 1460 cm−1 are 15 cm−1 at 90 K, 27 cm−1 at
180 K, and 40 cm−1 at 298 K. If, at 90 K, there is no effective rotation of methyl group, calculate the barrier height
of the methyl group in the molecule.
Solution Since there is no rotation of methyl groups at 90 K, Γ0 = 15 cm−1. Then the values ΓT − Γ0 at different
temperature are
1
ΓT Γ0 ( 1
) × 103
T
12 5.55
25 3.35
1
The values of U and A0 are evaluated by plotting ln (ΓT − Γ0) against . The plot is shown in Fig. 3.48. The
slope and intercept of this line are T
slope (m) = − 0.33 × 103 5.0
intercept (b) = 4.35 4.0

According to Eq. (3.282), the slope = − U/R and intercept = ln A0.
Therefore U = 0.33 × 1.987 × 103 = 0.662 kcal mole−1 = 232 cm−1 and
3.0
A0 = 77.47 cm−1.
In(Γ−Γ0)
(x) Enthalpy of H-bonded Adducts from Bandwidth The broad-

2.0
ening of the natural bandwidth may not only be due to Doppler effect
and molecular collisions but also due to weak interactions of the type
1.0
charge transfer and hydrogen bonding. Width of band, when affected
due to these types of interactions, can be correlated with the energet-
0
ics of adducts formed. Thus, if width of XH band where X = O, N, S, 0 1.0 2.0 3.0 4.0 5.0 6.0 7.0
etc., in some molecular systems with electron-loving and hating groups, I/T (x10+3)
changes as a result of formation of some adducts with other molecu- Fig. 3.48 The variation of bandwidth with
lar systems, then based on the experimental data on bandwidth of the temperature.
complex formed and enthalpy of complexation, empirical relations between these two quantities could be framed
by a least square treatment.
Problem 3.46: On the basis of infrared study of H-bonded complexes formed between phenol derivatives
having electron attracting and repelling groups and poly-functional bases, the following equations have been
established by a least square treatment.
(a) −ΔH = 3.66 + 0.070 ΔvOH(r = 0.985) for l-methyl-2-pyrimidone (3.283a)
(b) −ΔH = 9.61 + 0.055 ΔvOH (r = 0.984) for 3-methyl-4-pyrimidone (3.283b)
(c) −ΔH = 9.83 + 0.058 ΔvOH (r = 0.994) for 1,3-dimethyluracil (3.283c)

and
(d) −ΔH = −13.78 + 0.091 ΔvOH (r = 0.989) for pyrimidine, where units of ΔH are kJ mole−1
and of ΔvOH are cm−1 (3.283d)
By making use of these relations, calculate the enthalpy of H-bonded adducts formed between the following
systems, given the values of ΔvOH.
(i) Pyrimidine + 4-methoxyphenol (ΔvOH = 373)
(ii) 1,3-dimethyluracil + 3-flurophenol (ΔvOH = 253)
(iii) l-methyl-2-pyrimidone + 4-chlorophenol (ΔvOH = 33)
(iv) 3-methyl-4-pyrimidone + 3-nitrophenol (ΔvOH = 316)
Solution By substituting the values of ΔvOH (cm−1) of the adducts formed between the phenol derivatives and the
poly functional bases, in the respective equations, the following values of enthalpy of complexation are obtained.
Systems −ΔH (kcal mole−1)

Pyrimidine + 4-methoxyphenol 4.82
1,3-dimethyluracil + 3-flurophenol 5.86

l-methyl-2-pyrimidone + 4-chlorophenol 6.40
3-methyl-4-pyrimidone+ 3-nitrophenol 6.45
(xi) Mean Velocity of S-adatom from Bandwidth Chemisorption of hydrogen on metal surfaces is believed
to involve r and S-type adsorption. An S-type adatom may dissipate energy as a result of transfer from the excited
vibrational state to the ground state through collisions with the metal ions. The mean velocity nm of an S-adatom
is given by
1
⎛ 2κT ⎞ 2
vm = ⎜ (3.284)
⎝ π m H ⎟⎠
and if the mean free path of this adatom is Y Å, then the frequency f of its collision with the metal ion is
1
⎛ 2κT ⎞ 2
⎜⎝ π m ⎟⎠ (3.285)
f = H
Y
where mH = mass of hydrogen atom. The decay constant would then be
f
τ= (3.286)
c
where c is the velocity of light.
The bandwidth Δv1/2 of the adsorption band which is a measure of its dissipation energy is then given by
τ
Δv 1/ 2 = (3.287)
2
Thus, from observed value of Δv1/2, one can calculate mean velocity of an S-adatom.
Problem 3.47: During the chemisorption of hydrogen on Pt it is observed that there are two types r and S
of adatoms. If the mean free path of a S-adatom is 3 Å, calculate the temperature for Δv1/2 = 100 cm−1, κ = 1.380
× 10−16erg deg−1 and mH = 1.673 × 10−24g.
τ
Solution Since Δv 1/ 2 =
2
Therefore, τ = Δn1/2 × 2 = 100 × 2 = 200 cm−1,
f = c × τ = 200 × 3 × 1010 = 6 × l012 s−1
and vm = f × Y = 6 × 1012 × 3 × 10−8 = 18 × 104 cm s−1.

1
⎛ 2κT ⎞ 2
Since vm = ⎜
⎝ π m H ⎟⎠
( )
2
v 2 πmH × × 3.142 × 1.673 × 10 −24
Therefore, T = m = K = 617K
2κ 2 × 1.380 × 10 −16
or T = (617−273) = 344°C
3.16.3 Biological Applications

Infrared absorption spectroscopy in spite of its limitations due to absorption of water has been proved to be a
potential tool in conformational analysis of model polypeptides and proteins and of biopolymers, i.e. a-helical
and/or unordered conformations can be discriminated from b-structures. This technique also imparts information
about the interaction of lipid and protein components in cellular membranes. The technique of infrared dichroism
has also played a chief role in the conformational analysis of polypeptides and proteins, especially when ordi-
nary IR spectroscopy fails to distinguish between the various conformations. Before discussing the applications
in detail, the various factors affecting the conformation of model polypeptides and proteins demand attention.
The IR absorption bands of polypeptides and proteins are not only delocalised but are sensitive to the folding
of the peptide backbone and to H-bonding with other peptide linkages, i.e. conformation or 20 structure. Since
peptide backbone of polypeptides and proteins often includes arrays which can be described by symmetry space
groups, one can differentiate between the various conformations. Depending upon the nature of the media, pH,
temperature, segment length, ionic strength, etc., polypeptides and proteins may exist in one particular/specific
conformation or mixture of conformations, e.g. at low ionic strength, and near neutral pH, poly L-lysine exists-in
an unordered state but above pH 11, the a-helical form becomes predominant, since above pH 11, the electrostatic
repulsion between the side chains is small. At 55°C and pH 11.3, the polymer has b-pleated sheet structure. At
neutral pH and low ionic strength, poly L-aspartate and poly L-glutamate have unordered structure but exhibit
a-helical conformation at acid pH. It is to be noted that it is not always possible to shift all polymers from one
conformation to another by change of environment.
Plane polarised IR radiation is absorbed preferentially by chromophores whose transition dipole moments are
parallel to the polarisation plane. The absorption intensity of an infrared band depends upon the orientation of the
dipole moment of the chromophore relative to the incident beam of IR radiation. In the case of C == O stretching
vibration of esters, the dipole lies along the C == O axis. Hence the orientation of C == O group can be determined
by infrared dichroism.
Many polymers including polypeptides exhibit infrared dichroism. Consequently, the infrared absorption band
of a given vibration splits into two components of different frequency, preferentially absorbing radiation polarised
parallel and perpendicular respectively to the molecular axis. The selection rules for infrared dichroism as worked
out from group theory are valid for ordered polymers, helices in particular. Globular molecules do not exhibit
infrared dichroism since the molecules cannot be oriented properly with respect to the plane of polarisation of the
incident radiation.
In an infinite, single-stranded a-helix, coupling along the backbone and through hydrogen bond splits the
intensity of the normal vibration into two components parallel and perpendicular to the helical axis. In oriented
strongly coupled systems, it is found that
I⏐⏐
≈ cot 2 δ (3.288)
I⊥
where I|| and I⊥ are the intensity decrements for parallel and perpendicularly polarised radiation, to the screw axis
of the polymer and δ is the angle between the transition dipole moment of the vibration and screw axis of the
helix.
For a most general peptide chain where both inter- and intra-molecular hydrogen bonds exist, the vibration
frequency is given by
v (δ δ′ ) ν 0 ∑ D cos(sδ ) ∑ D ′ cos(s′δ′ )
s
s
s′
s (3.289)
where v0 is the unperturbed frequency and is found in randomly coiled H-bonded forms. Primed and unprimed
terms arise from inter- and intra-molecular interactions.
When Ds ′ = 0, then the frequency ν of a mode involving the coupled vicinal vibrators is given by
v (δ ) 0 ∑D
s
s (sδ ) (3.290)
Here d is the phase angle between vicinal vibrators and Ds = (Bs –As) v0; Bs and As, are the potential and kinetic
interaction factors between s neighbours.
For an infinite helical chain, δ can take on any value but only two yield infrared absorption: δ = 0, and δ = χ,
where χ is the angle between two consecutive groups relative to helix axis (2p/3.6 in the case a-helices). Since
hydrogen bonds here are formed between every third residue, D1 and D3 are dominant, but D2 is also significant.
Two frequencies can be observed:
v(0) = v(||) = v0 + Dl + D2 + D3
v(χ) = v(⊥) = v0 + D1 cos χ + D2 cos (2χ) + D3 (cos 3χ) (3.291)

Here, ν(||) and ν(⊥) are polarised parallel and perpendicular to the helix axis respectively.
(a) Parallel b-conformation In this conformation, the two adjacent peptide linkages in a chain can move ei-
ther in phase (δ = 0) or out of phase (δ = p). When groups in adjacent chains, coupled by H bonds, move in phase,
the vibration is infrared active. Two modes of vibration result with frequencies
v(0, 0) = vo+ D1 + D1′ (||) (3.292)
v(p, 0) = vo − Dl + D1′ (⊥)

Primed and unprimed terms arise from the inter- and intra-molecular interaction of groups in adjacent chains,
and the polarisation is relative to the backbone axis.
(b) Antiparallel b-conformation The spectrum arises from four peptide linkages, involving portions of two
arms of an antiparallel loop or two antiparallel chains. The frequencies of the resulting modes of vibration are
v(0, p) = vo+ D1 − D1′ (||)
v(p, 0) = v0 − D1 + D1′ (⊥, in plane) (3.293)
v(p, p) = v0 − D1 − D1′ (⊥, out of plane)
v(0, 0) = inactive
(c) Unordered State For an unordered state, the vibrational interactions between peptide groups average to
zero and only v0 is observed.
IR absorption band characteristic of a particular conformation of polypeptides and proteins are the
amide bands. In other words, amide bands can be fixed as the criteria to distinguish between the various
Table 3.18 IR bands associated with the peptide linkage.
Frequencies (cm-1) Assignment
3300 ⎤ NH str, 2 × amide II in Fermi resonance

⎥
3100 ⎥⎦
1650 Amide I (C == O str, 80%, C—N str, 10%, NH bend 10%)

1560 Amide II (C—N str, 40%, NH bend 60%)
1300 Amide III (C == O str, 10%, C—N str, 30%, NH bend 30%,
O == C—N bend 10%, other 20%)
625 Amide IV (O == C—N bend 40%, other 60%)

725 Amide V(NH bend, O/P)
600 Amide VI (C == O bend, O/P)

200 Amide VII (C—N twist, O/P)
str—stretch, O/P—out of plane
Table 3.19 Amide bands (in cm–1) In polyglycine-I and polyglycine-II.
Assignment Polyglycine-I Palyglycine-II

Amide I (C == O str) 1685 and 1630 1641
Amide II (NH bend) 1524 1558

Amide III (C—N str) 1297 1309
Amide IV (O == C—N bend) 589⎤

⎥
Amide V (NH out-of-plane bend) 710 ⎦ 740
Amide VI (C == O out-of-plane bend) 628 and 614 701 and 573
Amide VII (C—N twist, out-of-plane) 217 365
conformations. IR bands associated with a peptide linkage as well as in polyglycine-I and polyglycine-II are listed
in Tables 3.18 and 3.19 respectively.
In addition, the assignment of the amide bands corresponding to the different conformations in the model
systems needs to be described here for use to discriminate between the conformations in the biomembranes. In
case of a-helical poly-benzyl-L-glutamate, the amide I and II bands appear at 1656 and 1550 cm−1 respectively
while at 1630 and 1525 cm−1 in the b-structured polymer. a- and b-helical polymer-L-lysine exhibit amide-I bands
at 1652 and 1636 cm−1 respectively. The energy of the amide-I band may also be used to differentiate between
a-helical or unordered conformations of a given polypeptide under certain conditions, e.g. solid film IR spectra
of poly-L-lysine in the a-helical and unordered structures have amide-I bands at 1652 and 1657 cm−1 respectively.
The IR frequencies of the amide-I and II bands corresponding to different conformations of poly-L-lysine have
been summarised in Table 3.20.
Further, since the energy of the amide-I bands in the a-helical and unordered structures may be very close,
the position of band varies from one peptide to another and band broadening due to side chains may occur, the
amide-I band cannot always be used to distinguish between a-helical and unordered conformations until the poly-
peptide exhibits infrared dichroism, e.g. the amide-I band in the solid film spectra of a-helical-poly-L-glutamic
acid and poly-g-benzyl-L-glutamate appears at 1652 and 1656 cm−1 respectively. The latter in chloroform exhibits
Table 3.20 IR frequencies (cm−1) of amide-I and II bands of poly-L-lysine.
Helix Antiparallel-b Unordered

1516w ⎤ 1530s, II 1535, II
1535s ⎥⎦
, II
1650vs ⎤ 1636s ⎤ 1656, I

1652m ⎥⎦ , I 1685m ⎥⎦
,I
amide-I band between 1651−1656 cm−1.

The amide-II band is not very helpful to differentiate between the various conformations since a-helical and
b-structures both give bands near 1510 and 1540 cm−1. The vibrational modes of COO−-group in proteins may
also interfere since their energy lies in the 1550−1560 cm−1 range. Nevertheless, the amide-II band may be used
for conformational analysis in cases where polypeptides and proteins exhibit infrared dichroism in the said region.
In case of poly-L-glutamide at pH 7.0, COO− absorption even broadens the amide-I band.
Amide-V band proved to be very successful in differentiating between a-helical and unordered structures,
e.g. poly-g-methyl-L-glutamate in the a-helical, unordered and b-structures exhibit three bands at 600, 650 and
690 cm−1 in the amide-V region. The infrared spectra of lysozyme also give three amide bands at 600, 650 and
690 cm−1. Hence, simply by comparing the spectra of lysozyme with that of poly-γ-methyl-L-glutamate in the
amide-V region, it can safely be concluded that lysozyme is a mixture of a-helical, b and unordered structures.
Conformationally dependent bands also fall in the far infrared region of the spectra of the polypeptides and pro-
teins, hence this region may also be exploited for conformational analysis, e.g. a-helical-poly-L-alamine exhibits
four bands at 190, 167, 120 and 90 cm-1 while its b-form shows only two bands at 442 and 247 cm−1. The bands
are due to H-bond stretching and bending and the torsional oscillations about Ca—C and Ca—N axes whereas the
band at 247 cm−1 of b-structure may be due to CO—NH torsional oscillation.
Poly-b-benzyl-L-aspartate (insoluble in aqueous buffers) unlike most of the model polypeptides consists of
L-amino acids and forms left- rather than right-handed helices. However, when the polymer spread at an air–
water interface and is studied by infrared dichroism, it exhibits bands; at 1740 cm−1 (ester C = = O), 1658 cm−1
(amide-I), 1552 cm−1 (amide-II) which are not characteristic of the left-handed a-helical form but conforms to the
right-handed helices. These observations can be supplemented by the parallel polarisation of the C−−O band at
1168 cm−1 since the left-handed form shows perpendicular dichroism at this frequency. The IR dichroism results
also find support from the electron diffraction studies of the polymer spreaded at the air–water interface.
The information gathered from model systems can now be utilised for the conformational analysis of biomem-
branes. IR spectra of erythrocyte membrane dried from phosphate buffer of pH = 7.4 exhibit band at 1652 cm−1
but no absorption has been observed in the 1630−1640 cm−1 range. From this it can safely be inferred that the pro-
tein in the membrane exists in the a-helical and/or unordered conformation since the band observed at 1652 cm−1
is characteristic of these conformations. IR spectra of some of the biomembranes of animals and plants exhibit
strong absorption due to a-helical and/or unordered polypeptide but also a shoulder near 1630−1640 cm−1. The
appearance of band near 1635 cm−1 reveals that these membranes contain a significant amount of their peptide
linkages in a b-conformation also. The appearance of an additional band near 1690 cm−1 in the mitochondrial
membranes indicates the presence of antiparallel b-conformation in this membrane.
The structure of membrane proteins depends upon their interactions with membrane lipids. Thus, the IR bands
characteristic of membrane lipids may be utilised for the understanding of lipid-protein interactions. However,
all the IR bands of membrane lipids are hard to be characterised due to lipid polymorphism. Nevertheless, the
assignment of some of the major bands of fatty acid chains and phosphatides are listed in Table 3.21. The model
data can be exploited to study the lipid-protein interactions as is evident from the following examples. The aque-
ous dispersions of membrane lipids show a band due to CH2 rocking mode at 720 cm−1. However, this band does
not appear in the film spectra of plasma membranes of human erythrocytes indicating that in these membranes,
the planar, all trans configuration of hydrocarbon chains is not favoured. This may be due to apolar association
between membrane lipids and proteins that makes other configurations energetically more probable.
In some of the biomembranes, the intensity ratio of CH2 asymmetric stretching (2930 cm−1) and symmetric
stretching (2855 cm−1) modes of vibration respectively have been found to be 0.3 in case of whole membrane, 0.1
in case of lipid extract and 1.4 in the extracted residue, at liquid nitrogen temperature. The band at 720 cm−1 due
to CH2 rock has been observed only in the lipid extracts. These results can be fixed as criteria for the close range,
nonpolar interactions between apolar protein side chains and phospholipid hydrocarbon chains.
Table 3.21 Assignment of some major infrared bands (cm−1) of the fatty acid chains of phospholipids.
Expected Assignment Observed Remarks

Double near 1640 cm−1 in condensed lipids, due to
1460 CH2 bend and str 1640 interaction between fatty acid chains
1380–1180 CH2 wag
1300–1170 CH2 twist 1200 Ill defined

1150–870 C—C str and CH rock 1150–870 Not precisely localised
∼3000 C = C—H str
1580–1650 C = C str
690–980 C = C—H O/P bend
970 CH bend(ethylenic)
2926 CH2 asym str. (O/P)

2853 CH2 sym str. (I/P)
720 CH2 − rock 720 *
H-bonding shifts the P = O str to lower frequen-

P = O str 1350−1250 cies, e.g. P = O, 1227 in phosphatidyl-ethanolamine;
In phosphatidyl-serine P = O, 1220-1180
P−−O−−C str 980−1050
NH 3+ 2100 Weak band in dimyristoyl cephalin
N−−H str (lipid associated) 3330 In some lipoproteins
OH str 3600 Free cholesterol
OH str 3330 Cholesterol ester
O/P—out of plane, I/P— in plane, str—stretch, sym—symmetric, asym—asymmetric.

*This band reflects interaction between adjacent hydrocarbon chains. This band arises only when at least 4 successive carbon atoms in a hydrocarbon
chain are arranged in a planar all trans configuration. When interactions between chains are strong as in solids, this vibration yields a doublet separated by
6 cm−1. In liquid state, the 700–1350 cm−1 region of long chain hydrocarbons becomes less distinctive.
The shoulder at 1711 cm−1 due to lipid carbonyl in the IR spectra of erythrocyte membranes does not appear in
the lipid extracts. This may be due to lipid protein interaction.
The IR spectra of red blood cells membrane and its lipid extract exhibit band due to P = 0 stretching mode
of vibration at 1225 cm−1. This low value as compared to the expected value of 1250−1350 cm-1 may be due to
H-bonding. From these results, it is suggested that in these types of membranes, the phosphatide head groups lie
in water and are not ionically bonded to membrane proteins.
The IR absorption spectra of lipid free erythrocyte membranes exhibit a major band at 1652 cm−1 and a shoul-
der at 1628 cm−1. This may be taken as evidence for the transformation of some structures to b-conformation upon
lipid removal.
Infrared spectroscopy can also be used as an effective tool to observe ligands such as CO, NO, O2, CN−, 3−
and N2O directly in metalloproteins in vivo (intact tissue) or in vitro (isolated tissue) to obtain useful information
about the metal, the ligand and their environments. This is being elaborated here with the following examples:
The absorption band of free CO in the gaseous phase is observed at 2237 cm−1 (Δv1/2 = 100 cm−1) while in the
region 1905–1980 cm−1 (Δv1/2 = 4−30 cm−1) in hemeproteins. C—O stretch in carbonyl (HbA 12CI6O) appears at
1951 cm−1 whereas at 1907 cm−1 in case of HbA 13Cl6O and HbA 12C18O. The magnitude of the observed isotopic
shift in each case is 44 cm−1 and conforms to the shift expected for a simple diatomic molecule as computed from
the relationship
k
v = 1303.16
μm
Similarly, the frequency shifts observed between 14N16O (1615 cm−1)/and 15N 16O (1587 cm−1); 16O2(1155,1107 cm−1)
and 18O2(1096,1064 cm−1) in hemoglobin, and −l4N 14N I4N−(2046cm−1), −15N 14N 14N− (2040 cm−1), −14N l4N
15 −
N (2028 cm−1) in high-spin myoglobin provide evidence for the assignment of bands due to bound NO, O2 and
N 3− in nitrosyl, oxygenyl and azide hemeproteins respectively. In case of low-spin myoglobin, the respective azide
bands are observed at 2030, 2021 and 2004 cm−1. The two bands at 2046 and 2040 cm−1 due to azides are of equal
intensity because there is an equal probability of either the −14N end or −15N end bounding to Fe(III). However,
there is no evidence to suggest that N2O is serving as a ligand bound to different or any metal site since when N2O
is administered to brain tissue, the two bands appearing at 2230 and 2217 cm−1 fall in the infrared band frequency
region for N2O in different solvents with varying polarity.
The ligand infrared band frequency is affected by oxidation state of metal also. Cyanide bound to Fe(II) of
reduced cytochrome C oxidase is observed at 2060 cm−1 while at 2152 cm−1 when bound to Fe(III) of the oxidised
form of cytochrome C oxidase. The difference between the CN stretching frequencies in the reduced and oxidised
form of the protein may be because of greater back p-donation from Fe(II) than from Fe(III). Similarly, 15N 16O
when bound to Fe(II) and Fe(III) of hemeproteins exhibits N—O bands at 1587 and 1865 cm−1, respectively. Thus,
it is quite evident that IR spectra can safely be employed to differentiate between low and high oxidation states of
metals in proteins.
Infrared spectra have also revealed multiple species of hemoglobin carbonyls and myoglobin carbonyls and
azides that have not been detected crystallographically. These multiple carbonyls could also not be differentiated
by 13C-NMR spectra because the interconversion between the species is too fast for the NMR time-scale.
Problem 3.48: Calculate the B species (only Raman active) for amide-I and II bands in cyclictetraglycine
(CTG) if the A and E species (both are Raman and IR active) of amide-I band are observed at 1685 and 1640 cm−1,
and of amide-II band at 1575 and 1538 cm−1 respectively.
In CTG, there are four residues and therefore, δ = 0, p/2 or p. Take Ds' to be zero in CTG.
Solution For A(δ = 0), B(δ = p) and E(δ = p/2), species we may write from Eq. (3.290) the following equations
respectively
v(0) = v0 + 2D1 + D2 (3.294)
v(p) = v0 − 2D1 + D2 (3.295)
v(p/2) = v0 − D2 (3.296)
The unperturbed frequency v0 can be taken from nylon-66 because here the amide groups are sepa-
rated from each other through methylene groups. The amide-I and II bands in nylon-66 have been
observed at 1640 and 1540 cm−1 respectively. Using these frequencies together with the unperturbed frequencies
for amide-I and II bands from nylon-66, Eqs (3.294) and (3.296) give
D1 = 22.5 cm −1
amide I
−1
D 2 = 0.0 cm
D1 = 16.5 cm −1
amide II
D 2 = 2.0 cm −1
Using these values of D1and D2 for amide-I and II bands, the B species for amide-I and amide-II bands are
calculated at 1595 and 1509 cm−1 from Eq. (3.295). The corresponding values obtained from normal coordinate
analysis are 1713 cm−1 and 1534 cm−1, respectively.
3.17 FAR IR ABSORPTION SPECTROSCOPY

Molecules made up of only light atoms absorb in the far IR region. Pure rotational absorption of gases having
permanent dipole moment such as H2O, O3, HCl and AsH3 is observed in this region and can be used for the iden-
tification of such molecules. Moreover, as mentioned earlier, this technique can be used for the identification of
both inorganic and metalo-organic compounds since their vibrational frequencies fall below 670 cm−1.
3.18 NEAR IR ABSORPTION SPECTROSCOPY

Sources of radiation in near IR spectroscopy are tungsten–halogen lamps with quartz windows. Quartz or fused
silica cells with path lengths between 0.1−10 cm and are transparent up to 300 nm are used for absorption
measurements in this region. PbS photoconductors are used as detectors. Commercial spectro-photometers are
available to record the near IR spectra in the 180−2500 nm wavelength range. It is to be noted that in near IR
spectroscopy, the wavelength on the horizontal axis is reported in nanometers or micrometers in contrast to
mid-IR spectroscopy in which the abscissa is wave number in units of cm−1. Commonly used solvents in this
region are, CC14, CS2, CHC13, dioxane, heptane, C6H6, CH3CN and DMSO. However, out of these solvents only
CC14 and CS2 are transparent in the entire near IR region.
Further, in addition to the conventional IR absorption spectroscopy, the other infrared techniques that deserve
proper place in IR spectroscopy and are capable of solving numerous qualitative and quantitative problems which
are difficult to handle or not at all soluble by other spectroscopic methods are explained below.
3.19 MID IR REFLECTANCE SPECTROSCOPY (MIRS)

Reflectance Spectroscopy (RS) deals with the study of the spectral composition of surface reflected radiation with
respect to its angularly dependent intensity and the composition of the incident radiation. MIRS reflection spec-
troscopy is useful especially for solid samples that are difficult to handle such as polymer films and fibers, foods,
rubbers, agriculture products, etc. Mid-IR reflection spectra are similar in general appearance but not identical to
the corresponding absorption spectra. It provides the same information as do the absorption spectra. Four types of
reflection, viz., (a) regular (specular), (b) diffuse, (c) internal, and (d) attenuated total reflection, are dealt is RS.
(a) Specular Reflectance Spectroscopy (SRS) Specular reflection occurs from a smooth surface. Here the
angle of incident equals the angle of reflection. In case of infrared absorbing surfaces, the relative intensity of
reflection is less for absorbed wavelengths than for unabsorbed wavelengths. Thus, the specular reflectance spec-
trum consists of a plot of reflectance, R, i.e. the fraction of the energy of incident radiation that is reflected, on the
vertical axis against the wave number on the horizontal axis. SRS is used to examine and characterise surfaces,
virgin and coated solids. However, scope of SRS is limited as compared to diffuse and total reflection spectros-
copy.
(b) Diffuse Reflectance Spectroscopy (DRS) Diffuse reflection takes place from matle surfaces, i.e. ran-
domly oriented plane surfaces of a finely divided powder. Here, it is assumed that specular reflection occurs at
each of these plane surfaces and the intensity of the radiation reflected in all directions is isotropic, i.e. it is as-
sumed to be independent of the viewing angle. The relative reflectance intensity for a powder sample is given by
Kubelka–Munk function which is written as
( R′ ) K
F ( R′ ) = = (3.297)
2R ′∞ S
where R ′∞ is the ratio of the intensity of the reflected radiation by the sample to that of a non-absorbing standard
such as finely ground potassium chloride. The quantity, K is the molar absorption coefficient of the analyte and
S is the scattering coefficient. For a dilute sample, K is related to the molar decadic coefficient e and the molar
concentration of the analyte by the expression
K = 2.303 eC
Diffuse Reflection Spectra (DRS), thus, consists of a curve obtained by plotting F( R ′ ) versus wave number.
Diffuse Reflectance IR Fourier Transform Spectroscopy (DRIFTS) is particularly useful to obtain directly the
spectra of powdered samples with a minimum of sample preparation.
(a) (b)
1.00 0.30
Absorbance
⬘)
F(R∞
0.20
0.75
0.10
0.50
2000 1500 1000 2000 1500 1000

Wave number (cm−1) Wave number (cm−1)
Fig. 3.49 Comparison of the conventional infrared absorption spectrum (a) for carbazole with its diffuse reflectance spec-
trum (b).
Unlike in transmittance spectroscopy, the DRIFT spectrum is independent of the thickness of the layer, provided
it exceeds 5m. DRIFT spectrum of 5 per cent finely ground mixture of carbazole in KCl is compared with the
coventional IR absorption spectrum of carbazole in KBr pellet in Fig. 3.49.
The relative intensity of the bands are different but the position of the bands are the same. Water soluble
and water containing substances that are difficult to examine by IR transmittance measurements can also easily
be investigated by DRIFTS, e.g. two phase systems such as suspensions, emulsions or colloids, etc. Thus the
biochemical problems, i.e. spectra of polypeptides, amino acids, etc., can also be studied by this technique.
(c) Internal Reflection or Attenuated Total Reflectance Spectroscopy

(IRS/ATRS) Refractive index of material depends upon the wavelength of
incident radiation. In the absence of strong absorption bands, the refractive in-
dex decreases as the wavelength increases. In the region of strong absorptions,
very sharp changes in refractive index take place as shown in Fig. 3.50. This
Refractive index
phenomenon is of importance in the practical application of internal reflection
spectroscopy.
When a beam of radiation passes from high-refractive-index phase to low-
refractive-index phase, reflection occurs. If the sample absorbs radiation and the
angle of incidence exceeds a certain definite or critical value (θc), total reflection
takes place. Even when the angle of incidence exceeds the critical value, some
radiation penetrates the second phase, i.e. the less dense medium before the Frequency
reflection occurs and the penetrating radiation is called the evanscent radiation.
The depth of penetration is of the order of a wavelength of radiation and depends Fig. 3.50 The change in refractive
index of a material in the neighbour-
upon the refractive indices of both the phases, angle of incidence and wavelength hood of an absorption band. The solid
of incident radiation, l. Mathematically, the depth of penetration, also called sam- line represents the refractive index, the
pling depth, is described as follows. dashed line the absorption.
λ1
o =
Depth of penetration (3.298)
(
2p sin θ − 2
21)
2 1/ 2
Here l1 = l/n1 (cm), q is the angle of incidence with respect to the

surface normal, and n21 is the ratio of the refractive indices of the
sample (n2) and the Internal Reflection Element (IRE) (n1). From the
expression (3.298), it is clear that (i) penetration is greater at longer
wavelengths for the mid-IR region (2.5−25 mm), that means a factor
Depth of penetration
of 10 difference in sampling depth across the spectrum affects its pho-

tometry as compared to transmission; (ii) as q gets larger, the sampling
depth is smaller, and to the limit of grazing incidence, the sampling
depth approaches 0.1l; and (iii) for a fixed q, penetration is greater
as n21 approaches unity, i.e. where the refractive indices are matched.
The behaviour of depth of penetration which is of the order of few
micrometers with angle of incidence is shown in Fig. 3.51. From the 30⬚ 45⬚ 60⬚
behaviour it is inferred that incident angles of 30°, 45° and 60° can Angle of incidence
be chosen for practical purposes. When the low refractive index mate- Fig. 3.51 The behaviour of depth of penetration
rial absorbs the evanscent radiation, a fraction of radiation energy with the angle of incidence.
transferred is lost and the total reflection is attenuated Sample

at wavelength of absorption bands. This phenomenon
is called attenuated total reflectance (ATR). Mostly,
the ATR technique is restricted to plotting or directly
recording the apparent extinctions and/or transmit-
tances of the sample under investigation as a function Sample
of wavelength. Attenuated total reflectance spectrum
is similar but not identical to ordinary absorption Fig. 3.52 Multiple Internal reflection effect.
or transmission spectrum. Though same bands are
obtained, their relative intensities are different. Thus, the method of ATR provides a valuable tool for examin-
ing materials that are too opaque for transmission methods. A series of reflection is obtained by placing two
samples on opposite sides of an internal reflection transparent element with high refractive index as shown in
Fig. 3.52. Mixed crystal of thallium bromide/ thallium iodide is frequently used as an internal reflection ele-
ment. Plates of germanium and zinc selenide are also used as elements. When the refractive index of reflector
material is lower than the refractive index recorded at the peak maximum (Fig. 3.50) then anomalous band
shapes and band broadening are observed. The preparation of samples is not difficult in internal reflection
spectroscopy. For liquid samples, ATR crystal can be
dipped into a liquid. Water solutions can also be stud-
100
ied if the crystal is insoluble in water. Cells for liquid
(a)
samples are also available. With plastic materials such
as many resins or rubber and also threads, yarns, fab- 80
Per cent transmittance
rics, solids of limited solubility, films, pastes, powders,

etc., the optical contact between sample and reflecting 60
surface is made simply by pressing the sample and the
reflecting surface together. In case of hard materials, 40
(b)
the optical contact can be improved with a few drops
of a liquid of high refractive index such as CS2. Even
if CS2 evaporates, optical contact is still maintained. 20
The transmission spectrum of solid epoxy resin in KBr

is compared with the ATR spectrum of the said resin 0
1300 1200 1100 1000 900 800 700
in Fig. 3.53. Wave number (cm−1)
The method of internal reflection IR spectroscopy Fig. 3.53 Comparison of transmission spectrum of a solid epox-
is more suitable than DRIFTS for studying adsorp- ide resin according to the KBr method (a) with the attenuated total
tion of a particularly oriented crystal surface. Even in reflection spectrum of a solid epoxide resin (b).
case of mono molecular layers or only partially covered
surfaces, multiple reflection provides large extinctions
because of which qualitative analysis of surface layers can be carried out, e.g. surface spectrum of a polished silicon
plate that is exposed to atmosphere exhibits two bands at 3448 and 2135 cm−1 for angle of incidence = 45° and 165-
fold reflection (Fig. 3.54). They are assigned to —OH and —CH stretching vibrations of the adsorbed molecules
respectively. Broad band centred at 2060 cm−1 is observed during the chemisorption of hydrogen on a single crystal
of silicon. This is assigned to Si—H stretching vibration since SiH4 exhibits a Si—H stretching vibration at 2188
cm−1 also.
The surface reactions such as oxidation, reduction, the formation of intermediates (catalysis), electrode surface
reactions, etc., can also be investigated by this method.
Since semiconductors have high refractive indices and are 90
Per cent transmittance
also transparent to IR radiation, they can be studied by IRS.

70
Moreover, the free electrons in the semiconductor barriers
can absorb IR radiation. Thus, it is possible to measure the 50
number of free carriers and its change by irradiation as a
function of wavelength. Gas chromatographic fractions 30
O H
can also be studied by this method. The hot gas conducted 10
C H
directly through a capillary cell cooled from the side by 0
2500 5000 10000
means of the Peltier effect so that the fraction concerned Wave number (cm−1)
deposits on the reflector. Similarly the decomposition prod- Fig. 3.54 Surface spectrum of a polished silicon plate by
ucts of pyrolysis can be trapped for analysis by IRS. means of internal reflection.
3.20 NEAR IR REFLECTANCE SPECTROSCOPY (NIRS)

The first overtone of O—H and carbonyl stretching vibrational band absorbs near 7100 and in the region 3300−
3600 cm−1 respectively. So the absorption band near 7100 cm−1 can be used for the quantitative analysis of compounds
containing O—H groups, i.e. phenols, alcohols, organic acids, and hydrogen peroxides while the region in the 3300−
3600 cm−1 range can be used for the determination of esters, ketones and carboxylic acids. Water in glycerol, organic
films, hydrazine and fuming nitric acid can also be determined by NIRS. NIRS can also be used for the characterisa-
tion and determination of primary and secondary amines in the presence
of tertiary amines in the mixtures since the former amines have overlap-
ping absorption bands in the 3300−10,000 cm−1 region because of various
N—H stretching vibrations and their overtones. Tertiary amines have no 1940
1.0
such bands in this region. Primary amines absorb in the region of combi-
Log (1/R)
nation band of N—H stretching vibration at about 5000 cm−1. So primary 0.7
amines can be directly determined at this wave number. 2100
Near IR reflectance spectroscopy is also used for the quantitative esti- 0.4
mation of components of a finely ground solids, e.g. determination of 0.1

protein, moisture, starch, oil, lipids and cellulose in agricultural prod- 1400 1800 2200 2600
Wavelength (nm)
ucts such as grains and oilseeds. NIR spectrum is obtained by plotting
log (1/R) where R is the ratio of intensity of reflected radiation from the Fig. 3.55 Diffuse-reflectance spectrum of a
sample of wheat.
sample to the reflectance from a standard reflector as a function of wave-
length in nm. Finely ground solid sample is irradiated with one or more narrow bands of radiation having wavelengths
in the range 4000−10,000 cm−1. The diffuse reflectance spectrum of a sample of wheat that depends on the composition
of analyte is shown in Fig. 3.55. MgO or BaSO4 is used as a standard reflector in this case. The reflectance band at
about 1940 nm is due to water and can be used for the determination of moisture in the sample. The other signal
at about 2100 nm is due to the overlapping of signals of starch and protein. Thus, by making the measurements at
two wavelengths in this region, the amount of each of these components can be determined. Less time and sample
preparation are the major advantages of near IR spectroscopy.
3.21 PHOTOACOUSTIC OR OPTOACOUSTIC IR SPECTROSCOPY

(PAIS/OAIS)
The technique of PhotoAcoustic Spectroscopy (PAS) utilises the transformation of radiation energy by gases,
liquids and solids into acoustic energy. So, when there is absorption by the sample at a particular wavelength or
wave number, a sound is produced, called acoustic signal. PAS is not a new technique. The first person to observe
the photoacoustic effect (PA) was J Tyndall who, in 1881, made
Lens
a detailed study of the sound produced by gases when chopped Sun Lens Sample
light was made to fall on them. He attributed the effect to source Lamp black
the periodic heating of the gas by absorption of IR radiation. Chopper Prism
Alexander Graham Bell, in 1881, showed that sonorousness (monochromator) NO
2
Ear
under the influence of intermittent light is property common (Microphone)
to all matter and even visible light could produce PA effect.
He observed that loudest sounds are produced from substances Fig. 3.56 Bell’s ‘spectrophone’.
in a loose, porous, spongy condition and from those that have
darkest or most absorbent colours. Lamp black and soot, or
coloured gases such as NO2 gave sounds which were at times painful to the ear and could even be heard across the
room. The apparatus used by him called Bell’s spectrophone is shown in Fig. 3.56.
It consists of a light source (the sun), a chopper (a wheel with holes driven by a motor, generally a man), a mono-
chromator (the prism) and a sensitive pressure transducer (the ear). The sample to be studied is placed on the front of
the eye-piece of the spectroscope. The walls on the inside of the spectroscope are lined with lamp black and NO2 is
used as the gas to fill the inside of the spectroscope. This is to ensure maximum noise of photoacoustic signal. Only
the transmitted light produces a sound so that when there is absorption at a particular wavelength there will not be any
acoustic signal. One is thus able to hear bands of ‘sound and silence’ compared to the bands of ‘light and darkness’
in the conventional spectroscopic methods. Thus, the acoustic signal is useful noise unlike the usual noise which is
generally wasteful such as the noise produced by creaking wheels of a bullock-cart or the squeaking of a door.
According to modem analysis of PA effect, in case of a gas, the energy absorbed is converted entirely or
partially into kinetic energy of the gas molecules, thereby giving rise to pressure fluctuations that are detected as
audible sound. In case of solids, the radiation absorbed is converted in part or in whole into heat by non-radiative
de-excitation processes in time less than 10−6 s. This resultant periodic heat flows from the solid absorber to the
surrounding photoacoustic material, i.e. gas creates pressure fluctuations. Thus, PA effect is based on the detec-
tion of acoustic waves generated by a modulated radiation source. In order to obtain photoacoustic signal, the
intermittent or chopped radiation from the source assembly is directed towards photoacoustic cell that contains
a photoacoustic material such as air or any suitable gas, i.e, N2, C3H6 Ar, He, H2. A simple cell for photoacoustic
detection is an acoustically sealed chamber that contains a window, the sample and microphone. The periodic
heat flows from the sample creates in the cell pressure fluctuations that are detected by the microphone. Thus, the
photoacoustic spectrum consists of a curve that results from a plot between relative acoustic signal on the vertical
axis and wave number or wavelength on the horizontal axis.
There are several factors that affect the signal intensity. The most important factor is the internal cell volume.
Signal intensity is inversely proportional to the cell volume, so it is desirable to have the smallest possible volume.
Another important factor is the distance between the window and the solid sample. If this distance is less than the
thermal diffusion length of the gas, heat generated at the sample will be lost to the window, decreasing the signal
intensity. Signal intensity can also be lost due to acoustic damping that results from viscous friction in the gas and
heat losses to the walls during sound transmission. This is usually only a problem when the microphone is at the
end of a small diameter tube.
The background signal and noise should be minimised since they also affect the signal intensity. The two
factors that are important for minimising the background signal are (i) scattered radiation should not strike the
microphone diaphragm since microphone is sensitive to signals generated by modulated radiation absorbed by
diaphragm; and (ii) radiation absorption by interior surfaces of the cell must be minimised. This is achieved by
using cells made of either polished metal or optical grade glass. The two important sources of noise in a PA cell
are mechanical vibrations and acoustic leaks. The microphone diaphragm is partially sensitive to vibrations. This
effect is minimised by mounting the entire cell assembly on a base plate that is vibrationally isolated from the
table. The acoustic leaks in the cell not only increase the background noise but also decrease the signal intensity,
A back seal in a back-vented microphone increases the damping of this noise source. Even front-vented micro-
phones have acoustic leaks and must have the damping increased.
As with UV/V spectral studies, PAS has been used to obtain spectra in the mid-infrared region for qualitative
studies. This holds for the investigation of solid substances, as well as, chemically modified solid surfaces normally
inaccessible to IR conventional methods because of their tendency to scatter radiation. Such substances include
physically thick samples such as commercial catalyst pellets or opaque, such as chars and carbons, Mid-IR PAS
of CO on alumina supported platinum in the 1700–2200 cm−1 region is shown in Fig. 3.57. A strong band at 2060
cm−1 and weaker, broad band at about 1820 cm−1 correspond to the CO stretching vibrations of the adsorbed CO
molecules. PAIS of degassed alumina is compared with that of alumina exposed to acetic acid vapour in Fig. 3.58.
7.00
6.00 (a)
5.00
Relative PAS signal
Relative PAS signal
(b)
4.00
3.00 CH3
CO (1800)
2.00 H2O (1640)
1.00
3900 2100 2000 1400

2200 2000 1800
−1
Wave number (cm−1) Wave number (cm )
Fig. 3.57 Photoacoustic mid-infrared spectrum of Fig. 3.58 Photoacoustic infrared spectrum of (a) degassed
CO on alumina supported 5 per cent Pt. alumina, and (b) alumina exposed to acetic acid vapours.
Alumina shows the broad absorption characteristic of surface Al—OH groups and of bound water and the
deformation band of the latter. After esterification, decline in the intensity of O—H and H2O bands has been
observed and a band in the C—H region as well as a CO stretching band near 1800 cm−1 appeared. These two new
bands are attributed to the formation of a surface ester, Al—OCOCH3. In addition, the method has been used for
detecting the components of mixtures separated by thin layer and high performance liquid chromatography. Most
of these studies are done with FT-IR instruments because of their better signal-to-noise ratio, characteristics and
increased signal intensity. PAIS has been primarily used for the analysis of gases and for the study of vibrational
relaxation times. The concentrations of gaseous pollutants in the atmosphere are also monitored by this technique.
Here a tunable CO2 laser source is used in conjunction with photoacoustic cell filled with some photoacoustic
material that provides maximum noise. A system of this kind can analyse a mixture often gases with a sensitivity
of 1 ppb and a cycle time of 5 minutes. The vibrational relaxation process may be best understood as follows: The
exchange of energy between vibrational degrees of freedom and translational degrees of freedom can be charac-
terised by relaxation time, If the frequency of intermittent radiation is much less than the relaxation time, then
the transfer of energy from the vibrational to the translational modes would be very small. Since in photoacoustic
effect, it is the translational energy that contributes to the sound signal, a study of the change in the phase and
amplitude of the pressure signal with modulation frequency enables one
to understand the vibrational-translation process better. For such inves-
tigations, pulsed lasers are used.
PAS particularly in the near IR (PANIS) has the potential for provid- (a)
ing quantitative information about the surface coverage on chemically (b)

modified surfaces. Absorption bands arising from second overtone of
the aliphatic C—H stretch vibrational mode in octadecyl-, decyl- and
hexyl-modified silica gel are observed at 1188 nm as shown in Fig. 3.59. (c)
The second overtone of the aliphatic C—H stretch vibrational mode is
observed at 1710 nm. It is found that the magnitude of the signal ampli-
tude of both overtones when measured during modification as a function
of time has a linear relationship with carbon coverage, i.e. carbon con-
tent of modified silica gel. Similar behaviour has also been observed in
b-phenethyl modified silica gel at the first (1664 nm) and second (1130
1000 1200 1400
nm) overtones of the aromatic C—H stretch vibrational mode. The Wavelength (nm)
amplitude of the second overtone of the aliphatic C—H stretch vibra- Fig. 3.59 Photoacoustic near infrared spec-
tional mode in alkyl-modified silica gel, i.e. hexyl, octyl, decyl, dodecyl trum of (a) octadecyl-, (b) decyl-, and (c) hexyl
and octadecyl, varies linearly with chain length and carbon content of modified silica gel.
modified silica. The signal amplitude for the first overtone of the N—H
stretch vibrational mode for primary aliphatic amines at 1520 nm when recorded as a function of time in case of
aminopropyl-modified silica gel bears linear relationship with amine as well as silica contents of modified silica.
The advantages of PAS over the conventional methods may be summarised as follows: (i) there is no problem
about scattered radiation. This is taken as advantage in studying the spectra directly in TLC plate, (ii) small amount
of the sample can easily be studied, (iii) difference spectrum can easily be obtained, and (iv) the photoacoustic
spectrum is a power spectrum and hence depends not only on photo flux but also on the frequency (energy). Thus
the signal due to overtone would be twice as strong as that of the fundamental vibrational mode.
3.22 IR EMISSION SPECTROSCOPY(IMS)

Molecules that absorb IR radiation can also emit characteristic IR frequencies on heating. The lack of
applications of IMS is because of poor signal-to-noise ratio, characteristics of the IR emission signal
particularly when the analyte is at a temperature only slightly higher than its surroundings. This problem has
been solved with the aid of interferometric technique. Microgram quantities (1–10 mg) of pesticides such as
DDT, malathion and dieldrin can be identified by the interferometer of a FT spectrometer. Solutions of the
samples are prepared in a suitable solvent. The solution is then evaporated on a NaCl or KBr plate. Plate is
heated electrically near the spectrometer entrance. The components, i.e. SO2, CO2, etc., emitted from indus-
trial stacks can be identified with the aid of interferometric technique. The emitted radiations from a distant
industrial plumes are collected with a reflecting telescope and directed into the interferometer of FT infrared
spectrometer for detection.
PROBLEMS
1. Deduce the structure of a compound, C3H8O which has an 10. Calculate the amplitude of vibrations in V = 0 level of the
infrared band at 2941 cm–1 but none near 3333 cm–1 and following systems from their force constant data.
1724 cm–1.
[Ans: Methylethylether] Molecule H2 NO CO N2 O2
2. Deduce the structure of a compound, C2H5ON, which k (millidynes/Å) 5.2 15.5 18.7 22.6 11.4
has infrared bands at 3333, 3173 and 1678 cm–1
• Hint: For maximum displacement, potential energy =
[Ans: Acetamide] zero point vibrational energy, i.e.
3. Deduce the structure of a compound, C4H6O3 which has
infrared bands at 1750 and 1820 cm–1. 1 1 ⎛ 1⎞ 1
U k , A02 hνV EV = hνV V + , o V = 0, E0 hν 0
2 2 ⎝ 2⎠ 2
[Ans: Aceticanhydride]
4. Deduce the structure of a compound C8H8O which has [Ans: For CO, A0 = 4.78 × 106 cm]
an infrared band at 1689 cm–1 but no broad band near
11. Compute the maximal displacement above equilibrium
3333 cm–1.
distance as a percentage of re for the following systems
[Ans: Acetophenone]
from their vibrational frequencies.
5. Find out to what functional groups vibrations with the
wave numbers 3100, 2260, 1600, 1570, 750 and 686 cm−1
belong in the infrared spectrum of phenylisocyanate? Molecule Br2 CO FO K2 O2 N2
−1
[Ans: 686, 750 and 3100 cm–1 correspond to the group we (cm ) 325.3 2169.81 1060 92.96 1580.19 2358.03
≡ C—H, 2260 cm−1 to —N = C = O and 1600 cm–1 = to re(Å) 2.28 1.128 1.32 3.92 1.207 1.098
• Hint: Total potential energy = Zero point vibrational

C ] 1 1
energy, i.e. k (r– re)2 = hvV.
2 2
[Ans: 24 per cent for O2]
6. (a) Complete the vacant spaces in the following table.
12. The prominent maxima for different rotational levels in the
Molecule we (cm–1) R-branch of the rotation–vibration band in the absorption
spectrum of HCI vapour are given below. Determine the
H35C1
1
2990.95
number of molecules on each rotational quantum level if
H C1
2 35
— their extinction coefficients are independent of the rota-
H Cl
3 35
— tional energy.
H Cl
1 37
— J 0 1 5 8 10 12
H37C1
2
— −1
we(cm ) 2906.3 2925.8 2997.8 3044.9 3072.8 3098.4
H37Cl
3
—
D 0.155 0.418 0.362 0.065 0.011 0.001
(b) Write all the possible isotopes of C N and deter- 12 14
mine their frequencies if the vibration frequency of [Ans: J = 0, 3.07 × 1022 molecules; J = 12, 18.06 × 1019
l2
C 14N is observed at 2028.62 cm–1. molecules]
7. The three fundamental modes of vibration of MgF2 are observed 13. Compute the wave numbers of the first five lines in the
at 240. 447 and 870 cm–1. The intensities of the bands follow P-branch and R-branch of the rotation-vibration absorption
the order: I(870) > I(240) > I(447). Identify the types of vibration spectrum of HF using the needed data from the book.
associated with the said absorption bands. [Ans: P-branch, wave number of first line = 3915.93 cm–1
[Ans: nasym = 870 cm–1, nsym = 447 cm–1, d = 240 cm–1] for R-branch, wave number of third line = 4084.81 cm–1]
8. Three bands have been observed at 140 (weak), 14. Compute the wave numbers of thirteen absorption
129 (medium) and 403 (strong) in the infrared absorption bands in the R-branch of rotation-vibration spectrum of
spectra of SnCl4. Identify the types of vibrations associ- HCl, taking into consideration that the rotational constant
ated with the bands. BV varies. Given we = 2990.95 cm–1, xe = 17.65 × 10–3,
[Ans: nasym = 403 cm–1, nsym = 129 cm–1, d = 104 cm–1] Be = 10.59 cm–1, re = 1.274 Å. Find the value of a from that
of ν of the line J = 12 (see parameters of problem 3.12).
9. Determine (a) the maximal vibration quantum number,
(b) the vibrational energy on the zeroth level, (c) maximal [Ans: J = 0, ν = 2906.01 cm–1, J = 4, ν = 2981.65 cm–1,
vibrational energy, and (d) the bond energy for mole of J = 10, ν = 3075.51 cm–1]
the following systems. 15. The force constant for N2 molecule is 22.6 millidynes/Å.
How much energy is needed to increase the nuclear
Molecule DI FO H2 K2 N2 OH HF
separation by 0.5 Å?
we(cm−1) 1639.6 1060 4400.4 92.96 2358.03 3737.9 4138.7
1
• Hint: k A2 , A = 0 5 Å, leV = 8068 cm−1
xe × 103 12.1 14.6 27.4 3.82 6.00 29.7 21.7 2
16. The values of we and xe for some diatomics are listed
[Ans: For HF, (a) 22 (b) 4.0677 × 10–25 kJ below in the tabular form. Compute their zero point
(c) 94.154 × 10–25 kJ (d) 542.54 kJ mole–1] energies.
24. The fundamental band of HCl is observed at 2886 cm–1

Molecule Br2 Cl2 F2 I2 P2 and it fits into the empirical equation ν = 2885.90 +
20.577 m –0.3034 m2
we 325.3 559.75 917.55 214.55 780.69
where m = +1, +2, +3,... for R-branch
xe × 103 3.31 4.81 12.9 2.87 3.61 m = –1, –2, –3,... for P-branch.
Determine the values of Be, B0, B1 Given ae = 0.3312 cm–1.
17. The force constant for NO molecule is 15.5 millidynes/ Å.
Determine the frequency of vibration of the molecule • Hint: Compare the empirical relation with the general
and the spacing between its vibrational levels in eV. expression for the rotational lines of a band i.e.
18. Find the temperature at which the number of NO ν = ν0 + (BV″ + BV′)m + (BV″ – BV′)m2
1
molecules in the V = 1 level be th of that of V = 0 level.
4 • Hint: For fundamental band, V ″ = 1 and V′ = 0, BV =
we = 1904.4 cm–1, xe = 7.45 × 10–3.
Be – ae ⎛⎜ V + 1⎞⎟
N 1 ⎝ 2⎠ [Ans: Be = 10.5911 cm–1]
• Hint: V =1 = e −ΔE / T
=
NV = 0 4
19. (a) Compare the relative populations of the first six 25. Using the spectral data from the book, determine the
states of P2 molecules at 300 and 500 K. Use the thermodynamic functions at 298.15 K for the following
data from the book. diatomics: K2, P2, DI, Cl2, Br2, BrCl, F2, ClF, OH, HF, DF.
(b) Find the temperature at which the number of P2 26. Determine the thermodynamic functions for the following
molecules in the V = 1 state equals 1/e of number systems at 298.15 K.
of molecules in the ground state (V = 0)
(c) Draw a curve between population and vibrational Molecule H2S SO2 HCN CS2 N2O PH3
quantum number.
Frequencies 2611 1151, 712, 397, 589, 2327,
(d) Check whether the results are affected significantly 524, 2089 657 1285 991,
if we xe is ignored.
(cm−1) 1290 1361 3312 1523 2224 2421,
NV = 1 − (w e w e xe ) hc / T
• Hint: =e 2684 1121
NV = 0
Search other relevant data required in the calculations from
20. Explain why the infrared absorption spectrum of the book. Compare the results thus determined with the one’s
a-chloroketone shows two bands around 1720 and 1740 obtained by computer programs given in the book.
cm–1? Assign these bands also.
[Ans: Thermodynamic function (cal/mole.K)
21. Determine the vibrational component of the partition
function for the following molecules at 500 K from their SO2 CS2
vibration frequencies:
Free energy –50.95 –41.174
Molecule K2 BrCl ClF DI DF Enthalpy 2.53 2.386]
we(cm−1) 92.96 444.27 786.15 1639.6 2998.19 27. The fine structure of the fundamental vibrational band
of HCl shows the following lines:
22. Determine total number of molecules on the vibrational
2998.04 2981.00 2963.29 2944.90 2925.90 2906.24
quantum levels V = 1 and on the zeroth rotational quan-
tum state at 1000 K for the following systems. 2863.02 2841.58 2819.56 2796.97 2773.82 2750.13
Molecule we(cm–1) we xe(cm–1) Determine r0, r1 and re.

28. (a) Calculate the equilibrium constant KP , of the reaction
35
C119F 787.50 7.00 1 1
H2 + Cl2 = HCI at 900 K and 1 atmospheric pres-
2 2
37
C119F 777.99 6.83
sure from the following rotational moments of inertia and
35
Cl79Br 443.10 1.80 vibrational frequencies.
35
Cl81Br 441.42 1.79
Substance HCI Cl2 H2
1
H I
127
2308.09 38.98
I × 1040(g cm2) 2.612 116.3 0.459
2
H I
127
1640.14 20.16
we (cm−1) 2989.4 564.9 4396.6
3
H I
127
1345.50 13.57
The heat of reaction in the ideal gaseous state at absolute
zero is – 92.140 kJ mol–1
23. For the molecule HD, re = 0.7413 Å, we = 3817.1 cm–1 and
[Ans: Kp = 8.357 × 105]
wexe = 94.96 cm–1. Compute the vibrational energies on the
0th, 7th, 12th, 16th, 20th and maximal vibrational quantum (b) Calculate the equilibrium constant, Kp of the reaction of
levels. Calculate (r – re) and plot the Morse curve. thermal dissociation of iodine I2 = 2I, at 1274 K and 1.0133
[Ans: V = 0, E = 3.744 × 10–23 kJ, r – re = 0.6380 Å, 0.8706 × 105 Pa if H00 = 148.766 kJ mole–1, the vibrational fre-
Å, V = 20, E = 76.098 × 10–23 kJ, r – re = 0.3839 Å, 0.5689 Å] quency of I2 molecule is we = 214.25 cm–1, and the moment
of inertia is 750.1 × 10–40 g cm2. The statistical weight of 29. Which of the following vibrations of acetylene are infra-
the zeroth electronic level of iodine atom is 4, and that of red active.
the zeroth electronic level of the I2 molecule is unity. → ← → ←
[Ans: Kp = 0.173] (a) H C ≡ C H (C H − stretch)
(c) Calculate equilibrium constant, KP of the reaction NO = → ← ← →
1 N + 1 O at l000 K and 1 atmospheric pressure if the (b) H C ≡ C H (C H − stretch) , cm−1

2 2
2 2 → → ← ←
rotational moments of inertia of the molecules, vibrational (c) H C ≡ C H (C C − stretch)
frequencies, and statistical weights of the zeroth elec-
tronic levels are as follow. ↑ ↑
(d) H↓ C ≡ C↓ H (C C − H bend)
Substance NO N2 O2 ↑ ↑
(e) H↓ C ≡ C H↓ (C C − H bend) 729 cm
−1
I × 1040(g cm2) 16.43 14.01 19.34

[Ans: (b), (e) ]
we (cm−1) 1916.5 2359.4 1579.8 30. The frequencies for three fundamental modes of vibration
of CS2 have been observed at 396.7 (IR), 656.5 (R) and
g0, elec 4 1 3 1523 (IR). Calculate the frequencies of the bands for the
following arbitrary transitions: (i) V(0, 0, 0) → V(2, 0, 0);
(ii) V(0, 0, 0) → V(0, 1, 2); (iii) V(0, 0, 2) → V(2, 2, 0) and
The heat of reaction which proceeds in the ideal gaseous
(iv) V(0, 1, 0) → V(0, 0, 1).
state at absolute zero is –90.761 kJ mole–1.
[Ans: Kp = 2.532 × 105] 31. The vapour phase spectrum of methyl-thioformate
shows resolved rotational contours. The PR separa-
(d) Determine the equilibrium constant of the reaction CO tions as determined from the CH3 asymmetric stretching
+ H2O = H2+ CO2 at 300 and 1000 K and 1.0133 × 105 (3043 R, 3036 Q, 3023 P, in cm–1) and CH3 symmet-
Pa ( = 1 atm) if the moments of inertia of the molecules, ric stretching (2966 R, 2955 Q, 2946 P, in cm–1) bands
vibrational frequencies and heats of their formation both of type A have been found to be 20 and 20 cm–1
in the ideal gaseous state at absolute zero are as fol- respectively. Determine (a) moments of inertia of the
lows. molecule, (b) the values of the parameters l and ka and
(c) the configuration of the molecule if the rotational
Substance I × 1040 we(cm–1) Δ H00 constants of the cis and trans configurations are
(gcm2) (kJ mole–1)
CO2 67.8336 1351.2, 2396.4, 672.2 −393.229 Rotational constant cis trans
(doubly degenerate) (in MHz)
H2 0.459 4396.6 0 A 17941.60 17510.18
H2O 1.0296 3835.4 −238.906 B 4131.91 11401.15
1.9674, 3938.7,
2.9970 1647.6 C 3440.69 7252.33
CO 14.49 2170.0 −113.88
32. Using the value of B = 58000 MHz and v = 2160.0 cm–1
[Ans: Kp (1000 K) = 1.2787] for CO, calculate the frequencies of the first few lines
(e) Determine the equilibrium constant of the gaseous isotopic of the R and P-branches in the vibration–rotation spectrum
exchange reaction between hydrogen and deuterium of CO.
iodide, viz.,
⎡ Ans: ν R = 2160.0 + 3.87(J + 1); cm -1 ⎤
H2 + 2DI = 2 HI + D2 ⎢ ⎥
⎢⎣ ν P = 2160.0 − 3.87J; cm -1 ⎥⎦
at 298 K. From spectroscopic measurements, ΔH0 =
0
33. The frequencies of the first few vibronic transitions to an
83 cal, and the moments of inertia are as follows: H2 excited state 7Li2 are given below:
(45.9 × 10–42), DI (855 × 10–42), HI (431 × 10–42), D2 (92.0
× 10–42), g cm2, σ for H2 and D2 are each 2 and for HI and
DI they are 1. Vibronic transitions ν obs (cm–1)
[Ans: Kp = 1.22] 0→0 14020
• Hint : At room temperature, the four vibrational factors
0→1 14279
are not very different from unity. Thus, for metathetic or
double decomposition reactions, a very simple expres- 0→2 14541
sion can be derived for equilibrium constant, particularly 0→3 14805
if diatomic molecules only participate in the process,
e.g. in the case of the gas reaction 0→4 15074
AB + CD = AC + BD 0→5 15345
1 2 3 4 From these data calculate the values of we and wexe for

the exicited state of 7Li2.
ΔH 0
3 M3M 4 II σσ
ln K = − + I
0
+ In 3 4 + In 1 2 ⎡⎣ Ans : ω″e = 267 ω″ x″e = 0.04 cm -1 ⎤⎦
766 cm -1; ω
267.76
RT 2 M1M2 I1I2 σ 3σ 4
APPENDIX 3.1 A: HARMONIC VIBRATOR

The vibrational energies of molecule is obtained from the solution of the Schrödinger wave equation. The potential of a linear
vibrator is given by
U ( x ) = 2π 2mν 02 x 2 (3.1 A-1)
where x is the displacement of the particle of mass m from its equilibrium position x = 0. The one-dimensional time indepen-
dent Schrödinger wave equation is expressed by
Hˆ ( ) E Ψ( ) (3.1 A-2)
where Ĥ is the operator corresponding to the Hamiltonian of the system and the eigen value E is simply a number.
Thus the wave equation for a harmonic vibrator can be expressed as
d2 Ψ 8π 2m
+ 2 [EV (x)]Ψ = 0
U(x) (3.1 A-3)
dx 2 h
where U( x ) 2p 2mn 02 x 2
The wave function should be continuous, single-valued and finite throughout the region of values –∞ to +∞ for x.
In order to solve the differential equation we introduce two quantities,
8π 2mE
λ= and α [ 4π 2 0 ]
h2
Accordingly, Eq. (3.1 A-3) becomes
d 2Ψ
+ [λ − 2
x 2 ]Ψ = 0 (3.1 A-4)
dx 2
Let us now find the asymptotic solution of the wave function for large value of x . For any value of energy constant E, a value
of x can be found such that for it and all larger values of x , l is negligibally small relative to a2 x2, the asymptotic form of
the equation then becomes:
d 2Ψ
= α 2 x 2Ψ (3.1 A-5)
dx 2
This equation is satisfied asymptotically by the exponential function

α 2
± x
Ψ=e 2
(3.1 A-6)
α
dΨ ± x2
i.e., =±α xe 2
dx
α α
and d 2Ψ ± x2 ± x2
= α 2
x 2
e 2
± α e 2
d x2
a 2
+ x
The second term is negligible in the region considered. The asymptotic solution e 2
is unsatisfactory since it rapidly tends
to infinity with the increasing value of x . The satisfactory asymptotic solution of the equation throughout the configuration
space (–∞ < x < +∞ is now obtained by the power series method as follows:
α 2
−
Let Ψ=e 2
x
f (x) (3.1 A-7)
α
d 2Ψ − x2
∴ = e 2
[α 2 x 2f − α f − 2α xff ′ + f ″ ] (3.1 A-8)
dx 2
df d 2f
where f′ = and f ″ =
dx dx 2
α 2
− x
Dividing Eq. (3.1 A-4) by e 2 and comparing the resultant equation with Eq. (3.1 A-8) we get
f ″ 2α xff ′ + (λ α )f = 0 (3.1 A-9)
Let us now introduce a new variable ξ related to x by the expression:
ξ αx (3.1 A-10)
and to replace the function f(x) by H(ξ) to which it is equal. Consequently Eq. (3.1 A-9) becomes
d 2H dH λ
− 2ξ ( 1)H = 0 (3.1 A-11)
dξ2 dξ α
Let us now represent H(ξ) as a power series, i.e.

∞
H(ξ) = ∑
ν = 0,1,2
ν ξν = a + a ξ1 + a ξ2 + a ξ3 + ……….
0 1 2 3
(3.1 A-12a)
The first and second derivatives of the series are
dH
dξ
= ∑ νa ξ
ν
ν
ν −1
= a1 + 2 a2 ξ + 3 a3 ξ2 + (3.1 A-12b)
d 2H
dξ2
= ∑ ν(ν − 1)a ξ
ν
ν
ν −2
= 1.2a2 + 2.3 a3ξ + …….. (3.1 A-12c)
Substituting the set of Eqs. (3.1 A-12 a-c) into Eq. (3.1 A-11), we get
1.2 a2 + 2.3 a3ξ + 3.4 a4 ξ2 + 4.5 a5 ξ3 + …………
– 2 a1 ξ – 2.2 a2 ξ2 – 3.2 a3 ξ3 + …………..
⎛λ ⎞ ⎛λ ⎞ ⎛λ ⎞ ⎛λ ⎞
+ ⎜ − 1⎟ a0 + ⎜ − 1⎟ a1 ξ + ⎜ − 1⎟ a2ξ2 + ⎜ − 1⎟ a3 ξ3 = 0 (3.1 A-13)
⎝α ⎠ ⎝α ⎠ ⎝α ⎠ ⎝α ⎠
For H(ξ) to be the solution of Eq. (3.1 A-11), the seires will vanish for all values of ξ provided the coefficients of individual
power of ξ also vanishes, i.e.
⎛λ ⎞
ξ0 1.2a2 + 1 0
⎝α ⎠ 0
⎛λ ⎞
ξ1 2.3 3 + 1 2 a1 = 0
⎝α ⎠
⎛λ ⎞
ξ2 3.4a4 + 1 22 0
⎝α ⎠ 2
λ
ξ3 4.5 2 +( 1 2.3) 3 0
α
Accordingly we write the general expression for the coefficient of ξν,
λ
( )( )aν +2 + ( )aν = 0 (3.1 A-14)
α
λ
− 2ν − 1
aν α aν
or (ν )(ν + 2) (3.1 A-15)
From this recursion formula, one can determine the coefficients a2, a3, a4, a5,….. successively in terms of a0 and a1 which
are arbitrary.
For a0 = 0, the odd powers will appear.
For a1 = 0 only the even powers will appear, i.e.
a0 = 0 a1 = 0
λ ⎛λ ⎞
−( − 1) − ⎜ − 3⎟
α ⎝α ⎠
a2 a0 a3 a1
2 23
⎛λ ⎞⎛λ ⎞ ⎛λ ⎞⎛λ ⎞
−⎜ −5 1⎟ −⎜ −7 3⎟
⎝α ⎠ ⎝α ⎠ ⎝α ⎠ ⎝α ⎠
a4 a0 a5 a1
342 4523
or we can write
⎛λ ⎞ ⎛λ ⎞
− ⎜ − 1⎟ − ⎜ − 3⎟
⎝α ⎠ ⎝α ⎠
a2 a0, a3 a1
2 3
⎛λ ⎞⎛λ ⎞ ⎛λ ⎞⎛λ ⎞
−⎜ −5 1⎟ −⎜ −7 3⎟
⎝α ⎠ ⎝α ⎠ ⎝α ⎠ ⎝α ⎠
a4 a0 , a5 a1 (3.1 A-16)
4 5
Since a2, a4,….. are related to a0 and a3, a5 …… are related to a1, we can split the solution in even and odd series. Accord-
ingly, we can write
H(ξ) = [a0 + a2 ξ2 + a4 ξ4 +………] + [a1ξ + a3 ξ3 + a5 ξ5……..]
or
1+ γ 2 ( γ )( γ) 3−γ 3 ( γ )( γ)
H (ξ ) a0 [1 + ξ + ξ 4 ...] a1[ξ + ξ + ξ 5 + ....] (3.1 A-17)
2 4 3 5
λ
Here γ = . Since the values of a0 and a1 are arbitrary, we have obtained two independent solutions of Eq. (3.1 A-11) and
α
therefore Eq. (3.1 A-17) represents the most general solution of Eq. (3.1 A-11). It can easily be seen that when g = 1, 5, 9,…..
the even series becomes a polynomial and the odd series remains an infinite series.
Similarly, for g = 3, 7, 11……..the odd series becomes a polynomial and the even series remains an infinite series.
Thus when g = 2 n + 1; n = 0, 1, 2, 3,……. one of the solutions becomes a polynomial. Returning to eq. (3.1 A-15) we have
⎛ λ⎞
2ν + 1 − ⎟
aν + 2 ⎜⎝ α⎠ 2
= al es of ν.
= for l arg e values
aν (ν + 1)(ν + 2) ν
aν + 2
Thus lim = 0, i.e. the series always converges which shows that both the infinite series in Eq. (3.1 A-17) will converge
aν ν →∞
for all values of ξ.

2
Let us now compare the series for H and that for eξ .
∞
H(ξ ) ∑
ν = 0,2,4,...
aν ξν
⎛ λ⎞
2ν + 1 − ⎟
aν + 2 ⎜⎝ α⎠
=
aν (ν + 1)(ν + 2)
2 ξ 4 ξ6 ξν ξν + 2
eξ = 1 + ξ 2 + + + + + (3.1 A-18)
2 3 ν ν
+1
2 2
For large values of ξ, the first few term of these series is unimportant. Let the ratio of the coefficients of νth terms in the
2 a
expansion of H(ξ) and eξ equals c, which may be small or large, i.e. ν = c, if bν is the coefficient of ξν in the expansion of
2
bν
eξ . For large value of ν, the asymptotic relations will be:
2 2
aν a and bν b
ν ν ν ν
aν + 2 aν
Thus = =c,
bν + 2 bν
2
if ν is large. Therefore the higher terms of the series for H(ξ) differ from those of eξ only by a multiplication constant. But the
a 2
ratio of ν + 2 (in the limit of large ν) is the same as that of the coefficients of ξν+2 and ξν in the expansion of eξ . Consequently
bν + 2
2
if the series in Eq. (3.1 A-17) is not terminated, H(ξ) will behave as eξ for large value of ξ, therefore Ψ(ξ) and the product
2 2
e −ξ /2
. H (ξ) will behave as eξ /2
in this region, thus making it unacceptable as wave function, i.e. the boundary condition
Ψ (ξ) → 0 as ξ → ± ∞ (3.1 A-19)
will not be satisfied. The boundary conditions will only be satisfied by the wave function if the infinite series terminates to a
polynomial as mentioned earlier and this can happen only when g is an odd integer, i.e.
g = 2 n + 1; n = 0, 1, 2,…..
or l = (2n + 1) a (3.1 A-20)
The condition expressed in Eq. (3.1 A-20) for the existence of the nth wave function becomes:
E = En = (n + 1/2) hν0, n = 0, 1, 2, 3,………… (3.1 A-21)
hν 0
where a = , the zero point energy of the molecule. Equation (3.1 A-21) represents the discrete energy eigen values of
2
the operator H.
APPENDIX 3.2 A ANHARMONIC VIBRATOR

The wave equation for anharmonic vibrator can be expressed as
d 2 Ψ(ρ ) 8π 2Ie
+ 2 [E [EV U (ρ )]Ψ(ρ ) = 0
d ρ2 h
r − re
where ρ= (3.2 A-1)
re
To solve this equation advantage can be taken of the Morse potential of the type:
aρ 2
U(ρ) D(
D( e ) (3.2 A-2)
−2Da 2 ρ 2Da 2
For small r (small vibrations), the Morse function reduces to U (ρ ) Da 2 ρ 2 or f = = −krre ρ , where k = . In this
re re2
krre2 ρ 2
case the molecule is an harmonic oscillator. Thus substituting U (ρ ) Da 2 ρ 2 = into Eq. (3.2 A-1) we get the equation
of a harmonic oscillator. 2
By means of the Morse function, Eq (3.2 A-1) transforms as ( y e − aρ )
⎡ d 2Ψ d ⎛ d Ψ⎞ d ⎛ d Ψ dy ⎞ d Ψ d 2 y dy d dy d ⎛ d Ψ ⎞
⎢ 2 = = = + .
⎣ dρ d ρ ⎝ d ρ ⎠ d ρ ⎜⎝ dy d ρ ⎟⎠
⎜ ⎟ dy d ρ 2 d ρ d ρ d dy ⎜⎝ dy ⎟⎠
2
d Ψ d 2 y ⎛ dy ⎞ d 2 Ψ dΨ d 2Ψ ⎤
= +⎜ ⎟ = a2 y + a2 y 2 2 ⎥
dy d ρ ⎝ d ρ ⎠ dy
2 2
dy dy ⎦
d 2 Ψ 1 d Ψ 8π 2Ie ⎛ (Ev D ) 2D ⎞
+ + + − D⎟ Ψ = 0
dy 2 y dy a 2h 2 ⎜⎝ y 2 y ⎠ (3.2 A-3)
Now using the function y of the type

f (y ) (3.2 A-4)
Ψ=
y
dΨ d 2Ψ
and computing and from (3.2 A-4) and substituting the respective values into Eq. (3.2 A-3), we get
dy dy 2
Q 2D D
f ″ + λ 2( + − )f = 0 (3.2 A-5)
y 2 a2 y a2
8π 2Ie E −D 1
where λ 2 = and Q = V 2 +
h 2
a 4λ 2
The sought for asymptotic solution of Eq. (3.2 A-5) holds for any high value of y. Since we can ignore terms containing y in
the denominator, the sought for solution should satisfy the differential equation
λ2
f″ D f =0 (3.2 A-6)
a2
λ2
The solution to this equation has the form, as
±βy
, where β 2
= D
a 2 . The asymptotic solution e is unsatisfactory since it
f e +by
tends rapidly to infinity with increasing value of y. But solution e is satisfactory (f → 0 when y →∞) throughout the range
−by
of variation of the variable y.

The general solution of the equation throughout the configuration space is obtained by the function
f C( y )e − β y (3.2 A-7)
Compute f ″( C ″e
βy
βC e β y β 2Ce βy
) from Eq. (3.2 A-7) and substitute the values of f ′ and f ″ into Eq. (3.2 A-5) to
obtain a differential equation for C(y).
Q 2D
C ″ 2βC′
C′ + λ 2( + )C = 0 (3.2 A-8)
y 2 a2 y
The solution of this equation has the form:
C ∑b y
k
k
ν +k
(3.2 A-9)
Substituting the polynomial into Eq. (3.2 A-8), we get
∑b (k k )( k )y ν + k − 2 − 2β ∑ k ν k )y ν + k −1 + 2β 2 ∑ bk y ν 1
+ λ Q ∑ bk y
ν
2 2
=0 (3.2 A-10)
ν + k −1
Equating the coefficients of y we get
[((ν + k + 1)(
)( ) + λ2 ] [ β(ν
[2 ) 2β 2 ]bk (3.2 A-11)
Imposing the condition b0 = 0 and bV+2 = 0, we obtain from the recurrent equation
ν (ν + 1) = − l2Q (for b0 = 0, i.e. k = 0)
and ν + v + 1 = b (for b V + 2 = 0, i.e. V + 2 = k + 1 or k = V + 1)
Eliminating ν we get
−l2Q = (b − V )( b − V − 1) = b 2 − 2b (V + 1/2) + V (V + 1) (3.2 A-12)

⎛ λ2 ⎞
Substituing the value of Q from Eq (3.2 A 5) and also noting that β 2 = ⎜ 2 ⎟ D , we get
⎝a ⎠
a2 a2
EV = 2β (V + 1/2)
) − (V + 1/2)2 (3.2 A-13)
λ2 λ2
Also we know that
hcω e 1 k a 2D
xe = and ω e = 2π c μ
=
2π c Ie
4D
So Eq. (3.2 A-13) reduces to
E V = h c we (V + 1/2) − h c we xe (V + 1/2)2
or the vibrational term is

EV
GV (in cm 1) = = ω e (V
(V / ) e xe (V + 1// )2 (3.2 A-14)
hc
From spectroscopic data it follows that the best approximation is the formula
E V = h c we (V + 1/2) − h c we x e (V + 1/2)2 + h c we y e (V + 1/2)3 + …. (3.2.A-15)

APPENDIX 3.3 A ENERGY LEVELS FOR THE DIATOMIC ROTATOR–

VIBRATOR OR ENERGY LEVELS OF DIATOMIC NONRIGID
ROTATOR-ANHARMONIC VIBRATOR
The radial part of the Schrödinger wave equation for the diatomic nonrigid rotator-anharmonic oscillator can be written as
⎞ ⎡ (J + 1) 8π μ ⎤
2
1 ⎛
⎜⎝ r ⎟⎠ + ⎢ − + 2 { (r )}⎥ R = 0 (3.3 A-1)
r 2 dr dr ⎣ r 2
h ⎦
Here for simplicity the subscripts are omitted, i.e. E = EJ,V or Erot-vib and yJ, V = yrot-Vib = y, sub-scripts are omitted. The equation
is further simplified by putting
1
R(r ) = Ψ(r ) (3.3 A-2)
r
so that
1 d ⎡ 2 d ⎛ 1 ⎞ ⎤ 1 d ⎡ 2 ⎛ Ψ 1 d Ψ⎞ ⎤
⎢r Ψ ⎥= ⎢r − + ⎥
r 2 dr ⎣ dr ⎜⎝ r ⎟⎠ ⎦ r 2 dr ⎣ ⎝ r 2 r dr ⎟⎠ ⎦
1 d ⎡ d ⎤ 1 ⎡ dΨ dΨ d 2Ψ ⎤ 1 d 2Ψ
= − Ψ + r = ⎢ − + + r ⎥ =
r dr ⎢⎣
2
dr ⎥⎦ r ⎣ dr
2
dr dr 2 ⎦ r dr 2
Thus Eq. (3.3. A-1) in terms of function y becomes

d 2 Ψ ⎡ J(J + 1) 8π 2 μ ⎤ (3.3 A-3)
+ ⎢− + 2 {E U (r )}⎥ Ψ = 0
dr 2 ⎣ r2 h ⎦
1
The simple Hooke’s law potential energy function U(r) = k (r re )2 , fails to agree with experiment in that it yields equally
2
spaced levels, whereas the observed vibrational levels show a convergence for increasing values of V. On the other hand,
as has always been discussed, the Morse function,
a ( r re
U(r) = D[ e )]2 (3.3 A-4)
leads to very accurate values for the energy level of nearly all the molecules and only for a few molecules it is necessary to
consider further improvements. This function has a minimum value of zero at r = re and approaches a finite value for a large
value of r. At r = 0, U(r) is infinite, whereas Morse function is finite at all other values of r. The very large value of Morse func-
tion at r = 0 does not create any serious problem.
Lumping together Eqs.(3.3 A-3) and (3.3 A-4) we get
d 2 Ψ ⎡ J(J + 1) 8π 2 μ ⎤
+ ⎢−
dr 2 ⎣ r2
+ 2 E − D − De { 2 a r re
+ 2De a r re
}⎥ Ψ = 0 (3.3 A-5)
h ⎦
h2
Substituting z e −a(
a r − re )
and λ = J(J+1) into Eq. (3.3 A-5), we obtain
8π 2 μre2
d 2Ψ 1 d 8π 2 μ ⎡ E D 2D λ re2 ⎤
+ + ⎢ + − D − ⎥Ψ = 0 (3.3 A-6)
dz 2 z dz a 2h 2 ⎣ z 2 z z 2r 2 ⎦
2
Now the quantity re can be expressed in terms of z as
r2
z e −a(
a r − re )
ln z ar + arre
ln z r
or = − +1
+ arre re
r ln z
or =( − )
re arre
re2 1
or =
r 2 ⎛ ln z ⎞ 2
⎜⎝ 1 − arr ⎟⎠
e
re2 2 ⎛ 1 3 ⎞
or = 1+ (z − 1) + ⎜ − + 2 2 (z − 1)2 + .... (3.3 A-7)
r2 arre ⎝ arre a re ⎟⎠
Substituting the first three terms of the Taylor expansion into Eq. (3.3.A-6) we obtain
d Ψ 1 π μ −C D C
D C2 Ψ = 0 (3.3 A-8)
dz z dz z2 z
⎛ 3 3 ⎞
Here C 0= − +
ar
4 6 ⎞
C 1= − (3.3 A-9)
C − +
ar
Making the further substitutions:
Ψ e− y / ( )
where
y = 2 dz
8π μ
d = (3.3 A-10)
32π μ
b =− (E D − )
and proceeding as follows:

Ψ e − dz /2
L
dΨ ⎡ d
e dz
{(2d Lz + d b 2
Lz} e
dz dz dz
dL b ⎤
= d +
dz 2
dL
or
dΨ ⎡ dL b ⎤
= + z d Lz (3.3 A-11)
dz 2
d 2Ψ d ⎡
d L z) + ⎡
dL
= + z + z . ( − /2
)]
dz ⎣ dz dz
or finally
(3.3 A-12)
d Ψ ⎡d L dL ⎡
= d + 1 − +d Lz
dz 2 dz dz
The first term of Eq. (3.3 A-8), i.e.
d Ψ 1 dΨ ⎡ ⎤ dL
+ = [(2d b z− 2d
dz 2 z dz dz d
b
+ z + +d Lz (3.3 A-13)
4
and the second term of Eq. (3.3 A-8), i.e.
8π μ ⎡ E D C 2D C ⎤
2
+ − Lz
z z
8π μ 1 D C1 2
= 2d ) ] − Lz
32π μ z 8π μ
⎡ b 8 ⎤
= d) ] − + −d L( ) (3.3 A-14)
4z z
Finally the sum of the Eqs. (3.3 A-13) and (3.3 A-14) yields
d 2L dL ⎡ 8π 2 μ(2D C1) ⎤
+ [(b )z 1 − 2d
d] + ⎢ −(b dz 1 +
)d ⎥ L(z ) = 0
dz 2 dz ⎣ a 2h 2 z ⎦
or
d 2L dL ⎡ 8π 2 μ(2D C1) ⎤
+ [(b )z 1 − 2d
2d ] +⎢ − (b dz 1 ⎥ L( ) = 0
)d (3.3 A-15)
dz 2 dz ⎣ a 2h 2 z ⎦
Now
y = 2dz or z -1 = 2d/y,
dL 1 . dL
dy = 2d dz or dz = dy/2d; =
dz 2d dy
dy 2
and dz 2 =
4d 2
So Eq. (3.3 A-15) becomes

d 2L ⎡ ( 1)2d ⎤ dL ⎡ 8 2
μ((2 1 )2d 2d ⎤
4d 2 =⎢ − 2d ⎥ 2d . +⎢ −( 1) d . ⎥ L( y ) = 0
dy 2 ⎣ y ⎦ dy ⎣ a 2h 2 y y ⎦
or
d 2L ⎡ (b + 1) ⎤ dL 1 ⎡ 4π 2 μ(2D C1) 1 ⎤
2
+⎢ − 1⎥ + ⎢ 2 2
− (b + 1)⎥ L( y ) = 0
dy ⎣ y ⎦ dy y ⎣ ahd 2 ⎦
or finally
d 2L ⎡ (b + 1) ⎤ dL V
+⎢ − 1⎥ L( y ) 0 (3.3 A-15a)
dy 2 ⎣ y ⎦ dy y
in which
4π 2 μ 1
V = (2D C1) − (b + 1) (3.3 A-16)
a 2h 2d 2
We now try a power series solution of the form
∞
L(y) = ∑a y
ν =0
ν
ν
= a 0 y 0 + a1y 1 + a2 y 2 + a3 y 3 + ...
L′( y ) = a1 + 2a2 y + 3a3 y 2 + .....

and
L″( y ) = 2a2 + 3 2a3 y + .....
Substituting these values of L(y), L’(y) and L’’(y) into Eq.(3.3 A-15a) and arranging the coefficients of y in the ascending order
we get
[(b + 1) a1 + V a0] y −1 + [2a2 + 2(b+ 1) a2 + V a1 − a1] y 0
+ [3.2 a3 + 3(b+1) a3+ Va2 − 2a2] y1 = 0
For L(y) to be the solution of Eq (3.3 A-15a) the serie s will vanish for all values of y provided the coefficients of individual
powers of y also vanish separately, i.e.
y−1: 1 a1 (b + 1) + a0 V = 0
0
y: 2 a2 (b + 1 + 1) + a1 (V − 1) = 0
1
y: 3 a3 (b + 1 + 2) + a2 (V− 2) = 0
(ν + 1) aν+1 (b + 1 + ν) + aν (V− ν) = 0
aν +1 ν −V
or = (3.3 A-17)
aν (ν + 1)(ν + 1 + b )
Further, L(y) to be a polynomial, the series L(y) must terminate after a finite number of terms which is possible only if
ν − V = 0 or ν = V, i.e V = 0, 1, 2, 3, ……..
The solutions for V integral satisfy the boundary conditions when L → 0 as r → −∞ instead of as r → 0. Further in order to
obtain E from Eq (3.3 A-16), we proceed as follows:
8π 2 μ 16π 2 μ C 1
b= − 2 −1 − +
d d 2 2
or
2 2
16π μ ⎞ C1 1 64π μ 1
b = D− + + 1
V+ (3.3 A-18)
d 2 2 d 2 2
Substituting the value of b2 into Eq.(3.3 A-16) and simplifying the resultant equation,. we get
32π μ 16π μ ⎞ C1 1 64π μ C1 1

− D − − V+
d 2 2 d 2
or
2 2
8π μ C 1 2 C 1
C0 + + − V (3.3 A-19)
d 2 8π μ 2 d 2 2
Further substituting the values of d, we obtain:

2
C1 C1 1
D− D V+
8π μ 2 ah 1 2 2
C − − + +
8π 2 μ (D C ) 8π μ 2 D +C
or C1
2
C1 1
D− D− V+
ah 1 2 1
C0 − (3.3 A-20)
D C D + C2 8π μ 2
C1 C
Further simplification is carried out by expanding in terms of power series of and and substituting the values of the
D D
parameters C0, C1 and C2 in the resultant equation, i.e.
C1
D−
2 C ⎡ C ⎤
= − 1+
D C 2D D
⎡ C2 C C 2 ..⎤
= D 1 4D −
D D D
C 2
C2 C .. ⎤
=D 1 D − +
D 4D D D
Neglecting the higher terms in C1 and C2.
⎡ 1 1 ⎡C ⎤
= D 1− D + + + ...
D 4
1 C2 ⎤
= + + + ...
D 4
4 6 1 3 4 1 4
=D − + + − ,
ar D a e e e
the cube and higher terms in are are ignored,

2
3 3 1
=D are D a2 2
e
3 3
and C0 +
are e
So the sum of the first three terms of the Eq. (3.3 A-20) is
C1
D−
2 ⎛ 3 3 ⎞
− = + λ − −
D C ar ar a e e
2
h J ⎛ h
= J J+ −
Da 8 Da 2 e2 8 e
J ⎛ h 2D ⎞
= + Beh
Dre2 8 e 4π c μ
h3
+ Bh hc 6
128
e + e (3.3 A-21)
h h3
where B = and De =
8 r 128π μ 3 6
C C1
D D
2 2D C1 C 1 ⎤
= D 1- 1 = D 1− +
1 1
2D 2D 2D
D C 2 C2
D 1+
D
[Squares and higher terms in C1 and C2 are ignored.]
C
D−
2 ⎛ 4 6 1 3 ⎞ ⎛ 3 3
1 2 2 = D [1
2
D a 2D ar a
D C
3 1 1 3J J 1 h2
= D = D
2 D are D 8 e e
e
Accordingly the fourth term of Eq. (3.3 A-20) yields
C1
D −
ah 2 1 ah 1 3J + 1) h2 ⎞⎛ 1 1 ⎞
V+ D − − 2 2
π μ D C2 2 π μ 2 D 8
ah D 1 3J J + 1 ⎛ ah 3 ⎞ 1 1
= + − V+
π μ 2 2 D 8π μ e 2
ah ⎞ 2 c ⎞ 1 3h J J + 1 2 c 2 1 1 1 1
= V+ − V+
π μ a 2 2 16 2D 2 ar 2
= h c we (V+1/2) - ae h c J(J + 1) (V+ 1/2) (3.3 A-22)
where a 2D 3h 2 ⎛ 1 1
= −
16 2 2
e e
D r a e
The fifth term of Eq.(3.3 A-20) yields
1 ⎛4 ⎞ ⎛ h2 1
=− V+
8 2 2D 8π μ 2
2
⎛4 2 2
h2 4xe ⎞ 1
e
V+
2 8p 2
= - h c we xe (V+ 1/2)2 (3.3 A-23)
Lumping together Eqs’ (3.3 A-20), (3.3 A-21), (3.3 A-22) and (3.3 A-23) we obtain
1
c e − e x / + 1) e + J J + 1) (3.3 A-24a)
or rotational–vibrational term will be

EV ,J ⎛ 1⎞
TV ,J = = ωe ⎜V xe (V
(V / )2 J (J )Be De J 2 (J + 1)2 − α e (V / )J(J + 1) (3.3 A-24b)
hc ⎝ 2⎠
e
which is the equation for the rotational-vibration energy levels of a diatomic molecule.
Thus from Eq. (3.3 A-24b) we can also write the expression for the rotational terms for the nonrigid rotator as
EJ
FJ (cm 1) = = Be J((JJ ) De J 2 (J )2 − α e (V / )J(J + 1) (3.3 A-25)
hc
APPENDIX 3.4 A: MOLECULAR CONSTANTS OF DIATOMIC MOLECULES
Molecule we, (cm−1) xe x103 Be, (cm−1) re, Å D298, kJ mole−1

Br2 325.3 3.31 0.0821 2.28 192.9
BrCl 444.27 4.15 0.152 2.13 215.0
Cl2 559.75 4.81 0.244 1.987 242.3
ClF 786.15 7.84 0.516 1.63 252.5
CO 2169.81 6.11 1.931 1.128 1075.0
DF 2998.19 15.3 11.00 0.818 273.2
DI 1639.6 12.1 32.84 1.61 298.5
F2 917.55 12.9 0.89 1.41 159.0
FO 1060 14.6 1.104 1.32 215.7
H2 4400.4 27.4 60.86 0.741 435.9
HBr 2648.97 17.1 8.465 1.414 366.5
HCl 2990.9 17.65 10.59 1.274 431.4
HF 4138.7 21.7 20.95 0.917 566.1
HI 2309.01 17.2 65.11 1.61 298.4
I2 214.55 2.87 0.037 2.66 151.0
IO 681.47 6.3 0.340 1.868 222.0
K2 92.96 3.82 0.056 3.92 53.14
N2 2358.03 6.00 1.998 1.098 945.6
NO 1904.4 7.45 1.704 1.151 631.0
O2 1580.19 0.76 1.445 1.207 498.7
OH 3737.9 29.7 18.895 0.969 428.0
P2 780.69 3.61 0.303 1.893 489.1
we = Intrinsic wave number; xe = anharmonic constant; Be = rotational constants; re = equilibrium internuclear distance;
D298 = dissociation energy.
APPENDIX 3.5 A: RECIPROCAL TABLE FOR THE CONVERSION

OF MICRONS (m) TO WAVE NUMBERS (CM-1)
l, m ν , cm−1 l, m ν , cm−1 l, m ν , cm−1 l, m ν , cm−1

2.50 4000 2.74 3649 2.98 3355 3.44 2906
2.51 3984 2.75 3636 2.99 3344 3.46 2890
2.52 3968 2.76 3623 3.00 3333 3.48 2873
2.53 3952 2.77 3610 3.02 3311 3.50 2857
(Continued)
Reciprocal table for the conversion of microns (m) to wave numbers (cm-1)—Cont’d.
2.54 3937 2.78 3597 3.04 3289 3.52 2841

2.55 3921 2.79 3584 3.06 3267 3.54 2824
2.56 3906 2.80 3571 3.08 3246 3.56 2809
2.57 3891 2.81 3558 3.10 3225 3.58 2793
2.58 3875 2.82 3546 3.12 3205 3.60 2778
2.59 3861 2.83 3533 3.14 3184 3.62 2762
2.60 3846 2.84 3521 3.16 3164 3.64 2747
2.61 3831 2.85 3508 3.18 3144 3.66 2732
2.62 3816 2.86 3496 3.20 3135 3.68 2717
2.63 3802 2.87 3484 3.22 3105 3.70 2703
2.64 3788 2.88 3472 3.24 3086 3.72 2688
2.65 3773 2.89 3460 3.26 3067 3.74 2674
2.66 3759 2.90 3448 3.28 3048 3.76 2660
2.67 3745 2.91 3436 3.30 3030 3.78 2646
2.68 3731 2.92 3424 3.32 3012 3.80 2632
2.69 3717 2.93 3412 3.34 2994 3.82 2618
2.70 3703 2.94 3401 3.36 2976 3.84 2604
2.71 3690 2.95 3389 3.38 2958 3.86 2591
2.72 3676 2.96 3378 3.40 2941 3.88 2577
2.73 3663 2.97 3367 3.42 2923 3.90 2564
3.92 2551 4.84 2066 6.00 1667 8.40 1190
3.94 2538 4.88 2049 6.10 1639 8.50 1176
3.96 2525 4.92 2033 6.20 1613 8.60 1162
3.98 2513 4.96 2016 6.30 1587 8.70 1149
4.00 2500 5.00 2000 6.40 1563 8.80 1136
4.04 2475 5.05 1980 6.50 1538 8.90 1124
4.08 2451 5.10 1960 6.60 1515 9.00 1111
4.12 2427 5.15 1942 6.70 1492 9.10 1099
4.16 2404 5.20 1923 6.80 1471 9.20 1087
4.20 2381 5.25 1905 6.90 1449 9.30 1075
4.24 2358 5.30 1887 7.00 1429 9.40 1064
4.28 2336 5.35 1869 7.10 1408 9.50 1052
4.32 2314 5.40 1851 7.20 1389 9.60 1042
4.36 2293 5.45 1835 7.30 1370 9.70 1031
4.40 2273 5.50 1818 7.40 1351 9.80 1020
4.44 2252 5.55 1802 7.50 1333 9.90 1010
4.48 2232 5.60 1876 7.60 1316 10.00 1000
4.52 2212 5.65 1770 7.70 1299 10.20 980
4.56 2193 5.70 1754 7.80 1282 10.40 962
4.60 2174 5.75 1739 7.90 1266 10.60 943
4.64 2155 5.80 1724 8.00 1250 10.80 926
4.68 2137 5.85 1709 8.10 1235 11.0 909
4.72 2119 5.90 1695 8.20 1220 11.20 893
4.76 2101 5.95 1681 8.30 1205 11.40 877
l, m ν , cm−1 l, m ν , cm−1 l, m ν , cm−1 l, m ν , cm−1

11.60 862 13.40 746 16.60 602 21.00 476
11.80 847 13.80 725 17.00 588 22.00 455
12.00 833 14.20 704 117.50 571 23.00 435
12.20 820 14.60 685 18.00 556 24.00 417
12.40 806 15.00 666 18.50 541 25.00 400
12.60 794 15.40 649 19.00 526 26.00 385
12.80 781 15.80 633 19.50 513 27.00 370
13.00 769 16.20 617 20.00 500 28.00 357
APPENDIX 3.6 A: PROGRAM IN C++ FOR THE COMPUTATION

OF STATISTICAL A THERMODYNAMIC FUNCTIONS
FROM SPECTROSCOPIC DATA
Developed by Hardeep Kaur Randhawa, Department of Computer Science and
Applications KMV, Jalandhar
program name : Randhawa.cpp
purpose : program to compute statistical thermodynamic
functions from spectroscopic data.
//Define directives
#include <iostream.h>
#include <iomanip.h>
#include <string.h>
#include <process.h>
#include <math.h>
#include <conio.h>
#include <stdio.h>
#include <dos.h>
#define H 6.625600E-27
#define VL 2.997925E+10
#define R 1.987170
#define AK 1.380540E-16
#define CTR 7.2810
#define CTS 2.313
#define CROL 175.683
#define CROLS 177.67
#define CRONL 264.559
#define CRONLS 267.54
int i,j;
istream &bottom(istream &stream);
void eins(int m,float cons,int i,double *,double *,double *,double
*,double *,double *);
void fun(int i);
//output manipulator which specify float(fixed) format.

ostream &for2(ostream &stream)
{
stream.setf(ios::fixed);
stream.width(11);
stream.precision(6);
stream.setf(ios::showpoint);
return stream;
}
//output manipulator which specify integer format.

ostream &for3(ostream &stream)
{
stream.width(4);
stream.setf(ios::left);
return stream;
}
//Manipulator to press any key to continue.

istream &bottom(istream &stream)
{
cout<<”\nPress any key to continue :\n\n”;
getch();
return (stream);
}
//Manipulator to produce a sound and wait.
ostream &beep(ostream &stream)
{
cout<<’\a’; //produces a sound
delay(500); //waits for 500 milliseconds
cout<<’\a’;
return (stream);
}
//Manipulator to clear the screen.
ostream &clear(ostream &stream)
{
clrscr();
return (stream);
}
//This creates the class outl
class outl
{
int xl,yl;
public:
outl(int i,int j){xl=i;yl=j;} //Constructing outl
~outl(){ } //Destructing outl
friend ostream &operator<<(ostream &stream,outl ol);
};
//This creates the class out2
class out2
{
int x2,x3;
public:
out2(int k2,int k3) {x2=k2;x3=k3;} //Constructing out2
~out2(){ } //Destructing out2
friend ostream &operator<<(ostream &stream,out2 o2);
};
//This creates the class init
class init
{
double x;
public:
init(){x=0.0;} //Constructing init
~init(){ } //Destructing init
double getx(){return x;}
};
//This creates the class lnl
class lnl
{
int nta,q,noa;
char nom[20];
double mi,xyz[6];
protected:
int na[10],m,1,11,p,atom;
float am[10];
double ai;
public:
void setdatal();
void displayl();
void lin_nlin();
void ainer();
};
//Defines a derived class data from lnl.
class data:public lnl
{
int n;
float cons, htv[30], gtv[30], cpv[30], sv[30], htro[30], cpro[30],
gtro[30], sro[30], httr[30], cptr[30], gttr[30], str[30], ht[30],
cp[30], gt[30], s[30], optr[30], op[30];
double wm, sigma, t[30], w[30], eh[30], ecp[30], eg[30], es[30];
public:
data(){};
friend ostream &operator<<(ostream &stream,data o);
friend istream &operator>>(istream &stream,data &o);
void cal();
void output();
};
//Display *---------*----------*--------- and so on.
ostream &operator<<(ostream &stream,outl ol)
{
register i,j;
stream<<”\n”;
for(j=1;j<=ol.yl;j++)
{
stream<<”*”;
for(i=0;i<ol.xl;i++)
stream<<”-”;
}
stream<<”*”<<”\n”;
return stream;
}
//Display some spaces,one * and again some spaces i.e. “ * “.
ostream &operator<<(ostream &stream,out2 o2)
{
register k2,k3;
for(k2=l;k2<=o2.x2;k2++)
stream<<” “;
stream<<”*”;
for(k3=l;k3<=o2.x3;k3++)
stream<<” “;
return stream;
}
/*Input name of the molecule, nature of the molecule, type of atoms,
no. of atoms of the same kind, the atomic mass of the respective atoms,
and moment of inertia. */
void lnl:: setdatal()
{
cout<<”Enter name of the molecule(nom) : “;
cin>>nom;
cout<<”\nEnter the nature of the molecule(1) (linear-1, non_
linear-2): ”;
cin>>l;
cout<<”\nEnter the type of atoms in the molecule(atom): “;
cin>>atom;
cout<<”\nEnter the number of atoms of the same kind and their \n″
<<″ respective atomic mass (na[],am[])\n”;
for(p=l;p<=atom;p++)
{
cout«“na[“<<p<<”] = “;
cin>>na[p];
cout<<”am[“<<p<<”] = “;
cin>>am[p];
}
Cout<<″\nEnter the value of moment of inertia (mi) (scientific):“;
cin>>mi;
if(mi==0)
{
cout<<”\nEnter the number of axis (noa) : ″;
cin>>noa;
cout<<”\nEnter the values of the moments of inertia along”;
<<noa<<” axis(xyz[])(scientific) :\n”;
for(q=l; q<=noa; q++)
{
cout<<”xyz[“<<q<<”] = “;
cin>>xyz[q];
}
}
}
//Function to check whether the molecule is linear or non_linear.
void lnl::lin_nlin()
{
nta=0;
for(p=l; p<=atom; p++)
nta=nta+ *(na+p);
11=1-1;
if(11==0)
{ m=3*nta-5;
cout<<”\nThe molecule is linear and has “<<m
<<” vibrational frequency points.\n”;
}
else
{ if(11 < 0)
{
cout<<”\nThere is some error in the program.\n”;
cout<<”\nPress any key to exit.”;
getche();
exit(0);
}
else
{
m=3*nta-6;
cout<<”\nThe molecule is non-linear and has “<<m
<<” vibrational frequency points.\n”;
}
}
cin>>bottom;
}
//Function to calculate log(moment of inertia).
void lnl :: ainer()
{
if(mi==0)
{
ai=0.0;
ai=ai+log(*(xyz+q));
}
else
ai=log(mi);
}
/*Function to display name of the molecule, number of atoms of
the same type, atomic mass of the respective atoms and moment of
inertia. */
void lnl :: displayl()
{
cout<<”\nThe name of the molecule is : “<<nom;
cout<<”\n\nThe number of atoms and the atomic mass of the
respective”
<< “atoms are :”<<”\n\n”;
cout<<for3<<na[p]<<for2<<am[p]<<”\n”;
cout.setf(ios::scientific);
cout<<”\nMoment of inertia = “<<mi;
if(mi==0)
{
cout<<”\n\nMoments of inertia along “<<noa<<” axis are :\n”;
cout<<xyz[q]<<”\n”;
}
cout<<”\n\nlog(moment of interia) = “<<for2<<ai<<”\n”;
}
//Input symmetry number, fundamental vibrational frequencies,
total //no. of temperatures and the values of the five initial
temperatures.
istream &operator>>(istream &stream,data &o)
{
cout<<″\nEnter the symmetry number (sigma) : “;
stream>>o.sigma;
<prog>cout<<”\nEnter the values of the fundamental vibrational
frequencies”
<<”(w[]) : “<<”\n” ;
for(j=1;j<=o.m;j++)
{
cout<<”w[“<<j<<”] = ″ ;
stream>>o.w[j];
}
cout<<”\nEnter the total number of temperatures(n:5 - 27): “;
stream>>o.n;
cout<<”\n\nEnter the values of the five initial temperatures {t [
]) : \n”;
for(i=l;i<=5;i++)
{
cout<<”t[“<<i<<”] = “;
stream>>o.t[i];
}
cout<<beep;
return stream;
}
/*Display the symmetry no.,no. of vibrational frequencies and the
values of the fundamental vibrational frequencies, total no. of
temperatures and the values of the five initial temperatures.
ostream &operator<<(ostream &stream,data o)

{
stream<<clear;
stream<<”\n\nThe value of sigma is : “<<for2<<o .sigma<<”\n\n”;
stream<<”The no. of vibrational frequencies is : “<<for3<<o.m<<”\
n\n”;
stream<<”The fundamental vibrational frequencies are : “<<”\n\n”;
for(j=1;j<=o.m;j++)
stream<<setw(11)<<setprecision(4)<<setiosflags(ios::fixed)
<<setiosflags(ios::left)<<setiosflags(ios::showpoint)<<o.w[j];
stream<<”\n\nTotal no. of temperatures is : “<<for3<<o.n<<”\n\n”;
stream<<”Five initial temperatures are : “<<”\n\n”;
for(i=l;i<=5;i++)
stream<<for2<<o.t[i];
stream<<”\n”;
return stream;
}
//Function to calculate contributions of various degrees of freedom to
//the thermodynamic functions and the total thermodynamic functions.
void data::cal()
{
double x,xl,x2,he,cpe,ge,se;int k;
wm=0.0;
wm=wm+ *(na+p)* *(am+p);
cons=H*VL/AK;
for(i=l;i<=n;i++)
{
if(i-5 >0)
*(t+i) = *(t+i-1)+100;
*(httr+i) = 2.5*R;
*(cptr+i) = 2.5*R;
*(gttr+i) = -1.5*R*log(wm)-2.5*R*log(*(t+i))+CTR;
*(str+i) = 1.5*R*log(wm)+2.5*R*log(*(t+i))-CTS;
if(11==0)
{
*(htro+i) = R;
*(cpro+i) = R;
*(gtro+i) = -R*ai-R*log(*(t+i))+R*log(sigma)-CROL;
*(sro+i) = R*ai+R*log(*(t+i))-R*log(sigma)+CR0LS;
}
else
{
if (11>0)
{
*(htro+i) = 1.5*R;
*(cpro+i) = 1.5*R;
*(gtro+i) = R*log(sigma)-0.5*R*ai-l.5*R*log(*(t+i))-CRONL;
*(sro+i) = -R*log(sigma)+0.5*R*ai+l.5*R*log(*(t+i))+CRONLS;
}
else
{ cout<<”error\n”; exit(O);}
}
eins(m,cons,i,t,w,eh,ecp,eg,es);
*(htv+i)=R* *(eh+i);
*(gtv+i)=R* *(eg+i);
*(cpv+i)=R* *(ecp+i);
*(sv+i) =R* *(es+i);
*(ht+i) = *(httr+i)+ *(htro+i) + *(htv+i);
*(cp+i) = *(cptr+i)+ * (cpro+i) + *(cpv+i);
*(gt+i) = *(gttr+i)+ *(gtro+i) + *(gtv+i);
*(s+i) = * (str+i)+ * (sro+i) + *(sv +i);

} //END OF FOR LOOP
}
/*Function to display Einstein functions, contributions from various
degrees of freedom to enthalpy, heat capacity, Gibb’s energy,
entropy functions and thermodynamic functions of the molecule in the
gaseous state. */
void data::output()
{
outl a(13,5),*pa;
out2 b03(0,3),b24(2,4),b43(4,3),b44(4,4),b40(4,0),b01(0,l), b
ll(1,1),blO(1,0),b21(2,1),bl2(1,2),b22(2,2),b23(2,3),
b20(2,0),b00(0,0),b30(3,0);
out2 *pb03,*pb24,*pb43,*pb44,*pb40,*pb01,*pbll,*pbl0,*pb21, *pbl2,
*pb22,*pb23,*pb20,*pb00,*pb30;
pa = &a; pb03=&b03; pb24=&b24; pb43=&b43; pb44=&b44;
pb40=&b40; pb01=&b01; pbll=&bll; pbl0=&bl0; pb21=&b21;
pbl2=&bl2; pb22=&b22; pb23=&b23; pb20=&b20; pb00=&b00;
pb30=&b30;
cin>>bottom;
cout<<clear;
cout<<”The following are the values of the various Einstein
functions\n″
<<”at different temperatures....... \n” ;
cout<<*pa;
cout<<*pb03<<”TEMP.[K]”<<*pb24<<”EH[I]”<<*pb43<<”ECP[I]”<<*pb44
<<”EG[I]”<<*pb44<<”ES[I]”<<*pb40;
cout<<*pa;
for(i=l;i<=n;i++)
{
cout<<*pb01<<for2<<t[i]<<*pbll<<for2<<eh[i]<<*pbll<<for2<<ecp[i]
<<*pbll<<for2<<eg[i]<<*pbll<<for2<<es[i]<<*pbl0; cout<<*pa;
fun(i);
}
cin>>bottom;
cout<<clear;
cout<<”\nThe following are the values of the Thermo functions for\n”
<<”each of the various statistical contributions and also
their\n”
<<”total values\n”;
cout<<”\nEnthalpy function(cal/mole.K)...... \n”;
cout<<*pa;
cout<<*pb03<<”TEMP.[K]”<<*pb21<<”TRANS.CONT.”<<*pbl2<<”ROT.CONT.”
<<*pb22<<”VIB.CONT.”<<*pb23<<”ENTHALPY”<<*pb20;
cout<<*pa;
for(i=l;i<=n;i++)
{
cout<<*pb01<<for2<<tt[i]<<*pbll<<for2<<httr[i]<<*pbll<<for2<<htr
o[i]
<<*pbll<<for2<<htv[i]<<*pbll<<for2<<ht[i]<<*pblO;
cout<<*pa;
fun(i);
}
cin>>bottom;
cout<<clear;
cout<<”\nHeat capacity(cal/mole.K)........ \n”;
cout<<*pa;
<<*pb22<<”VIB.CONT.”<<*pb20<<”HEAT CAPACITY”<<*pb00; cout<<*pa;
for(i=l;i<=n;i++)
{
cout<<*pb01<<for2<<t[i]<<*pbll<<for2<<cptr[i]<<*pbll<<for2<<cpro[
i]
<<*pbll<<for2<<cpv[i]<<*pbll<<for2<<cp[i]<<*pblO;
cout<<*pa;
fun(i);
}
cin>>bottom;
cout<<clear;
cout<<”\nGibbs energy function(cal/mole.K)....... \n”;
cout<<*pa;
<<*pb22<<”VIB.CONT.”<<*pb20<<”GIBB′S ENERGY”<<*pb00; cout<<*pa;
for(i=l;i<=n;i++)
{
cout<<*pb01<<for2<<t[i]<<*pbll<<for2<<gttr[i]<<*pbll<<for2<<gtro[i]
<<*pbll<<for2<<gtv[i]<<*pbll<<for2<<gt[i]<<*pblO;
cout<<*pa;
fun(i);
}
cin>>bottom;
cout<<clear;
cout<<”\nEntropy(cal/mole.K)..... \n”;
cout<<*pa;
<<*pb22<<”VIB.CONT.”<<*pb23<<”ENTROPY”<<*pb30;
cout<<*pa;
for(i=l;i<=n;i++)
{
cout<<*pb01<<for2<<t[i]<<*pbll<<for2<<str[i]<<*pbll<<for2<<sro[i]
<<*pbll<<for2<<sv[i]<<*pbll<<for2<<s[i]<<*pblO; cout<<*pa;
fun(i);
}
cin>>bottom;
cout<<clear;
cout<<”\nIdeal gas thermodynamic functions..... \n”;
cout<<*pa;
cout<<*pb03<<”TEMP.[K]”<<*pb20<<”HEAT CAPACITY”<<*pb00<<”GIBB•S
ENERGY”
<<*pb03<<”ENTHALPY″<< *pb23<<”ENTROPY”<<*pb30;
cout<<*pa;
for(i=l;i<=n;i++)
{
cout<<*pb01<<for2<<t[i]<<*pbll<<for2<<cp[i]<<*pbll<<for2<<gt[i]<<*pb
ll
<<for2<<ht[i]<<*pbll<<for2<<s[i]<<*pbl0;
cout<<*pa;
fun(i);
}
}
//Function to compute Einstein functions.

void eins(int m, float cons,int i, double *pt, double *pw, double *peh,
double *pecp, double *peg, double *pes)
{
double x, xl, x2, he, cpe, ge, se;
int k;
float d;
init q;
*(peh+i)=q.getx{); *(pecp+i)=q.getx();
*(peg+i)=q.getx(); *(pes+i)=q.getx();
for(k=l;k<=m;k++)
{
x=cons* *(pw+k)/ *(pt+i);
xl=exp(x);
x2=xl-l.0;
he=x/x2;
cpe=x*x*xl/(x2*x2);
ge=log(1.0-exp(-x));
se=x/x2-ge;
*(peh+i)= *(peh+i)+he;
*(pecp+i)=*(pecp+i)+cpe;
*(peg+i)=*(peg+i)+ge;
*(pes+i)=*(pes+i)+se;
}
}
//Function to pause the output.
void fun(int i)
{
out1 ob(13,5);
if(i==7 || i==14 || i==21)
{
cin>>bottom;
cout<<ob;
}
}
//Function main()
main()
{
char ch;
data x;
void (data::*psetdatal)();
void (data::*pdisplayl)();
void (data::*plin_nlin)();
void (data::*painer)();
void (data::*pcal)();
void (data::*poutput)();
data *px;
psetdatal=&data::setdatal;
pdisplayl=&data::displayl;
plin_nlin=&data::lin_nlin;
painer=&data::ainer;
pcal=&data::cal;
poutput=&data::output;
px=&x;
cout<<beep<<clear;
do
{
cout<<clear;
(px->*psetdatal)();
(px->*painer)();
(px->*pdisplayl)();
(px->*plin_nlin)();
cin>>*px;
cout<<*px;
(px->*pcal) () ;
(px->*poutput)();
cout<<clear;
gotoxy(1,10);
cout<<”\nDo you want to process for some other molecule?\n”
<<”\nEnter ‘y or Y’ for yes, anyother key for no.: “;
ch=getch();
} while(ch==’y’ || ch==’Y’);
return 0;
}
/* END OF THE PROGRAM */
Description Of The Input Variables.
1 .nom Name of the molecule

2.1 Nature of the molecule
3.atom Type of atoms
4.na[] No. of atoms of the same type
5.am[] Atomic mass
6.mi Moment of inertia ( gm.cm 2)
7.noa No. of axis
8.xyz[] Moments of inertia along different axis (gm.cm)”)
9.sigma Symmetry number
10.w[] Fundamental vibrational frequencies (cm -1)
11.n Total no. of temperatures
12.t[] Temperatures
Testing data
1. For carbondioxide(CO 2)
nom - carbondioxide or CO 2
1 -1
atom -2
na[l] -1
am[l] -12.0
na[2] -2
am[2] -16.0
mi -71.1 e-40
noa -3
sigma -2.0
w[l] -667.03
w[2] -667.03
w[3] -2349.03
w[4] -1330.00
n -any value from (5 - 27)
t[l] -100.00
t[2] -200.00
t[3] -273.15
t[4] -298.15
t[5] -300.00
2. For water (H 2O)
nom -water or H 2O
1 -2
atom -2
na[l] -2
am[l] -1.008
na[2] -1
am[2] -16.0
mi -0.0
noa -3
xyz[l] -1 929e-40
xyz[2] -1 007e-40
xyz[3] -3.013e-40
sigma -2.00
w[l] -3651.07
w[2] -1595.00
w[3] -3755.08
n -any value from (5 - 27)
Temperatures (i.e. t[1] t[5] ) are same as for carbondioxide
Output at 298.15K (in units cal / molc.k)
Molecule Heat capacity Gibb’s energy Enthalpy Entropy
co2 8.879392 -43.545517 7.509235 51.054661
H2O 8.002221 -37.025913 7.955631 44.981861

CHAPTER 4 RAMAN SPECTROSCOPY
The secret of those who make discoveries is to look upon nothing as impossible.
—Humboldt
4.1 INTRODUCTION
Raman effect is an optical analog of the Compton effect which arises due to the scattering of x-ray photons by
electrons. It was discovered by Sir CV Raman of India in 1928 and Raman was awarded the Nobel Prize in 1932
for this discovery, which led to a period of intense experimental activity for the elucidation of molecular structure,
for locating the various functional groups or chemical bonds in molecules, and for the quantitative analysis of
complex mixtures.
Recall from Chapter 1 that when electromagnetic radiation outside the x-ray region, i.e. visible region, interacts
with pure chemical substance in any of its forms, i.e. solid, liquid, or gas, it may undergo scattering in addition
to absorption and transmission. If I0 is the intensity of incident radiation and Is, It and Ia are the intensities of the
scattered, transmitted and absorbed radiations respectively then mathematically we can write
I0 = Ia+ Is+ It (4.1)
Raman spectroscopy deals with the measurement of intensity of radiation, scattered by the molecules of chemi-
cal substances as a function of wavelength, and the Raman spectrum is thus a graph on which the intensity is plot-
ted on the vertical and wavelength on the horizontal axes. The spectrum of scattered radiation not only consists of
a central line (also called Rayleigh line), the frequency of which belongs to the frequency of source (v0), but also
a series of displaced lines with frequency, say vR on both sides of the central line. This phenomenon of scattering
with modified frequency due to the Kramer–Heisenberg process is known as the Raman effect, combination scat-
tering or Raman scattering.
Raman scattering is generally weak and is ≈1 per cent of the incident radiation or ≈0.01−0.001 per cent of the
Rayleigh scattering for many simple molecules. The Raman scattering process, unlike fluorescence, does not
involve absorption of radiation. Scattering occurs in 10−12 second after excitation whereas fluorescence emission
occurs in ≈10−6−10−8 second after excitation. It also differs from Tyndall scattering (which is due to colloids and
emulsions) and Rayleigh scattering (which is due to ordinary liquids, gases, crystals and glasses) in the sense
that there is no change of frequency or colour of the scattered beam from that of the incident beam. Rayleigh
scattering is valid for particles whose size is not more than 0.1 of the light wavelength of 40−70 nm, while
Tyndall scattering is valid for particles of the size between 1−100 nm. When white light is used in Rayleigh
scattering, there is enhancement of different colours already present in the incident beam. The blue colour of the
sky is due to Rayleigh scattering of the sunlight by the particles of the air with dimensions less than one tenth
of the wavelength of light. The scattering intensity is proportional to v04 law. The lines in the Raman scattering
are frequently arranged symmetrically on both sides of the Rayleigh line. The lines with frequency greater than
that of the Rayleigh line are called anti-Stokes lines while those with lower frequency are called Stokes lines.
The difference between the frequencies of Raman (vR) and Rayleigh (v0) lines matches with the fundamental
vibrational (vvib) or the rotational (vrot) frequency of the molecule of the scattering material. In other words, we
can say that the frequency shift, also called the Raman shift, which is the difference between the frequencies of
λ λ0
Rayleigh line and a particular Raman line, i.e. ΔvR = v0 − vR = R , corresponds to one of the vibrational or
λ R λ0
rotational frequencies of a particular bond within the molecule that gives rise to that Raman scattering line. If vs
and va are the frequencies of the Stokes and anti-Stokes Raman lines respectively then we may write
Raman Spectroscopy 225
Rayleigh Line
Stokes Region Anti-Stokes Region
n s < n0 n a < n0
n0
Fig. 4.1 A sketch showing Raman effect.
νs ν 0 − ν vib (4.2)
ν a = ν 0 + ν vib
νs ν 0 − ν *rot (4.3)
ν a = ν 0 + ν *rot
and the pictorial view of Raman scattering is shown in Fig. 4.1.
4.2 CLASSICAL THEORY OF RAMAN SCATTERING

The origin of Raman scattering can best be understood by considering the behaviour of a molecule under the
influence of an electric field of electromagnetic radiation.
When a molecule is under the influence of a static electric field E, a dipole moment induced into it is given by
M = aE (4.4)
where a, the polarisability of the molecule, occurs because of the distortion of the electron cloud of the molecule
under the impact of E.
Thus, a molecule under the influence of an electric field of the electromagnetic radiation will also have induced
fluctuating dipole moment M, such that
M = a E sin(2p v0t) (4.5)
where v0 is the frequency of incident radiation. This induced moment oscillates with the frequency of the incident
radiation. These forced vibrations of the molecular dipole are the cause of ordinary Rayleigh scattering of radia-
tion. However, by virtue of specific dynamic conditions, a change can occur in the polarisability of the molecule,
this change being due to certain participation also of nuclei in the vibrations of the dipole. These dynamic condi-
tions are due to a forced coupling of electrons and nuclei, as a result of which the forced vibrations of the electrons
(with frequency v0) cause vibrations of the nuclei. The vibrations of the latter which are much slower than those
of the electrons (because of the great difference in the electron and nuclear masses) are what lead to a change in
the polarisability of the molecule. This latter factor alters the frequency of radiation, i.e. light being scattered.
Consequently, displaced lines on both sides of the Rayleigh line appear in the Raman spectrum of molecules.
The polarisability of a molecule (given vibrations of frequency v) is designated by a = a0 + aA sin 2p vt, where a0
is the polarisability of a nonvibrating molecule and aA is the amplitude of change in polarisability during vibra-
tions. Thus, Raman effect in a molecule is due to changes in size, shape or orientation of the polarisability due to
internal motions, i.e. rotation or vibration of the molecule. The molecular polarisability of most of the molecules
is anisotropic and its variation as a function of distance from the centre of mass of the molecule in any direction
is governed by the hypothetical law:
1
r∝ (4.6)
αθ
where aθ is the polarisability in each direction.
The origin of rotational Raman scattering lies in the anisotropy of the polarisability of a diatomic molecule.
The component of polarisability a with respect to the incident radiation along the internuclear axis (say
X-direction) ax, is larger than the components ay and az along the Y-and Z-directions respectively, i.e. ax > ay = az.
Now if νrot is the rotational frequency of the diatomic molecule then
α α y + (α x α y )sin(2πν *rot t ) (4.7)

M [ y (α x α y )s
) i 2πν rot
*
r t ]E
]E ssii ( 0 t)
1
= α y E i ( 2πν t ) + (α x − α y )E [ π (ν − ν r*ott )t ] − cos π (ν 0 + ν rot
*
)] (4.8)*
2
*2 sin A sin B = cos (A− B) − cos (A + B)
The lines in the rotational spectrum of a diatomic molecule will, thus, be positioned at frequencies
ν 0 ν 0 − ν *rot , d ν 0 + ν *rot . Here, ν *rot is the frequency of rotational motion, the frequency with which the polaris-
ability changes twice that of rotational frequency. The polarisability of the molecule returns to its initial value after
a rotation of 1800 as shown in Fig. 4.2 (i.e. returns to its initial values twice on every revolution). Consequently
ν *rot in Eq. (4.7) is 2 νrot. But this is not so in case of vibrational motion of molecules.
E; 0° E
E; 90° a
a
a E; 180°
Fig. 4.2 Changes observed in the polarisability during the complete rotation.
The vibrational Raman scattering of diatomic molecules is because of the change in the molecular polarisabil-
ity a, along the internuclear axis during vibration. The variation of a with Q, the normal coordinate of vibration
with frequency v, is expressed by the relation,
⎛ dα ⎞
α = α0 + ⎜ Q (4.9)
⎝ dQ ⎟⎠ 0
where a0 is the polarisability when the nuclei are in their equilibrium position and (da/dQ)0 is the rate of change
of a0 with Q for infinitesimal nuclear displacements. The hypothetical variation of a with internuclear bond dis-
tance, r is shown in Fig. 4.3.
Now if the change (a−a0) is denoted by Δa and vvib is the vibrational frequency of a diatomic molecule, then
from Eq. (4.7) we write
a = a0 + Δa sin (2p vvjb t) (4.10)
M = [a0 + Δa sin (2p vvib t)] E sin (2p v0 t)
Δα E
= a0 E sin (2p v0 t) + ⎡cos 2π ( − )t − cos 2π ( + )t ⎤⎦ (4.11)
2 ⎣
aθ
Polarisability
X
re
Distance
Fig. 4.3 Hypothetical variation of polarisability of diatomic molecule as a function of internuclear distance.
Thus, the scattered radiation will appear at frequencies v0, v0+ vvjb and v0 − vvib in the Raman vibrational spectrum
of a diatomic molecule.
The classical theory though explains the origin of Raman effect in relation to the position of the Raman lines
in the spectrum, yet it fails to solve the problem of the intensity of the Raman lines as well as the discrete nature
of the spectrum. According to this theory, the intensity of both the lines, i.e. Stokes and anti-Stokes, should be the
same which, of course, is contrary to the experimental facts. Secondly, the Raman spectrum should be continuous
according to the classical theory since the molecular dipole rotates to the electric field of radiation wave, but in
reality Raman spectrum has a discrete linear structure.
The answers to these problems came from the quantum mechanical treatment of Raman scattering.
4.3 QUANTUM MECHANICAL PICTURE OF RAMAN SCATTERING

This model is based upon the Kramer–Heisenberg theory of light dispersion. Accordingly, when a photon of fre-
quency v0 interacts with the molecule of the scattering substance, it may jump from any of its initial vibrational or
rotational level V′ ′ J ′ , where V ′/J ′ = 0, 1, 2, 3,... to any higher vibrational or rotational level EV″ ″ J ″. The higher
level is an unstable level of both the photon and the molecule and its lifetime is of the order of 10−17 second. The
′
molecule in the virtual level subsequently returns to any one of the initial vibrational or rotational levels EV ′ J ′
with the emission of a photon. This process of release of photon from the virtual level to the initial level may be
divided into the following three stages (Fig. 4.4).
(a) Stage I The molecule may return from the virtual state to any of the initial states such that the frequency of the
″ ′
released photon is equal to that of the incident photon, i.e. EV ″ J ″ − EV ′ J ′ = hv0. This will be true for 0 → 0, 1 → 1,
2 → 2, etc., transitions. This phenomenon of emission of photon is called Rayleigh scattering and is due to the
interaction of photon with the nucleus. The energy level diagram for Rayleigh scattering is given in Fig. 4.4 (a).
(b) Stage II The molecule in state V ′/J ′ = 0 absorbs the incident photon and jumps to the virtual level with the
subsequent return to any one of the initial states with the release of photon with lower frequency ( ν s ) than that of
″
the incidence photon, i.e. EV ″ J ″ − EV′ ′ J ′ = hν s . This will be true for 0 → 1, 0 → 2, 0 → 3, etc., transitions. This
is known as Stokes–Raman scattering. The schematic diagram for Stokes–Raman scattering or effect is presented
in Fig. 4.4 (b).
(c) Stage III The molecule in any of the higher initial levels, say V ′/J ′ = 1 gains energy from the incident pho-
ton and jumps to any one of the virtual levels. The molecule subsequently returns to any of the initial levels with
″ ′
the release of the photon with frequency (v a) higher than that of the incident photon, i.e. EV ″ J ″ − EV ′ J ′ = hν a .
This is true for 1 → 0, 2 → 0, etc., transitions. This process of release of photon is called anti-Stokes Raman
effect, and with lower frequency (v s) by the mechanism of Stage II. This will be true for 1 → 2, 1 → 3, etc., transi-
tions. The schematic diagram for the mechanism of release of a photon in Stage III is given in Fig. 4.4 (c).
(a) (b) (c)
3 3 3
2 2 2
Ex (Ev″/J ″)
1 1 1
0 0 0
ns ns na
3 3 3
Ei (Ev ′/J ′) 2 Es 2 Es 2
1 1 1
hn0
0 0 0 Es
Fig. 4.4 Energy level diagram of Raman effect: (a) Rayleigh scattering, (b) Stokes–Raman scattering,
and (c) Anti-Stokes and Stokes Raman scattering.
The three-stage process may also be discussed from the general expression deduced below.
We know that (Fig. 4.4)
hv0 = Ex − Ei (4.12)
hvR = Ex – Es (or Es – Ex) (4.13)
Es Ei
v0 ± vR = (4.14)
h
For Stage I, Es=Et ; for Stage II, Es < Et ; and for Stage III, Es >Ei .
The use of Eq. (4.14) will be made later on while discussing the vibrational and rotational Raman spectra of
diatomic molecules.
4.4 CHARACTERISTIC PARAMETERS OF RAMAN LINES

Normal Raman scattering, also called spontaneous or first-order Raman scattering, occurs when the intensity of
the incident radiation corresponds to that of the source. Normal Raman scattering is linear since the induced dipole
moment varies linearly with the electric field of the incident radiation, i.e. M ∝ E. The important parameters of
Raman lines in the normal Raman scattering are the frequency shift, the degree of polarisation, the intensity, the
width and the shape of the Raman lines.
(a) Intensity of Raman Lines According to Placzek’s theory, the intensity of Raman line IR varies directly as
(v0 − vvib )4 except in the vicinity of Raman resonance scattering, i.e. the intensity of Raman lines depends upon
the frequency of incident radiation. Stokes lines are invariably stronger than their counterparts, i.e. anti-Stokes
lines, since at room temperature the higher levels are thinly populated than the lower levels. The intensity of a
fundamental Raman band is given by
I R = I nm =
const.N ( v0 − vnm )
4
( ′ ′ )⋅ h
(4.15)
1 − exp ( − hv nm T ) 8π v nm
2
where N is the number of molecules in the initial energy state m, and n is the higher energy state a ′ represents
the average of the derivatives of the three principal polarisabilities and b ′ is the magnitude of the derivative of
the anisotropy part of the polarisability of the molecules. The constants 45 and 13 arise in the averaging over all
configurations in the experimental arrangement when the scattered light is observed per unit solid angle at right
angles to the incident direction.
It is found that the ratio of intensities of anti-Stokes and Stokes lines is governed by the Maxwell–Boltzmann
law and is expressed as
− hvrot/vib
Ia
=e κT (4.15a)
Is
where k is the Boltzmann constant and Ia and Is are the intensities of anti-Stokes and Stokes lines, respectively.
However, for accurate determination of Ia/Is, the frequency factor, which is (v0 − vvib)4 for the Stokes and (v0 + vvib )4
for the anti-Stokes lines should be taken into account.
The intensity of anti-Stokes lines grows with the growth of temperature since at higher temperature, the density
of population of higher level increases. Thus, by raising the temperature, we observe anti-Stokes lines which are
missing at low temperatures. The small quanta of radiation favour Stokes as well as anti-Stokes lines while large
quanta favour only Stokes lines.
(b) Width and Shape of Raman Lines Raman lines are generally narrow in gases, liquids and crystals while
in amorphous substances, the lines are broad and diffuse. The width of Raman lines increases with the increase of
temperature and roughly varies directly to the square root of absolute temperature. This increase is because of the
increase in the anharmonicity of vibrations. The shape of Raman lines is independent of the frequency of incident
radiation. An unsymmetrical continuum called Raman wing, adjacent to and on both sides of the Rayleigh line
with continuously decreasing intensity, appears in the case of liquids. The Raman wing arises because in case of
liquids, the rotational motion is hindered. The intensity and width of a Raman wing varies from liquid to liquid
and depends upon the anisotropy of the molecules of the liquid.
(c) Frequency Shifts of Raman Lines Frequency shifts of Raman lines from the Rayleigh line vary from
4,000 to a few wave numbers (cm−1) and are independent of the frequency of the incident radiation. Frequency
shifts grow with the growth of temperature. The frequency-shifts-versus-temperature plot runs parallel to the
width-versus-temperature plot.
(d) Polarisation of Scattered Radiation Raman scattered radiation is at least partially polarised irrespective
of the polarisation of the incident radiation. The extent of polarisation of the scattered radiation depends upon
the structure of the molecule as well as on the symmetry of vibrational motion. It also helps in the determina-
tion of phase changes in liquid crystals and solids when these substances are heated. When a beam of radiation
is incident on a molecule with isotropic polarisability, it scatters. The induced dipole moment is supposed to be
perpendicular to the direction of incident radiation. The scattered radiation may be polarised in a plane parallel
and perpendicular to the plane of polarisation of the incident radiation as shown in Fig. 4.5. The plane of vibration
or displacement is always perpendicular to the plane of polarisation of radiation. Here we are now going to define
a very important quantity called the degree of depolarisation of scattered light r. It is the ratio of the intensity of
scattered light polarised in a plane perpendicular to the plane of polarisation of the incident light to the intensity
of scattered light that is polarised in the parallel plane. Mathematically, it is expressed as
(4.16)
Here, the direction of observation of scattered light with an analyser is perpendicular to the Z-axis in the XY
plane. I ⊥ is the intensity of scattered light parallel to the ZX plane and I is that parallel to the XY plane (Fig. 4.5).
On the other hand, if the incident light is travelling in the −Y direction and the scattered light is observed in the
X-direction, and if the incident light is polarised in the Z-direction, the degree of depolarization is defined by the
equation
(4.16a)
where I (Y) and I (Z) are the observed intensities of light polarized in the Y- and Z-directions respectively. If the
incident light is unpolarised then the degree of depolarisation of the scattered light is given by
(4.16b)
When the polarisability of the molecule remains isotropic during vibration, i.e. Δa = 0 then = 0. However,
when the polarisability of the molecule is anisotropic during vibrational motion then ≠ 0 and the scattered
I⊥
III
X
ion
diat Molecule
t Ra
id en
Z Inc
Fig. 4.5 The parallel and perpendicular components of the scattered radiation. The incident beam is along
the Z-axis and the molecule is positioned at the origin.
radiation will be polarised in the parallel and perpendicular planes. For I⊥ = I , ρp′ = 1, the Raman lines will be
completely depolarised. Thus the value of lies between 0 and 1. The symmetrical breathing vibrations of mol-
ecules such as CO2, CH4 and C6H6 have completely polarised Raman lines, i.e. = 0. In particular, Roman lines
of totally symmetric fundamentals are polarised and those of other fundamentals are depolarised.
Depolarisation ratio plays a major role in ascertaining the symmetry of a vibrational mode participating in the
scattering event and the form of the molecules, i.e. more precisely, the form of the polarisability ellipsoid of the
molecule. In other words, investigations of the depolarisation of Raman lines are important in the assignment
of vibrations and in the final analysis of molecular structure. Let us now frame the rules for the polarisation of
Raman lines. The degree of polarisation depends on the anisotropic character of the molecules. Experimentally,
it has been established that normally the plane polarised Raman lines are sharp and intense, whereas depolarised
lines are diffuse and of weak intensity.
The state of polarisation of the Raman lines is studied by measuring the depolarisation ratio r which is defined as
6β ′ 2
ρ= (4.16c)
45α 2 +7β 2
for unpolarised incident radiation, and for the polarised incident radiation, the depolarisation is defined as
(4.16d)
6 3
For totally asymmetric vibration a ′ = 0, the scattered light is totally unpolarised and ρ = and ρp′ = . How-
7 4
ever, the symmetric vibrations are polarised and the depolarisation ratio lies between 0 and 6/7 for unpolarised
incident radiation and between 0 and 3/4 for polarised incident radiation. Often the former criterion is employed
6 6
for practical purposes. r = 0 − polarised ≥ depolarised.
7 7
Further, it is observed that symmetric vibrations give rise to intense Raman lines and asymmetric vibrations
appear as weak lines and sometimes are unobservable.
In certain cases, polarisation is inverted and r → ∞. The inverted polarisation does not occur in normal Raman
scattering but occurs in resonance Raman scattering. This arises when the polarisation plane of radiation scattered
at right angles to the incident beam lies perpendicular to that of the incident radiation, when this is perpendicular
to the direction of scattering, e.g. some of the bands in the resonance Raman spectra of cytochrome C and
haemoglobin exhibit inverse polarisation.
4.5 EQUIVALENCE OF BEER–LAMBERT LAW OF ABSORPTION

IN RAMAN SCATTERING
Let us suppose that a single molecular species is irradiated with radiation of intensity I0 (w cm−2). The intensity of
the scattered radiation Is at a single Raman frequency will be given by
Is = s I0 (W) (4.17)
where s has the dimensions of cm2.
For N molecules, the total intensity of the scattered radiation is
Is N σ I 0 (w ) (4.18)
If ΔN = the number of molecules/ml, A = cross-sectional area illuminated in cm2, Δx = path length of the illuminated
volume in cm then
N = ΔN A Δx (4.19)
Combining Eqs (4.17), (4.1 8) and (4.19), we get
Is N ) ( AI
( ΔN ) ( Δxx ) kkI Δx (4.20)
where I A I 0 and k = s ΔN is the loss factor in cm−1.

Equation (4.20) is equivalent to the Beer law for the loss of energy from the exciting beam due to Raman
scattering into a given Raman band.
Raman cross-sections are not easily determinable, and a few data are available. The value of s is 10−28 cm2.
Systems with large polarisable atoms or biologically important groups such as I—, —C == C— are likely to be
strong scatterers. Ionic bonds tend to scatter poorly.
4.6 SECOND-ORDER RAMAN SPECTRUM

The Raman spectrum due to overtones and combinational tones is called second-order Raman spectrum. Raman
frequency of the first overtone is not exactly equal to twice the Raman frequency of the fundamental. Similarly, a
combination tone is not equal to the sum or difference of the two fundamental frequencies. This is due to the fact
that the vibrational levels are not spaced uniformly as in the case of a harmonic vibrator. The order of intensity of
overtones and combination bands predicted by theory is vanishingly small as compared to the intensity of funda-
mental vibrations. Consequently, it is very difficult to observe second-order Raman spectra using a conventional
source of radiation, i.e. Hg arc. However, with the advent of lasers it is now possible to record second-order
Raman spectra also.
4.7 GENERAL SELECTION RULE FOR RAMAN SCATTERING

nm
The general selection rule is worked out from the transition probability M for Raman effect, i.e.
E ∫ ψ n α ψ m dτ
nm
M (4.21)
If a is constant, the integral vanishes due to orthogonality of wave functions. Therefore, in order that a particu-
lar vibration or rotation be Raman active, a must change during vibration or rotation, i.e. Δa ≠ 0.
4.8 RAMAN SPECTRA OF DIATOMIC MOLECULES

Raman effect is employed to find the vibrational, rotational, rotational–vibrational spectra as well as the rotational
and vibrational constants not only of homonuclear, say H2, N2, O2, etc., but also of heteronuclear, say HCl, NO,
CO, etc., diatomic molecules. Thus, we can say that Raman spectroscopy complements infrared spectroscopy.
4.8.1 Pure Rotational Raman Spectra of Diatomic Molecules—

Selection Rules for Pure Rotational Spectra
(i) The first selection rule, as worked out from the transition probability for Raman effect, is Δa ≠ 0.
(ii) The second selection rule is established by means of a three-stage scheme of the scattering process (Fig. 4.4).
Considering the transition from Ei → Ex and the selection rule Δ J = ±1, we get two possible values of Jx for
the rotational quantum number of the transition state, i.e
Jx = Ji ± 1 (4.22)
Here Ji refers to the initial state. Upon transition to the final state, i.e. for the transition Ex → Es the rotational
quantum number of Es, must be equal to Js given by
Js = Jx ± 1= (Ji ± 1) ± 1 = Ji ± 2 (4.23)
Hence, the second selection rule for pure rotational Raman spectrum is Δ J = 0, ±2 but not Δ J = ±1 as in the
case of far infrared and microwave spectra. The second factor (i.e. ±2) appears since during a complete rotation,
the polarisability ellipsoid rotates twice as fast as the molecule. It is to be noted that any linear molecule belonging
either to the C∞h or D∞h point group has a rotational Raman spectrum.
Transition between Two Levels We know from Eq. (4.14) that
Es Ei
v0 ± vR =
h
and it is also known that EJ = BhcJ(J+1) for rigid rotator diatomic molecule and EJ = Bhc J(J + 1) − DhcJ 2 (J + 1)2
for nonrigid rotator diatomic molecule.
Let EJ ′, and EJ″, be the energy of initial and final states and J′ and J″ be the corresponding quantum numbers; then
EJ Bh ″(J ″ + 1) (4.24)
EJ Bh J ′ ( J ′ + 1) (4.25)
From Eqs (4.14), (4.24) and (4.25), we get the energy difference ΔE between the two states, i.e.
ΔE = EJ″ − EJ′
ΔE = Bhc( J ″ − J ′ )( J ′ + J ″ + 1) (4.26)
EJ − EJ ′
or Δ = = Bc( J ″ − J ′ )( J ′ + J ″ + 1) (4.27)
h
= v0 ± vR
Now applying the selection rule Δ J = 0, we get v0 = ± vR which is the case of Rayleigh scattering, i.e. the fre-
quency of incident radiation is equal to the frequency of the scattered radiation. When
Δ J = +2, J ″ J ′ + 2, then, vR = v0 − 2Bc (2J′+3)
or v s = v0 − 2Bc (2J′+3) (4.28)
This represents the S-branch of the rotational band. Similarly, when Δ J = −2, we get
v a = v0 + 2Bc(2J ′−1) (4.29)
and it corresponds to the O-branch of the rotational band.
The position of the rotational lines in the O- and S-branches of the Raman band as worked out from Eqs (4.28)
and (4.29) are compared with the experimental observations as under.
J′ Theory Experimental
S O
0 v0 − 6Bc v0 − 2Bc v0 + 6Bc
1 v0 − 10Bc v0 + 2Bc v0 + l0Bc
2 v0 − 14Bc v0 + 6Bc v0 + 14Bc
3 v0 − 18Bc v0 + l0Bc v0 + 18Bc
It is evident from the above data that the position of the first line (S-branch) on the left-hand side of the excit-
ing line is spaced 6Bc from it, while the first line of the anti-Stokes region of the spectrum is positioned at 2Bc
(O-branch) on the left-hand side of the Rayleigh line which seems to be unusual. However, it is observed that in
the anti-Stokes region of the spectrum or the O-branch of the Raman spectrum, J ′ ≠ 0, 1, 2, 3,.. but J ′ = 2, 3, 4,...
Accordingly, the first line in the O-region will also be at 6Bc from the Rayleigh line. The spacing between the rota-
tional lines in the Stokes and anti-Stokes region of the spectrum is 4Bc but not 2Bc as is observed in the infrared
and microwave regions of the spectra of diatomic molecules. The schematic diagram for the theoretical rotational
spectrum of diatomic molecule is presented in Fig. 4.6.
Similarly, we can show that for a non-rigid rotator diatomic molecule, v0 = ±vR for Δ J = 0
and for Δ J = +2 2
⎛ 3⎞ ⎛ 3⎞
v s
v0 −( B D )c J ′ + ⎟ − 8D
Dc J ′ + ⎟ (4.30)
⎝ 2⎠ ⎝ 2⎠
and for Δ J = −2
2
⎛ 3⎞ ⎛ 3⎞
v a = v 0 + (4 B 6 D )c J ′ + ⎟ − 8D
Dcc J ′ + ⎟ (4.31)
⎝ 2⎠ ⎝ 2⎠
Equations (4.30) and (4.31) may also be expressed as
2
⎛ 3⎞ ⎛ 3⎞
vrot ∓ (4 B − 6 D ) J ′ + ⎟ − 8 D J ′ + ⎟ (4.32)
⎝ 2⎠ ⎝ 2⎠
Rayleigh Line
−14B −10B −6B −2B 0 +2B +6B +10B +14B
(a)
Rayleigh Line
−14B −10B −6B 0 +6B +10B +14B
Frequency Shifts from the Rayleigh Line (cm−1)
(b)
Fig. 4.6 Theoretical rotational spectra of diatomic molecule (a) for J’ = 0, 1, 2, 3, ... and (b) for J’ = 2, 3, 4,....

v rot ⎛ 3⎞
= ∓ (4 B 6 D ) − 8D J ′ + ⎟ (4.33)
⎛ 3⎞ ⎝ 2⎠
⎜⎝ J ′ + ⎟
2⎠
Thus, a plot of vrot /(J′+ 3/2) against (J′ + 3/2) will give a straight line with slope of −8D and intercept
= ∓ (4B − 6D). From the slope and intercept, the rotational constants, i.e. B and D, can be determined. The preci-
sion of the measurement does not normally warrant the retention of the term D in Raman spectroscopy.
For molecules Λ > 0 and also for symmetric-top molecules, the selection rules are Δ J = 0, ±1 in the infrared region
of the spectrum. The second selection rule for the rotational Raman spectrum for these types of molecules have been
worked out by the three-stage scheme of scattering process (as done earlier for diatomics, Fig. 4.4) and is
Δ J = 0, ±1, ±2
This rule also holds good for asymmetric-top molecules.
Thus, the rotational Raman spectrum of these type of molecules has in addition to S- and O-branches (Δ J = 0, ± 2),
the usual R- and P-branches (Δ J = ±1). By employing these selection rules, we get from Eq. (4.27), the frequencies
of the S- and O-branches, i.e.
S( J 2)
2) s
0 2Bc (2J ′ + 3) (4.34)
O( J 2)
2) a
0 2Bc (2J ′ − 1) (4.35)
and as well as of P- and R-branches, i.e.
P( J 1)
1) s
0 2Bc (J ′ 1) (4.36)
R( J 1)
1) a
0 c ′
2BcJ (4.37)
The positions of lines in S- and O-branches [from Eqs (4.34) and (4.35)] and in P- and R-branches [from Eqs (4.36)
and (4.37)] are
J′ S O P R
0 v0−6Bc v0−2Bc v0−2Bc v0
1 v0−10Bc v0+2Bc v0−4Bc v0+2Bc
2 v0−14BC v0+6Bc v0−6Bc v0+4Bc

3 v0−18Bc v0+10Bc v0−8Bc v0+6Bc
In order that O- and R-branches of the spectrum should coincide with the experimental spectrum then
J ′ = 2, 3, 4,.. for O-region and J ′ = 1, 2, 3,... for R-branch of the spectrum. The spacing between the rotational
lines in S- and O- regions is 4Bc while it is 2Bc in the P- and R- branches of the spectrum as expected. The sche-
matic diagram for the rotational–vibrational spectrum of these types of molecules is presented in Fig. 4.7.
Thus, knowing B from the rotational spectrum of diatomic molecules, we can determine the intensities of
Raman lines and the bond length of diatomic molecules, e.g. in case of HCl, B =10 cm−1, and we know that
h
B=
8π 2 Ic
where the terms and symbols have their usual meanings, since I = μmr 2.
Therefore,
6 6 × 10 −27 (erg s )
r2 =
8 × 22 / 7 × 22 / 7 × 3 × 10100 ( ) × 35 / 36 × 1.66 × 10 24
)10( −1 )
((g)10(cm
= (1.27×10−8)2 (cm2)
or r = 1.27 ×10−8 cm = 1.27 Å = 0.127 nm
Rayleigh line
S O
R P P R
6B
4B
2B
−2B
−4B
−6B
−14B
−10B
−6B
+6B
+10B
+14B
−6B
−4B
−2B
2B
4B
6B
Shift (cm−1)
Fig. 4.7 Schematic diagram of rotation-vibration spectra of symmetric top molecules and for molecules L > 0.
Using Eq. (4.34), i.e.

v s = vo − 2Bc(2J′ + 3)
we get vrot = 2B(2 J′ + 3)
Since J′ = 0, 1, 2, 3,...
Therefore, for
J = 0 → 2, vrot = 6B = 60 cm–1
1 → 3, vrot = 10B = 100cm–1
2 → 4, vrot = 14B = 140 cm–1
Further from Eq. (4.15), we write
− hcvrot
Ia
=e κT
Is
−6.6 ×10 −17 ( erg s ) ×3×1010 ( cm s −11 ) v rot

Ia 1.38054 ×10 −16 ( erg K −1 )
×
=e
T
Is
vrot
−1.434 ( cm K )
=e T
The intensity ratios for these transitions at room temperature, i.e. 298 K, are
a
J = 0 → 2; I =7.49 × 10 –3
Is
a
1 → 3; I = 6.18 × 10 –3
Is
a
2 → 4; I = 5.09 × 10 –3
Is
It is clear from the data that the intensities of anti-Stokes lines decrease as the rotational quantum number
increases as expected. Similarly, we can also depict the effect of temperature on the intensities of Raman lines for
fixed quanta of radiation. Let us calculate the intensity ratio at three different temperatures, say 300, 310 and 320 K,
on the 0→2 transition, i.e. vrot = 60 cm–1. The ratios at these temperatures are
⎛ Ia⎞ −1.434 × 60
⎜⎝ I s ⎟⎠ =e 300
= 7 50 × 10 −3
300
⎛I ⎞ a −1.434 × 60
⎜⎝ I s ⎟⎠ =e 310
= 7 57 × 10 −3
310
⎛I ⎞ a −1.434 × 60
⎜⎝ I s ⎟⎠ =e 320
= 7 64 × 10 −3
320
From the above data it is clear that as the temperature grows, the intensities of anti-Stokes lines increase while
those of Stokes lines decrease as expected.
The effects of nuclear spin have also been observed as in the infrared spectrum in the case of linear molecules
with centre of symmetry, e.g. O2, H2 and CO2. In case of O2 and CO2, every alternate rotational level is absent
since spin of oxygen is zero, e.g. in O2, every level with even values of J will be missing. In case of H2 and other
molecules with nonzero spin, the spectral lines show an alternation of intensity. Further, in case of linear mol-
ecules with more than three heavy atoms, the rotational fine structure is often unresolved in the Raman spectrum
due to large moment of inertia. Consequently, in such cases direct structural information cannot be obtained from
the rotational Raman spectrum.
The pure rotations of spherical-top molecules (Ix = Iy = Iz), viz. CH4, SiH4, are inactive in the Raman spectrum,
since Δa = 0, i.e. no change in polarisability takes place due to rotation of such type of molecules. However, in
the case of asymmetric-top molecules (Ix ≠ Iy ≠ Iz), all the rotations will be Raman active. The Raman spectrum
of such molecules is complex and is often analysed by assuming the molecule as intermediate between the oblate
(Ix < Iy = Iz) and prolate (Ix > Iy = Iz) types of symmetric-top molecules. For symmetric-top molecules, e.g. CH3F,
CH3CI, etc., (Ix, = Iy ≠ Iz), the rotation about the top axis produces no change in the polarisability but end-over-end
rotations will produce a change in the polarisability. The selection rules are
ΔK = 0, Δ J = 0, ±l, ±2 (except for K = 0 levels when Δ J = ±2 only)
where J = 0, 1, 2,...; K = ±J, ± (J − 1) ,..., 0 and K will have total (2J + 1) values. K is the rotational quantum
number for the axial rotation. Therefore, the selection rule ΔK = 0 reveals that rotation about the top axis will be
Raman inactive. The restriction, K = 0 when Δ J = ±2 imposed on the selection rule means that Δ J = ±1 for transi-
tions involving J = 0 since K = ±J, ± (J − 1)..., 0, Thus, for J ≠ 0, K ≠ 0, the transition according to selection rule
Δ J = ±1, is not forbidden.
4.8.2 Pure Vibrational Raman Spectra of Diatomic Molecules

The selection rules for pure vibration of Raman spectrum are the following:
(i) Δa ≠ 0, i.e. the spectrum is due to a change in the polarisability along the internuclear axis.
(ii) The second selection rule is identical to that of infrared vibrational spectra, i.e. ΔV = ±1, ±2, ±3,...
Transition between Two Vibrational States The energy of a vibrational level is given by the expression
2
⎛ 1⎞ ⎛ 1⎞
EV h e V
hcw hcwe xe V + ⎟
⎝ 2⎠ ⎝ 2⎠
If V′ and V″ are the vibrational quantum numbers for lower and upper states respectively then we may write
2
⎛ 1⎞ ⎛ 1⎞ (4.38)
EV we V ′
hcw h e xe V ′ + ⎟
hcw
⎝ 2⎠ ⎝ 2⎠
2
⎛ 1⎞ ⎛ 1⎞ (4.39)
EV ″ = hcwe V ″ + ⎟ − hcwe xe V ″ + ⎟
⎝ 2⎠ ⎝ 2⎠
From Eqs. (4.14), (4.38) and (4.39), we get
ΔE
= we (V ″ − V ′ ) − we xe (V ″ − V ′ )(
) (V ′ + V ″ + 1) = v0 − vR (4.40)
hc
where ΔE = EV ″ – EV ′
or vR = v0 – [we (V″−V′)−wexe (V″− V′) (V′+ V″ + 1)] (4.41)
Applying the selection rules, we get
(i) For ΔV = +l
vR = v0 – [we − 2wexe(V ′ + 1)] (4.42)
or vs = v0 – [we − 2wexe( V ′ + 1)]

Since v0 – vs = vvib
Therefore, vvib = we – 2wexe(V′+ 1) (4.43)

Since for the 0→1 transition, V′= 0, consequently,
vvib = we (1 – 2xe) (4.44)
For the 1→2 transition
vvib = we (1 – 4xe). This is called the first hot band.
Similarly, for ΔV = –1,
vvib = –we (1 – 2xe) (4.45)
Thus, the frequency of the first line in the vibrational Raman spectrum of diatomic molecules observed at
vvib = ±we (1 – 2xe) (4.46)
shows that the fundamental vibrational line lies on both sides of the excitation line at a space of we (1 – 2xe).
This is true for solids and liquids which have no rotational fine structure.
(ii) For ΔV = ±2,
vvib = ±2we (1 – 3xe) (4.47)
and
(iii) For ΔV = ±3,
vvib = ± 3we (1 – 4xe) (4.48)
It follows that the fundamental, its first, second … overtones are positioned at we (1 – 2xe), 2we (1 – 3xe) and
3we (1 – 4xe) respectively on both sides of the central line as expected. The Raman frequency shifts conform
to the fundamental and its overtones in the infrared spectrum of diatomic molecules. The spacing between two
consecutive lines is not constant, but decreases as ΔV increases. This is due to the anharmonic nature of the
diatomic vibrator. However, if the diatomic vibrator behaves as a harmonic oscillator, the vibrational lines are
observed at frequencies, we, 2we and 3we and the spacing between the consecutive lines is we as expected. The
intensity of these lines decreases sharply and follows the order IΔV (± 1) >> IΔV (± 2) >> IΔV (± 3). Consequently, over-
tones are observed with great difficulties. At room temperature, Is >> Ia since at this temperature, the population
Rayleigh Line
−3ωθ −2ωθ −ωθ 0 +ωθ +2ωθ +3ωθ
νvib(cm−1)
Fig. 4.8 Schematic vibrational Raman spectrum of diatomic harmonic vibrator.
of V = 0 level is higher than V = 1 level. The schematic diagram of a pure vibrational Raman spectrum of
diatomic harmonic oscillator is presented in Fig. 4.8.
Thus, from the position of Stokes or anti-Stokes lines and the Rayleigh line, we can predict the vibrational
frequency and force constant of diatomic molecule, e.g. when H35Cl is irradiated with 4358 Å Hg line, the Stokes
line has been observed at 20057 cm−1. The vibrational frequency of HC135 will be
1
vvib v0 − v s = ( −11
) − 20057( −1
) = ( 22946 − 20057) cm −1 = 2889 cm
( 4358 × 10 −8 )
We know that
o
k ( millidynes /A )
v vib (cm ) = 1303.16
1
μ m (amu )
2889 × 2889 × 35
Therefore, k=
1303.16 × 1303.16 × 36
= 4.77 millidynes /Å
4.8.3 Rotational-Vibrational Raman Spectra of Diatomic Molecules

The rotational fine structure appears in the vibrational spectra of a sample in the gaseous phase. The selection
rules of vibrational (ΔV = ±1, ±2, ±3,...) and rotational (Δ J = 0, ±2) are valid in this case also. Because of the
rule Δ J = 0, a central line (Q-branch) appears in the fine structure which, of course, is missing in the vibrational
structure of the infrared vibrational band since in the infrared spectra, Δ J = 0 is not allowed. Further due to
the rule Δ J = ±2, the analog of P- and R-branches with a spacing of 2B in infrared spectra appears as O and S
with spacing of the order of 4B in the Raman rotational–vibrational spectra. The Stokes lines for V = 0 →1
transition are
ΔJ = 0 v rot vib v 0 − w e (1 2x e ) for all J Q-branch
Δ J = +2 vrot vib v0 − we (1 2 xe ) + B(4J + 6) J = 0, 1, 2,... S-branch

Δ J = −2 vrot vib v0 − we (1 2 xe ) + B(4J − 2) J = 2, 3, 4,... O-branch
The scheme diagram of the theoretical rotational–vibrational spectrum of a diatomic molecule is presented in
Fig. 4.9. The theoretical structure pattern of Fig. 4.9 corresponds to the experimentally observed spectra of O2
and HCl as shown in Fig. 4.10.
Q-Branch
S-Branch O-Branch
Fig. 4.9 Theoretical Raman rotational–vibrational spectrum of a diatomic molecule. Stokes lines for V = 0 → 1 transition.
Q-Branch (1556)
S-Branch O-Branch
1700 1600 1500 1400

Frequency (cm−1)
(a)
Rayleigh Line
Rotation-Vibration Rotation-Vibration
Q Q
Pure Rotation
O S O S
+wθ −wθ
+14B
+10B
+6B
− 6B
−10B
−14B
−1)
Δνvib−rot (cm
(b)
Fig. 4.10 (a) Vibration–rotation band of oxygen, and (b) Schematic diagram for rotation-vibration Raman spectrum of HCI
vapour; B =10 cm−1; we = 2860 cm−1.
4.9 VIBRATIONAL RAMAN SPECTRA OF POLYATOMIC MOLECULES

The vibrational Raman spectra are generally simpler as compared to infrared since fewer combination and
overtone bands appear. Just like vibrational infrared spectra, the vibrational Raman spectra of polyatomic
molecules are more complex than diatomic molecules. However, the basic features of the spectra can be
understood by extending the treatment of the Raman vibrational spectra of diatomic molecules to polyatomic
molecules.
For polyatomic molecules, as in the infrared spectra, we need to know the number of fundamental modes of
vibrations and their symmetry in order to ascertain the Raman activity of each vibration. All molecules having
centre of symmetry, say CO2, exhibit the principle of mutual exclusion according to which all vibrations of a
molecule, with centre of symmetry, which are Raman active are infrared inactive and vice versa. In other words,
if a molecule has centre of symmetry, no fundamental frequency can appear in both the Raman and infrared spec-
tra. The asymmetric stretching (2349 cm−1) and bending (667 cm−1) vibrations of CO2 are infrared active, while
symmetric stretching (1330 cm−1) is Raman active only. In general, symmetrical vibrations give rise to Raman
scattering while antisymmetrical vibrations usually yield strong infrared absorption. Just like infrared spectra, the
different symmetry properties of normal modes are important when considering the rotational fine structure of
the vibrational bands.
4.10 VARIOUS FORMS OF RAMAN SCATTERING

4.10.1 Electronic Raman Effect/Scattering (ERE/ERS)
Normal Raman scattering is associated with the excitation and de-excitation of rotational and vibrational quanta
and is a two-photon process. The two-photon process is feasible provided the transitional levels have the same
parity. This is possible only if the energy gap between the transitional levels is small as in the case of rotational
and vibrational energy levels. Due to this reason, the Raman scattering is suited for low-lying energy levels with
the same parity as the ground state. Since these conditions are not met easily by electronic levels, the electronic
Raman scattering was observed at a much later stage than vibrational and rotational Roman scattering. However,
the electronic energy levels of paramagnetic ions particularly those of rare earths qualify the parity rule, there-
fore, these types of ions exhibit electronic Raman scattering. Electronic Raman scattering has been observed in
Ce, Yb, Eu, Nd, Pr, etc., ions. The NO molecule also exhibits electronic Raman scattering in the ultraviolet and
visible region. To summarise, electronic Raman scattering occurs when the differences (v0 − vR) corresponds to
the energy of electronic excitation of the molecules or ions. In NO, besides the lines vR = v0 ± vvib/rot, we also find
the lines,
vR = ν0 ± (2Π3/2 − 2Π1/2) (4.49)
where 2Π3/2 and 2Π1/2 are the components of the ground state of the NO molecule, i.e. 2Π. The direct combination
of these terms is not allowed by the selection rule ΔΣ = 0. However, the general selection rule for the rotational
and vibrational terms of symmetric molecules according to which two terms that combine with the third do not
ordinarily intercombine, holds good in this case.
4.10.2 Resonance Raman Scattering/Effect (RRS/RRE)

Electronic absorption spectrum of a molecule arises because of the transition between two electronic states of a
molecule. The energy separation between two electronic states is very large. Each electronic state has vibrational
and rotational energy levels associated with it so that electronic transition is spread over a wide range of frequen-
cies and hence the spectrum is called band spectrum.
When the frequency of the exciting radiation in Raman scattering falls in the electronic absorption band spec-
trum of the molecule, there is a dramatic gain in the scattering efficiency (∼102−106). The intensity of Raman
scattering increases as the frequency of the exciting radiation is continuously varied and becomes maximum
at the frequency of maximum electronic absorption. This is known as resonance Raman effect or scattering.
When the exciting frequency is close to or coincides with an absorption transition between discrete energy levels,
e.g. rotational–vibrational level in a molecule of the gas phase, then Resonance Raman Effect (RRE) is called
Discrete Resonance Raman Effect (DRRE). Due to high intensity of scattered radiation observed in RRE, very
low concentration of analyte, i.e. as low as 10−8 mole lit−1, has been observed in contrast to the concentration limit
of normal or spontaneous Raman effect which is greater than 0.1 mole lit−1. Since resonance enhancement is
restricted to the Raman lines associated with a chromophore, Raman resonance spectra consist of a few lines.
The schematic diagram for resonance Raman scattering process is compared with fluorescent, phosphorescent
and absorption emission processes in Fig. 4.11. The RRE differs from normal Raman effect in that the electron
is promoted to an excited electronic state E xe followed by an immediate relaxation to a vibronic or vibrational
e
level, EsV of the electronic ground state, Ei . Fluorescence arises from the radiative transitions from the lowest
vibrational level of the higher electronic state to the various vibrational levels ( sV ) of the ground electronic
state ( ie ). Thus, RRE differs from fluorescence in that the relaxation to the ground state is now preceded by
prior relaxation to the lowest vibrational level of the excited state. Fluorescence is due to the transition between
the states of the same multiplicity, i.e. singlet (S1) → singlet (S0), while phosphorescence is due to transition
between electronic states of different multiplicity, i.e. singlet (S1) → triplet (T1) → singlet (S0). However, in
certain cases, triplet–triplet transitions are also possible, e.g. in the phosphorescence spectra of fluorescein, two
absorption bands corresponding to T1 ← T0 and T2 ← T0 transitions have been observed. Vibrational analyses
of triplet–triplet absorptions have imparted information about the geometry of the triplet molecules and their
interaction with the environment. The energy released in the nonradiative transitions occurring in fluorescence
and phosphorescence goes into the vibrational energy of the molecules. The spacing between the bands in the
fluorescence spectrum is equal to the difference in energy between the vibrational levels of the ground state
while the spacing between the bands in the absorption spectrum is equal to the difference in energy between the
vibrational levels of the excited state. If the spacing of the vibrational levels in the excited state are similar to
those in the ground state, there will be an approximate ‘mirror image’ relationship between the absorption and
fluorescence spectra (Fig. 4.11). The time scale for these processes is quite different. The time for Raman relax-
ation is less than 10−14 second, for fluorescent emission is of the order of 10−6−10−8 second, for phosphorescence
is 10−3−102 second, for vibrational transition is ∼10−13 second, for rotational transition is 10−12 second and for
electronic transition is ∼10−15 second.
The intensity rule of normal Raman effect does not hold good in RRS. The intensity of 1088 and 288 cm−1
Raman lines of calcite increases manifold in the ultraviolet region than given by (v0−vis)4. This effect may be
explained on the assumptions: (i) all the molecules are in nondegenerate electronic levels, and (ii) all are oriented
identically in space. Now, if a single excited electronic level x is effective, the intensity of Raman line of displace-
ment vis depends upon the frequency of the exciting radiation v0 by the relation
2
⎡ 2 ⎤
v xi + v 02 ⎥
C (v 0 ± v is ) ⎢
4
I is (4.50)
(
⎢v ) ⎥
2
⎣ xi v ⎦
where the term C is independent of v0.
S1
Exv
E ex
v″=0 E xv (of Triplet State)
v ″=0
hnix
hnxs
h(n0−nis)
S0
E vs Esv Esv
v ′=0 v ′=0
hn0
hnis
E ei E ei
E ei
(a) (b) (c)
Emission Intensity
Absorption or
Absorption Fluorescence Phosphorescence
Wave Number (cm−1)

Fig. 4.11 Schematic energy level diagram of (a) resonance Raman scattering, (b) fluorescence emission and absorption processes, and
(c) phosphorescence process. The wavy arrows stand for radiationless relaxation.
This effect is used to supplement absorption and fluorescent studies and for investigating the structural changes
in molecules and excited electronic states. This process has been used to study biological molecules, viz., haemo-
globin, oxyhaemoglobin, de-oxyhaemoglobin, cytochrome-C, rubridoxin, carotenoid pigments, vitamin A type
molecules and bacteriorhodospin, etc., under physiologically significant conditions. Oxidation state and spin of
iron atoms in haemoglobin and cytochrome-C have been determined. The resonance Raman bands are due solely
to vibrational modes of the tetrapyrrole chormophore group of these molecules. A major limitation of RRE-like
normal Raman effect is the interference by the fluorescence either by the analyte or by the other species present
in the sample.
4.10.3 Stimulated Raman Scattering/Effect (SRS/SRE)

This is a nonlinear Raman scattering involving a four-photon process, i.e. two with the frequency of incident
radiation and one each with the frequency of Stokes and anti-Stokes Raman lines. SRS is coherent and can be used
as a source of laser radiation. This process occurs when one or more high-intensity lasers are used as exciters. The
intensity of the exciters should be greater than the threshold intensity for normal Raman scattering, i.,e. >1 mW.
The intensities of the stimulated Raman lines grow dramatically under these conditions and is greater than that of
the normal Raman lines by a factor of e10 to e20. When the intensity of incident radiation is very high, higher order
Stokes lines with progressively diminishing intensities are observed with frequency shifts corresponding to 2vvib,
4vvib, etc. Anti-Stokes lines may also be observed. The Stokes and anti-Stokes lines are highly directional. The
fundamentals and octaves are not equally spaced in SRS. Since a limited number of Raman lines are observed in
SRS, it is not good for study of structure of molecules and chemical processes.
The first such effect was observed in nitrobenezene excited with a ruby laser at 6943 Å which gives rise to
Stokes radiation at 7650 Å with an intensity of over 25 per cent of ruby laser. The direction of scattered radiation
was almost collinear with the incident laser radiation.
The vibrational bands obtained in some of the molecules by SRS are CC14—460, CS2—655, C6H6—-990,
3064, styrene −999, 1631, 3056, benzonitrile—1002, 2229 and acetone—2921 cm−1.
4.10.4 Spin-Flip Raman Scattering/Effect (SFRS/SFRE)

In this process, Raman scattering is because of the flipping of the spin of the electronic state between two opposite
directions under the action of a photon. The shift in this process is twice the cyclotron frequency of the electron.
4.10.5 Continuous Wave Stimulated Raman Gain Spectroscopy (CWSRGS)

The gain in the Raman scattering is achieved via the polarisability term b, using two laser sources of high intensities.
The intensity of one of the laser beams (probe) is enhanced by the other (pump). High resolution is attained with
two continuous single-frequency lasers. This process of achieving Raman gain is called continuous- wave stimu-
lated Raman gain spectroscopy.
4.10.6 Inverse Raman Effect/Scattering (IRE/IRS)

When a sample is excited simultaneously by a monochromatic radiation of high intensity and a continuous
radiation beam with a well-defined absorption band (continuum), a sharp absorption spectrum corresponding to
the Raman transitions is observed on the anti-Stokes side. This is called inverse Raman scattering. The absorption
spectrum is due to the absorption of vibrational quanta from the continuum. Since no threshold is found in IRE,
the complete Raman spectrum may be obtained in principle. IRE may be employed to study gases, liquids, solids
and even free radicals provided a suitable broad continuum and pulse lasers are available.
4.10.7 Hyper Raman Scattering/Effect (HRS/HRE)

The dipole moment induced in a molecule by high electric field of incident laser radiation of high intensity is
given by the expression
1 1 3 ...
M α E + βE 2 γE + (4.51)
2 6
The molecular distortions caused by the weak electric field of radiation are insufficient to change the bond
energies in the molecule. Consequently, the induced dipole moment varies directly as E and the first term a is the
cause of normal Raman effect. However, under high electric field it is not so and the higher terms, viz., b, g, etc.,
i.e. second- and third-order hyperpolarisability tensors, etc., which perturbs the linear behaviour appears. The
higher terms become significant only when E is very high. The second term b is the cause of hyper Raman effect
and the intensity of Raman lines in hyper scattering depends upon this quantity.
This is a very weak nonlinear incoherent scattering process and is observed along with stimulated Raman
effect. This is due to high electric field of incident laser radiation with frequency v0 (Ruby laser) and occurs in
the neighbourhood of frequency 2v0 as exciting radiation with frequency shifts corresponding to vibrational,
rotational or phonon frequencies, i.e. the Raman lines in this effect appear as 2v0 ± vvib rather than in the normal
Raman form v0 ± vvib, or in other words, hyperfine scattering occurs when non-linear inelastic scattering takes
place with higher harmonics of the incident radiations as exciter. The scattering rules of normal Raman scatter-
ing are not valid in the case of hyper Raman scattering. Consequently, the normal modes of vibration which are
infrared and normal Raman inactive may be hyper Raman active. H2O, CC14, CH3CN, etc., exhibit hyper Raman
effect. Soft mode of SrTiO3 (Tc = 110 K) is inactive in normal Raman scattering but is active in hyper Raman scat-
tering (a strongly anharmonic vibration with the frequency tending to zero for T → Tc is called a soft vibration),
e.g. the soft mode of Hg2Cl2 appears at 12.5 cm−1 at room temperature (T) and tends to zero for T → Tc (185 K),
Tc being the phase transition temperature.
4.10.8 Coherent Anti-Stokes Raman Scattering/Effect (CARS/CARE)

CARS is a nonlinear Raman scattering process which arises due to the coherent polarisability term g of the mol-
ecule. A molecule is simultaneously excited with laser exciters with frequencies v0 and v′0 such that v0 > v′0 coher-
ent scattered radiation will be produced at frequency 2v0 − v′0 only if (v0 − v′0) corresponds to one of the Raman
frequencies vRa of the molecule, i.e. v0 − v′0 = vRa . CARS is very intense, coherent and directional. Therefore, due
to its coherency and directionality, it can be isolated from incoherent fluorescence if any in the medium by spatial
filtering. Further, CARS will be free from fluorescence since the frequencies of anti-Stokes lines are on the higher
side of the excitation frequencies. Coherent anti-Stokes Raman spectrum is recorded by keeping the frequency
v0 (pump) constant and varying v′0 (probe) to cover the entire range of the Raman spectrum. CARS is used to
determine the vibrational and rotational constants of molecules very accurately. It is used for measurement of
air pollutants in the environment and study of transient molecules, i.e. short-lived species produced in various
chemical and physical processes. Temperatures in combustion and jet engines, industrial furnaces and their
exhausts can be measured from the temperature dependence of intensity of scattered radiation.
4.10.9 Coherent Stokes Raman Scattering/Effect (CSRS/CSRE)

This is similar to CARS except that CSRS is produced at frequency 2 v′0 − v0 provided v′0 − v0 coincides with a
Raman frequency vRs of the medium, i.e. v′0 − v0 = vRs .
4.10.10 Higher-Order Raman Spectral Excitation Scattering/Effect

(HORSES/HORSEE)
This is higher-order mixing process, e.g. six photons or second-order four photons mixing, which occurs when
the intensities of the coherent anti-Stokes Raman ftequencies, i.e. 2v0 − v′0 and Stokes Raman frequencies, i.e.
2v′0 − v0 are very large. These frequencies may further interact with the frequencies of incident beams to produce
scattered radiations at still higher frequencies, e.g. 3v0 − 2v′0. This higher-order scattering process is known as high-
er-order Raman spectral excitation scattering. The second-order Raman spectra may be observed by HORSES.
4.10.11 Surface-Enhanced Raman Scattering/Effect (SERS/SERE)

The intensity of Raman scattered radiation grows by a factor of 103–106 when the molecules adsorbed on metal
surfaces are involved in scattering. This is known as surface enhanced Raman scattering. Consequently, detection
limits in the range 11−9−10−12 mole lit−1 have been observed. When pyridine is adsorbed on a silver surface, the
intensity increases by a factor of 6. The enhancement depends upon (i) type of adsorbent and adsorbate, (ii) nature
of metal surface whether it is a film or colloidal particle, and (iii) on the nature of adsorption, i.e. chemical adsorp-
tion or physical adsorption. This effect arises because of the interaction between the conduction electrons in metals
and adsorbate, i.e. the adsorbed molecule on the metal surface. The coupling occurs through an electric field (called
image electric field) generated in the metal by the adsorbate. The induced dipole moment is then expressed as
M = a(E + Ead) (4.52)
The higher polarisability terms are negligible. Ead becomes very large when the frequency of incident radiation
is in resonance with the normal mode of the conduction electrons in the metal. Now since M ∝ (E + Ead) rather
than E, the intensity increases manifold. This effect is used for studying corrosion of metal surfaces, vibrational
structure of chemical entities and biomolecules available in small quantities since the cross-section increases on
adsorption. Surface microstructure of industrial materials may also be determined by this process.
4.10.12 Time-Resolved Raman Spectroscopy (TRRS)

When the normal Raman scattering is recorded instantaneously using picosecond pulses from high-intensity lasers,
the spectroscopy is called Time-Resolved Raman Spectroscopy (TRRS or TR2S) while when the time-resolved
resonance Raman scattering is observed using picosecond pulses, the spectroscopy is known as Time-Resolved
Resonance Raman Spectroscopy (TR3S). Time-resolved Raman spectroscopy is used for the study of mechanism
involved in a number of photochemical, photophysical and photobiological processes. Transient species with
lifetimes of the order of 0.1 picoseconds have been studied.
4.10.13 Raman Optical Activity (ROA)

The small differences between the intensity of Raman scattered radiation from left and right circularly polar-
ised radiation is known as Raman Optical Activity (ROA) and is shown by optically active substances. ROA is
expressed in terms of a parameter called Circular Intensity Differential (CID) and is defined as
IR − IL
δ= (4.53)
IR + IL
where IR and 1L are the intensities of scattered radiation from right and left circularly polarised incident radiation.
4.10.14 Raman-Induced Kerr Effect Spectroscopy (RIKES)

When a medium is irradiated simultaneously with a weak linearly polarised beam of radiation with frequency
v0 (probe) and a circularly polarised beam of frequency v′0 (pump) from a high-power laser, a new beam with
frequency v results. The new beam is because of the nonlinear interaction between the frequencies v0 and v′0 and
is perpendicularly polarised with respect to probe. The new beam exhibits resonance only when ν ′ 0 ν 0 is near
the frequency of the Raman active vibration of the medium, i.e. ν ′ 0 ν 0 = ν vib . This process of induction of Kerr
effect, only at Raman shifted frequencies of a medium by a laser beam is known as Raman induced Kerr effect
spectroscopy. Some organic liquids exhibit this effect.
4.11 BASIC PRINCIPLES OF A RAMAN SPECTROMETER

To obtain a good Raman spectrum, the intensity of incident radiation should be very strong since the Raman scatter-
ing is generally weak and secondly the irradiating beam should be highly monochromatic as Raman lines appear on
both sides of the Rayleigh line. The flowsheet diagram of a basic laser Raman spectrometer is shown in Fig. 4.12.
M
RF
S FL SC
S : Source
RF : Radiation Filter CL
FL : Focusing Lens
SC : Sampling Cell
M : Mirror for Multireflection of
Incedent Radiation
CL : Collecting Lens SP
SP : Spectrometer
AN : Analyser
AN
Fig. 4.12 Flowsheet diagram of a simple Raman spectrometer.
It consists of (i) a suitable stable gas laser source (S), and (ii) an optical system comprising RF, FL and M to
focus the beam on or into the sample as well as CL to feed the scattered radiation to SP for counting the scattered
photons as a function of frequencies. This analysis is then displayed as a plot of scattering intensity versus fre-
quency by the recorder. Typical hypothetical and experimental Raman spectra are presented in Figs 4.13 to 4.15.
Before the advent of a lasers, mercury arcs were used as a source of radiation, i.e. light- and most of the work
was carried out at 4358 (2s → 2p2) and 5461 (2s →2p1) Å lines of mercury. An aqueous solution of sodium nitrite
or quinine sulphate filters out the undesirable 4047 Å (2S →2p3) line of mercury. Saturated solution of iodine in
carbon tetrachloride filters out 4358 Å line of mercury. These days, the commonly used lasers for Raman excita-
tion are (i) continuously operating He–Ne (6328 Å), Ar+ ion (4880, 5145 Å) and Hg-Cd (4415, 3250 Å) gas lasers;
(ii) pulsed and the continuously operating forms of Ruby (6943 Å) and the Nd-YAG (1.06 micron) solid state
lasers; (iii) tunable dye lasers, e.g. rhodamine 6G laser, with continuously variable wavelength in the visible region
(0.43−0.65 μm); and (iv) UV lasers (184−260 nm) have been developed by the principle of repeated stimulated
anti-Stokes Raman scattering in hydrogen under pressure and thus shifting the harmonics of Q-switched Nd-YAG
laser. The power of lasers used for linear Raman spectroscopy, i.e. M ∝ E, is of the order of 50 mW−2 W.
Rayleigh Line
Stokes Anti-Stokes
− +
0
Δν–(cm−1)
Fig. 4.13 Hypothetical laser Raman spectrum.
459
Stokes Lines
313
217
Exciting Line 6328 Å
Anti-Stokes Lines
217 459(cm−1)
313
(cm−1)791 761
Fig. 4.14 Laser Raman spectrum of carbon tetrachloride.

1004
786
28 A
Exciting Line 63
(cm−1) 1030
521
216
623
993 814 344
730
Fig. 4.15 Laser Raman spectrum of toluene in the region of up to 1100 cm−1.
The choice of lasers which, of course, are interchangeable for specific type of studies are most important
because of their high cost. The He–Ne (100 mW, 6328 Å) laser is frequently used due to its narrow wavelength.
With this laser source we need not (i) cool the sample, and (ii) use filters to get monochromatic beam of radiation
as is done in case of Hg-arc source. Further, Raman spectra can be recorded using 0.03 ml or less of the ana-
lyte whereas 10 ml of liquid or several litres of gaseous sample are required when Hg arc is used as a source of
radiation. Fluorescence emission in the sample which could be excited by 4358 Å line of Hg complicates Raman
spectroscopy and may be completely eliminated by careful sample and reagent purification and using 6328 Å
line of He–Ne laser as a source of excitation. The other main advantages of laser sources are (i) the second-order
Raman spectra can be recorded; (ii) the polarisation of laser beam is well-defined and may be controlled within
0.1 per cent; (iii) the broadening due to Doppler effect can be minimised; and (iv) the width of the laser line is of
the order of 0.005 cm−1 or less; therefore, precise information could be obtained on the width and fine structure
of Raman bands.
The Krypton-ion laser with lines at 5682 and 6471 Å is useful for the samples absorbing at 5000 Å except for
resonance Raman spectral work. Generally, the best choice is an argon laser with lines at 5145 or 4880 Å with
high output power. The additional advantage of argon-ion laser is that the detector efficiency and Raman scat-
tering cross-section is great near 5000 Å. These factors and high power levels are a must for work with dilute
solutions. When the background fluorescence is low, the signal-to-noise ratio of Raman band approximates to the
square root of the product of analyte concentration, time of measurement and laser power. Fluorescence declines
most rapidly at higher laser power and low irradiating wavelength. The presence of fluorescence indicates absorp-
tion of radiation, some of which is dissipated as heat and causes localised heating and hence analyte damage,
especially when the analyte is a biological sample.
The sample cells used in Raman spectroscopy are made up of nonfluorescent fused quartz and have well-
polished flat bottoms. The top of the cell is sealed with a transparent cap to avoid dust and is designed to permit
deoxygenation of the sample. The cell is completely filled to avoid reflections from the curved air–liquid inter-
face. The spectra of clear coloured or colourless liquids, suspensions or moderately turbid solutions are recorded
by focusing the beam vertically through the bottom window into the sample and then collecting the scattered
radiation in a direction perpendicular to the incident beam. The spectra of solids are recorded in capillaries, but
solids give too much Rayleigh background.
The samples which are sensitive to laser wavelengths, e.g. CBr3COCl, CCl3COCl, etc., are likely to be degraded
and hence the spectra of such samples are recorded by the rotating cell technique. The Raman spectrum of CBr3-
COCl recorded by this technique is shown in Fig. 4.16.
The irradiation of biological substances absorbing at laser wavelength leads to local heating which may
degrade some biological molecules. In these type of biological systems, e.g. cytochrome-C, haemoglobin, etc., the
resonance Raman scattering of laser radiations from vibrational modes of biological molecules can be recorded
because of the high detection efficiency (10−8 mole lit−1) of RRS.
The characteristic Raman wave numbers for certain univalent and multivalent bonds which are averages of a
large number found for the homologous series are listed in Table 4.1.
1800 1600 1000 800 600 400 200 0

Frequency (cm−1)
Fig. 4.16 Laser Raman spectrum of CBr3COCl.
Table 4.1 Raman vibrational frequencies (in cm−1)

of some univalent and multivalent bonds.
Bond Bond Frequency
H—H 4158
C—H (aromatic) 3050
C—H (aliphatic) 2924
N—H 3370
O—H 3650
S—H 2572
S—D 1865
C—C 993
C=C 1620
C≡C 2120
C—N 1033
C=N 1650
C≡N 2150
C—O 1030
C=O 1700
C≡O 2146
S—S 509
C—S 650
C=S 1225
C—Cl 710
C—Br 597
C—I 526
4.12 APPLICATIONS OF RAMAN SPECTROSCOPY

Due to simplicity of this technique, studies of Raman spectra have many advantages over infrared spectra. The
possibility of detecting directly the frequencies of infrared inactive vibrations makes the method of Raman
scattering particularly valuable. This method is likewise very important for studying the structure of molecules, the
structural changes occurring in molecules due to association, dissociation, solvation, etc., in studies of chemical
equilibria, for analysing complex mixtures, for identifying compounds, and to find the energy difference between
rotamers, etc. Raman technique is also useful for studying the kinetics of fast reactions.
Laser Raman spectroscopy enjoys a significant biological advantage over infrared spectroscopy since water
interferes to a far less extent. Consequently, the scattering from water can largely be ignored, e.g. the scattering
from the water OH bending vibration in the amide 1 region (1650 cm−1), although significant in intensity, is rela-
tively weak compared to its infrared absorption. The water background in Raman spectroscopy can be overcome
either by using high concentration of biological samples, e.g. polypeptides like poly-L-alanine, poly-L-proline
and polyglycine or protein, or by using computer-assisted difference techniques. With the latter technique, it is
possible to analyse even very low concentrations of the analyte in solution or suspensions. The Raman vibrational
bands of H2O and D2O are recorded in Table 4.2.
Table 4.2 Raman vibrational bands (in cm−1) of H2O and D2O.
Vibration H20 D20
0—X stretching 3493 (vs, vb), 3300 (s), 2532 (vs, vb), 2400 (s),
3600,3500 (mc) 2600 (rnc)
Association vw,b —
X—O—X' bending 1640 (m) 1280(m)
Libration (three bands) ~450−780(w) 375−550 (w)
Hindered translation ~152−175 (w), 60 (w) ~1 52−1 75 (w)
b—broad, vb—very broad, s—strong, vs—very strong, m—medium, mc—minor components, w—weak,
vw—very weak, X = H or D.
4.12.1 Study of Environmental Effects on Molecular Systems

The molecules are generally affected by their environments. The shift in the Raman lines of the molecule with
change of environment may, therefore, provide valuable information about the environmental effects since Raman
lines are sensitive to inter as well as intra-molecular interactions, e.g. the appearance of two SH bands at 2568/1863
and 2552/1853 and two C==S bands 1225/1212 and 1220/1198 cm−1 in the Raman spectra of CH3CSSH/CH3CSSD
in CH2Cl2 and a single band at 2498 cm−1 in the Raman spectra of neat acid is an evidence that dithioacids exist as
an open-chain dimer (I) in the solution phase while cyclic dimer (II) in the virgin liquid phase as shown below.
1225 cm−1 S
CH3 C
S H S 1220 cm−1
2552 cm−1 C CH3
H S
2568 cm−1
(I)
2498 cm−1
S H S
CH3 C C CH3
S H S
(II)
A comparison of laser Raman spectra of NbOCl3 in the gas (at 300 °C) and solid phase reveals that a
shift occurs in the Nb = O stretching frequency from 997 cm−1 in the gas phase, characteristic of Nb = O double
bond, to 770 cm−1 in the solid phase characteristic of a bridged oxygen. Thus, it appears that in gas phase, the
NbOCl3 molecule is a simple tetrahedron, although it has polymeric structure in the solid state as shown by
x-ray crystallography.
4.12.2 Mechanism of Tautomerism and Polymerisation

The tautomerism and polymerisation established by other methods may be confirmed by Raman spectroscopy.
The two Raman lines at 1632 and 1725 cm−1 which are characteristic of ethylenic bond have been observed in
the Raman spectrum of ethylacetoacetate (CH3COCH2COOCH2CH3) but are absent in the spectrum of ethyldim-
ethylacetoacetate [(CH3)2HCCOCH2COOCH2CH3)] which shows two bands characteristic of the carboxyl group
at 1707 and 1738 cm−1. From this we infer that ethylacetoacetate exhibits keto-enol tautomerism. The intensity of
the Raman band due to the double bond of methyl-methacrylate decreases gradually during polymerisation and
the band almost disappears in the final polymers. This leads us to infer that polymerisation occurs because of the
opening of the double bond. Further, by monitoring the intensity of double bond during the course of polymerisa-
tion, we can study the kinetics of polymerisation also.
4.12.3 Conformational Equilibria H
X
The results on conformational equilibria may be supplemented
H
by Raman spectroscopy as well. Let us consider the conforma-
tional equilibria. X
where X = F, Cl, Br, I, OH, etc.
The frequencies of the Raman bands due to C—X link in the two conformers say axial (a), and equatorial
(e) are different due to the steric interactions with the adjacent protons, i.e. vC X (e) > vC X ( a) . The intensity is
directly proportional to the amount of the scattering species, i.e.
Ie ∝ Ce
I a ∝ Ca
I e Ce αe ⎡ α ⎤
Thus, K= = ⋅ ⎢assume e ≈ 1⎥ (4.54)
I a Ca αa ⎣ αa ⎦
Here, Ie and Ia are the intensities of the bands at vC X (e) and vC X ( a) respectively, Ce and Ca stands for the concen-
trations of the respective conformers. Thus, by simply monitoring the intensities of the vC−X bands corresponding
to the two conformers, we can determine K, the conformational equilibrium constant. The free energy difference
for the two conformers in most of the compounds lies below 1 kcal mole−1.
4.12.4 Study of Ionic Equilibria

Let us consider the dissociation of a mineral acid (HX) in water,
K
HX + H 2 O H 3O + + X −
where K is the dissociation constant of the acid. The scattering intensity is directly proportional to the concentra-
tion of the scattering species, i.e.
−
IX− ∝ [X ]
IHX ∝ [HX]
I X⎯⎯ α ⎡⎣ X − ⎤⎦ ⎡ α ⎤
K HX = = assume ≈ 1⎥ (4.55)
I HX α ′ [ HX ] ⎣⎢ α′ ⎦
−
Thus, by simply monitoring the scattering intensities of the vibrational bands of X and HX respectively, we
can determine the value of K from Eq. (4.55).
Similarly, we can study the following equilibria.
H2 X + H2O H 3O + + HX −
β ⎡⎣ HX ⎤⎦ ⎡
−
I HX − β ⎤
K H2 X = = ⎢ assume ≈ 1⎥ (4.56a)
I H2 X β ′ [H 2 X ] ⎣ β′ ⎦
and HX − + H 2 O H 3O + + X 2 −
I X 2− γ ⎡⎣ X 2 − ⎤⎦ ⎡ γ ⎤
K HX − = = ⎢ assume ≈ 1⎥ (4.56b)
I HX − γ′ ⎡⎣ HX ⎤⎦ ⎣−
γ′ ⎦
Raman spectra of aqueous solutions of nitric and sulphuric acids of moderate concentrations give definite proof
of the existence of unionised molecules. It is observed that sulphate ions are formed in very dilute solutions of
sulphuric acid.
4.12.5 Study of Hydrogen-bonded Equilibria

Let us consider the following hydrogen-bonded equilibria.
K
A H+B A H...B
where A or B = O, N, S, F, etc.
The A—H stretching frequency will be modified as a result of H-bond formation. Thus, by measuring the
scattered intensity of the free A—H and H−bonded A−H bands as function of concentration of B, the value of
equilibrium constant K can be determined by the expression
K=
[ − ] (4.57)
[ − ][ ]
o-chlorophenol has two non-equivalent configurations in solution H H
of carbontetrachloride, i.e. O O
The cis form is stabilised by the influence of OH ••• CI interac- Cl Cl

tions. Consequently, two bands corresponding to free OH (trans)
and H-bonded OH, i.e. O—H ••• CI (cis) should appear in the
Raman spectrum of o-chlorophenol. The hydrogen bond equilib- trans cis
rium constant K can then be determined from the expression
I cis
K= (4.58)
I trans
where Icis /Itrans is intensity ratio of the OH stretching bands of the H-bonded and free OH groups. The intensity
ratio of the first infrared overtones of the OH stretching band at 6910 cm-1 for the cis and at 7050 cm−1 for the trans
isomer has been found to be of the order of 10.
4.12.6 Nature of Chemical Bond

The totally symmetric vibrations of the tetrahedral complex ML4 ( ZnCl 2− −1 2− −1
4 , 275 cm ; CdCl 4 , 260 cm , PO 4 ,
3−
−1 2− −1 − −1 −1 −1
938 cm , SO 4 , 982 cm ; ClO 4 , 928 cm ) and octahedral complex ML6 (SF6, 775 cm ; UF6, 666 cm ) are
Raman active only. The force constant for the M−L bond can be estimated from the frequency of the totally symmet-
ric vibration employing the equation, vM−L = 1303.16 k / μ m where k is the force constant in millidynes/Å and μm is
the reduced mass in amu provided νM−L is in cm−1. The high values of force constants for 3−
4 (5.06), SO 2−
4 (6.04);
− 2− 2−
ClO 4 (5.58) as compared to those of ZnCl 4 , (1.02); CdCl 4 , (1.07), SF6 (4.21) and UF6 (4.60) indicate the pres-
ence of dπ − pπ bond between the central atom and oxygen atom in addition to the σ bond in the former systems.
4.12.7 Molecular Structure

The structure of molecules and complex ions can be deduced from the activity and depolarisation ratios of the
Raman bands.
Let us first explore how depolarisation ratio could be helpful for structural determination of XYZ2 type molecular
systems. These systems can have either planar or pyramidal structure. All of its six fundamental modes of vibration
are Raman as well as infrared active. For the planar structure, the three modes should be polarised and three should
be depolarised while four should be polarised and two should be depolarised for the pyramidal structure. Accord-
ingly, the number of polarised and depolarised Raman bands is structure specific. Therefore, simply on the basis of
depolarisation selection criterion we can assign either planar or pyramidal structure to XYZ2 systems, e.g. out of
the six fundamental modes of vibrations of SOCl2, the four bands at wave numbers 1230, 490, 344 and 192 cm−1 are
polarised and two at 445 and 284 cm−1 are depolarised; consequently SOCl2 has a pyramidal structure.
Structural elucidation by the activity selection criterion is also possible. Let us apply this criterion to XY3
systems which can have either trigonal or pyramidal structure. Out of the six modes of vibrations, only four are
observable, since the two fundamental modes are doubly degenerate. From the activity of Raman bands listed in
Table 4.3, we can take decision regarding the structure of XY3 systems.
Table 4.3 Activity of XY3 systems.
Pyramidal XY3 Trigonal XY3
No. of Modes Activity No. of Modes Activity
2 R(p) (||) 1 R(p)

1 IR(||)
2 R(dp),IR(⊥) 2 R(dp), IR(⊥)
p = polarised; dp = depolarised
For example, the four observable normal modes of vibration of SO3 are at wave numbers 1332, 1068, 530 and
653 cm−1. The bands at 1332, 1068 and 530 are Raman active only. As a consequence thereof, SO3 has a planar
or trigonal structure. On the other hand, in case of NH3, out of its four Raman observable bands, viz., 3337, 950,
3414 and 1628 cm−1, only two bands at 3337 (p) and 950 (p) cm−1 appear in the spectrum. The fact that the two
doubly degenerate vibrations at 3414 and 1628 cm-1 have not been observed in the Raman spectrum is in agree-
ment with the usual weakness of Raman lines corresponding to non-totally symmetric vibrations. Consequently,
NH3 has pyramidal structure.
The criterion of ‘similarity’, i.e., molecules having similar spectra will have similar structural pattern, can also
be employed to establish the structure of molecular systems, e.g. the Raman spectrum of tetramethylammonium
ion [(CH3)4N]+ is similar to that of tertiary butane (CH3)4C which has a regular tetrahedral structure. This lends
support to the regular tetrahedral structure established by stereochemistry of [CH3)4N]+ ion. Thallous ions may
either exist as Tl+ or Tl2++ in aqueous solutions. Only a single Raman band appears in the Raman spectrum of this
ion and is assigned to Tl—Tl bond stretching vibration. From this it is inferred that thallous ions exist as Tl2++ but
not as Tl+ in aqueous solutions. Mercurous (Hg2++) and argentous (Ag2++) ions also show similar behaviour.
4.12.8 Energy Difference Between Rotamers

The intensity of Stokes lines is directly proportional to
−1
⎛ − hv
⎞
⎜⎝ 1 − e κT
⎟⎠
−1
s ⎛ − hv
κT
⎞
i.e. I 1 e ⎟⎠ (4.59)
⎝
Let It, and Ig be the scattered intensities, vt and vg the vibrational frequencies and Nt and Ng, the number of mol-
ecules of the trans and gauche forms respectively. Then
−1
⎛ − hv t
⎞
It N t ⎜ 1 − e κT ⎟ (4.60)
⎝ ⎠
−1
⎛ − hvg
⎞
Ig N g ⎜1 − e κ T ⎟ (4.61)
⎝ ⎠
From Eq. (4.60) and (4.61), we get
N ⎛1− e g ⎞
− hv / κ T
It
= t ⎜ (4.62)
I g N g ⎝ 1 − e − hvt / κ T ⎟⎠
It N 0 e − Et / T
⎛ 1 − e − hvg / κ T ⎞
= ⎜ 1 − e − hvt / κ T ⎟
I g N 0 e − Eg / T
⎝ ⎠
Taking logarithm of both sides, we obtain
− hv g
⎛ I ⎞ ΔE 1 − e κT
ln ⎜ t ⎟ = + ln (4.63)
⎝ I g ⎠ κT
− hv t
1 − e κT
where ΔE = Eg − Et, Ng = No e − Eg / T Nt = N0 e − Et / T and N0 = Nt + Ng

Differentiating Eq. (4.63) with respect to (1/T), we get
∂ ⎡ ⎛ I t ⎞ ⎤ ΔE E
⎢ln ⎥= + f (T ) (4.64)
∂ ( / T ) ⎢⎣ ⎜⎝ I g ⎟⎠ ⎥⎦ κ
− hvg / κ T ( hv g / κ ) ( hvt / κ )
where f (T ) = e − hvg / κ T
− e − hvt / κ T
( e ) (1 − e − hvt / κ T )
The plot of ln (It /Ig ) versus 1/T will be a straight line with intercept = ΔE/κ. From the intercept we can calculate ΔE.
Thus, the energy difference between the rotational isomers can be determined simply by monitoring the intensi-
ties of the Raman bands corresponding to the two rotamers at different intervals of temperatures. It /Ig depends
upon the choice of the baseline and also on the method of separation of two bands due to two rotamers overlap-
ping each other. Consequently, the selection of vibrational frequencies corresponding to the two rotamers is vital
for these types of studies. In addition to these two factors, ΔE also depends upon f (T), e.g. the energy difference
between the rotational isomers of 1,2-dibromoethane in the liquid state using 4880 Å line of Ar+ laser, as deter-
mined by monitoring the relative intensity of Raman bands at 660 (trans form) and to that at 551 cm−1 (gauche
form) in the temperature range 24.5–105.1°C is of the order of 955 ± 80 cal mole−1 ignoring f (T) factor. When
f (T) is taken into account, ΔE decreases by 38 cal mole−1 at 24.5°C and 54 cal mole−1 at 105.1°C. Consequently,
ΔE = 9 10 ± 80 cal mole−1. The value of ΔE determined by other methods lies in the range 650-1000 cal mole−1.
4.12.9 Structure of Water

Water is the principal constituent of all living organisms and the most abundant compound on the surface of our
planet. The human body is about 65 per cent water by mass, and some tissues such as the brain, lungs, and muscles
contain close to 80 per cent water. The unique properties of water make it essential to the existence and continua-
tion of life, and it is likely that water has played a major role in the formation of prebiological living systems. The
thermodynamic properties of liquid water, its volume, density, heat capacity, compressibility, and so on, its static
dielectric constant, x-ray diffraction pattern, refractive index and NMR chemical shift, can provide information
only about structures with relatively long lifetimes. On the other hand, infrared and Raman spectroscopy, and
inelastic neutron scattering tools where the radiation or particles interact with the liquid for very short times and
hence provide information concerning molecular processes in the liquid with very short life spans (Fig. 4.17). The
dielectric relaxation and ultrasonic absorption bridge the gap between groups of methods. The average lifetime
of a given arrangement of molecules in a liquid must exceed by a factor often the vibrational period of the bonds
which define ‘the structure, in order to be considered structurally significant. The period of an O−H stretching
vibration is approximately 10−14 second, while that of an intermolecular H bond [v (AH... B) is 200 cm−1 ] is about
2 × 10−13 second. Thus the lifetimes of the structural entities considered must be kept in mind in structural descrip-
tions of liquid water.
The models of water structure can be divided roughly into two types: the uniformistic or continuum models,
which assume the existence of only one type of molecule corresponding to the average of all the molecules need
be considered; and mixture models, which require the existence of two or more different structural units. Evidence
for a model of water structure is generally indirect. One or more of the physical properties of water are calculated,
based on the assumptions of the given model, and compared to the experimentally known data. The best model
+4
thermodynamic x-ray
properties diffraction
+2
0 NMR
chemical shift
−2
Log time (sec) −4
−6 dielectric
relaxation ultrasonic
−8 absorption
−10
inelastic neutron
−12
scattering
infrared and
−14 Raman spectroscopy
period of an O − H
stretching vibration
Fig. 4.17 Schematics diagram illustrating the time intervals for which various experimental methods provide
information on molecular processes in ice and water.
would be the simplest model that is in agreement with known facts and that predicts the properties of liquid water
with the use of the minimum number of adjustable parameters.
The infrared spectra of dilute solutions intervals for which various experimental methods provides information
on molecular processes in ice and water of D2O in H2O at −7.6 and 4°C has shown that while the ν01 (O−D) band
of ice is sharp (Δv1/2 ∼ 31 cm−1) and Lorentzian in shape, on melting the band broadens (Δν1/2 ∼ 150 cm−1) and
becomes Gaussian in shape. Thus, the simplest band shape occurs in ice though there is maximum H−bonding
and maximum physical coupling, while in water, the band broadness remains symmetrical even though there are
fewer H-bonding interactions. Moreover, Gaussian distribution has been found to fit the broad v01 (O—D) band of
water as well as 10 to 15 narrow, Lorentzian components. This suggests that O—D band at 4°C is due to continuous
distribution of the perturbed O—D vibrators.
The Raman spectroscopic studies of solutions of D2O in H2O in contrast to infrared spectroscopic studies, pro-
vides strong support for a mixture model of liquid water. Photo-electric Raman spectra of D2O in H2O obtained
with an argon ion laser source, exhibit a broadband at 2525 cm−1with a pronounced shoulder at 2645 cm−1. The
former peak has been attributed to H-bonded, and the latter peak to non-H bonded, OD vibrators respectively. The
enthalpy for the breakage of one H-bond obtained from the temperature variation of the ratio of the intensities of
the 2525 and 2645 cm−1 bands and from the temperature dependence of the intermolecular Raman vibrations (H
bond bending and stretching vibrations near 60 cm−1 and 160 cm−1, respectively.) is 2.55 kcal/(mole of O−H….O).
The corresponding entropy change is −8.5 cal/deg (mole of H bond). Thus, it is likely that a model able to cor-
relate and account for all the known properties of water will include some of the salient characteristics of both the
mixture and continuum type models.
4.13 BIOLOGICAL APPLICATIONS

Raman spectroscopic data on synthetic polypeptides, proteins and lipids not only supplemented the results of
infrared spectral studies of biological systems but also came up with a number of new correlations for clarifying
the conformational states of proteins and lipids in the biological membranes. This is important since structure
of membrane proteins depends upon their interactions with membrane lipids. The assignment of some of typical
Raman bands of poly-lysine and lipids are recorded in Tables 4.4 and 4.5 respectively. Some of the important corre-
lations are (i) the low-frequency bands, i.e. <50 cm−1 in certain proteins depend upon the protein configuration but
not only on the state of the sample, i.e. film, solution or crystals. These low-frequency vibrations can be used for
the detection of transitional motion of proteins in membrane since they reflect motions of all or very large portions
of molecules in a system. (ii) Denaturation of insulin alters the amide I and III regions as well as scattering from
S-S and C-S stretching and skeletal bending. The data suggest major conformational changes towards b-structure
upon denaturation of insulin. (iii) Raman data on long chain fatty acids, fatty acid esters and phosphatides reveal
that bands between 1000 and 1200 cm−1 arise from skeletal vibrations where alternate carbons move in opposite
directions in a hydrocarbon chain. The symmetric C—H stretching band from CH3 group (2870 cm−1) is intense
as compared to that of CH2 group (2850 cm−1) in the crystalline state and vice versa in liquid state
Table 4.4 Raman and Infrared frequencies (1500-1700 cm-1) of poly-lysine in different conformations.
Helix Antiparallel-b Unordered
Raman IR Raman IR Raman IR

1517w 1511 w 1521 m
1516w ⎫
⎬ II
1537m 1535s ⎭ 1535w 1530s II 1535 II
1547
1564 m 1565
1600m
1617m 1617m
1631 vs s ⎤
1639 s 1653 vs
⎥ I 1656 I
1647 vs 1650 vs⎫ I ⎥ 1665 vs
⎬ ⎥
1652m ⎭ 1672 vs 1683 vs
1685 m ⎥⎦
s−strong; vs−very strong; m−medium; w−weak; I−amide I band; and II−amide II band
Table 4.5 Raman lines (cm−1) of fatty acids and their esters.
Frequencies Assignment
720 Head group deformation vibration

~845 CH2 rock
967 =C—H (out of plane) deformation
~1085 C—C stretching
1261 ==C—H (in-plane) deformation
1298 CH2 twisting
1370 CH2 wagging
1445 CH2 deformation
1731 C=O (ester) stretching
~2100 C—D stretching
~ 2850 CH2 symmetric stretching
~2870 CH3 symmetric stretching
~2930 CH2 asymmetric stretching
~2960 CH3 asymmetric stretching
The weak band near 1080 cm−1 relative to the 1125 cm−1 band is indicative of trans structure in the crystal-
line state. This gets reversed upon melting due to the appearance of gauche isomer. Accordingly, the intensity
of 1080/1125 cm−1 bands may provide information about the structural changes in these types of systems in
moving from crystalline to molten state or vice versa, e.g the intensity ratio of bands near 1066 and 1130 cm−1 in
the spectra of aqueous suspensions of micellar dipamitoyl lecithin corresponds to the all trans structures in the
paraffin chain. The broad band at 1089 cm−1 appearing in the melted paraffin is assigned to the gauche rotations
of the melt. Thus, from the ratios of the scattering intensity of bands at 1089/1130 or 1089/1066 cm−1, we can
infer about the extent of ordering of paraffin chains in passing from micellar to molten state. Based on this cri-
terion, the Raman lines near 1100 cm−1 and the intensity ratio of 1125/1085 cm−1 bands in the spectra of lecithin
and lecithin-cholesterol mixtures, the hydrocarbon chains of dipalmitoyl phosphatidyl ethanolamine are mainly
in trans configuration in the crystalline state while those of egg phosphatidyl choline are in the liquid state. The
insertion of cholesterol into egg lecithin liposomers results in the decrease of contribution of gauche isomers
while increase in that of the trans isomers. Cholesterol prohibits the formation of some but not all gauche isomers,
thereby reducing the chain fluidity.
The Raman resonance spectrum due to the chromophore group of retinal cell is not affected by the primary
photoreaction since the time scale of Raman resonance state of this system is smaller, i.e. 10−14 second than that
of photoreaction which is of the order of 10−12 second. This time lag makes the technique of Raman resonance
spectroscopy an ideal tool for investigating the transients formed in this biological process, e.g. the conjugated
rhodospin protein of the retinal cell transports protons across the membrane on illumination of 11−cis retinal. The
prosthetic group of this protein not only forms covalent bond with protein but also the carboxyl group reacts with
amino group to form an imine. The rhodospin simply absorbs light, which then transforms the less stable 11−cis
retinal into the more stable 11− trans retinal (Fig 4.18).
As a consequence, the conformational changes occur in the entire rhodospin molecule in order to accommo-
date the new guest, i.e. 11− trans- retinal in it. The energy from biological reactions changes the trans form back
to the cis form. These conformational changes have been supplemented by Raman resonance scattering. The C=C
stretching mode of the retinal moiety ≈1555–1529 cm−1 not only differs in wave number but also in band shape
and width in the course of transformation of rhodospin to other transients in the photochemical processes. This
Light
11 11
CHO
12
hv from
12
rhodospin
11-cls-retinal CHO 11-trans-retinal

Fig. 4.18 Structural changes in 11-cis retinal on shining radiation.
property of C = C stretching vibration has been exploited for kinetic studies of such transformations in biological
processes. The C = C stretching mode of intact retinal is lower than that of free retinal or its basic Schiff base. This
lends support to the hypothesis that chrome is bound as a protonated Schiff base, i.e. by a —C = N+H-bond. It is
also observed that the purple colour of chrome of purple membrane is because of unprotonated Schiff base. The
chromophore consists of a charge transfer complex formed between retinyllysine and a protein tryptophan.
This technique is used to probe the active site in a large number of metalloproteins including the haeme pro-
teins, non-haeme oxygen carriers, transferrin, chlorophylls, vitamin B-12, iron-sulphurproteins and blue copper
proteins and to investigate the reaction dynamics of chlorophyll, haeme group as well as of various helical forms
of DNA and the transitions among them.
Resonance Raman studies of the Fe—O stretching region for haemoglobin provide structural information on
the mode of dioxygen bonding. Two possible structures are the side on model I, and the end-on-model II.
ESR studies of oxycoboglobin and x-ray structure of a synthetic iron porphyrin provide evidence for structure
(II). The Fe−O stretch (568 cm−1) in the latter, i.e. dioxygen iron porphyrin, as determined by RRS is close to that
of oxyhaemoglobin (567 cm−1). This Raman data in conjunction with the x-ray structure of the dioxygenated syn-
thetic iron porphyrin, suggest that dioxygen is end-on in haemoglobin. RRS of pure 16O18O haemoglobin exhibits
two peaks separated by 20 cm−1 in the 550 cm−1 region. This observation corresponds to model (II) where two
stretching frequencies separated by 20 cm−1 are expected for the Fe−18O−16O and Fe−16O−18O structures. The struc-
ture (I) is ruled out since only one iron–oxygen stretch is possible in the 550 cm−1 region because an end-for-end
flip of 16O18O in (I) does not produce a second isotopic isomer.
O⎯O stretching frequencies in oxygen carriers may also be fixed criteria for the O
assignment of oxidation states. The range of O−O stretching frequencies in simple O O O
compounds and metal complexes is 1075–1164 cm−1 for O1− 2 and 738–879 cm
−1
2−
for O 2 . The value of v (O—O) in oxyhaemeerythrin and haemocyanin has been Fe Fe
observed at 844 and 742 cm−1, respectively. These frequencies fall into the O 2−2 per-
oxide class suggesting thereby that these proteins have a peroxide type-O2. (I) (II)
PROBLEMS
1. A compound is irradiated by 4358 Å line of mercury. ⎛ h ⎞ the Stokes and anti-
Raman lines are observed at wavelengths of 4420, 4435, 11. Compute in units of
4498 and 4620 Å. (a) Compute the value of the Raman
⎝ 8π 2 Ic ⎠
Stokes Raman frequencies which will result from rota-
shift for each line in terms of wave numbers. (b) What
would be the wavelengths of the anti-Stokes lines corre- tional transitions into a final state of quantum number (a)
sponding to the Stokes lines at 4620 Å? (c) At what wave J = 4, and (b) J=6. [Ans: (a) Stokes: 22; anti-Stokes: 14]
numbers will their Raman lines appear if the compound 12. The fluorine molecules were excited by 4047 Å and 4358
is irradiated with monochromatic light of 5460 Å? Å Hg lines. The Stokes lines are observed at 4199 Å and
(a) Hint: n0 − nR = ΔnR; nR < n0 for Stokes lines; nR > n0 for 4534 Å respectively. Compute the ground state vibra-
tional frequency of F2 and its force constant also.
anti-Stokes lines; n0 = 22946 cm−1.
[Ans: 893 cm−1]
[Ans: 322,398,714,1301cm−1, 4124 Å]
13. The Raman spectrum from liquid chloroform excited by
(b) Hint: ΔnR will remain the same; vR = 4620 Å = 21645cm−1,
the 4358.3 Å line of mercury consists of lines at 4289.9,
[Ans: 4124 Å]
4309.1, 4358.3, 4408.7, 4429.0, 4489.1, 4507.9, 4602.2,
(c) [Ans: 17993, 17917, 17601, 17014 cm−1]
5018.7 Å. What are the six fundamental vibrational modes
2. The displacements from the Rayleigh line in the rota- of chloroform?
tional Raman spectrum of HCl are expressed by ΔnR = [Ans: 262, 366, 668, 761, 1216 and 3019 cm−1]
± (62.4 + 41.6 J) cm−1. Compute the moment of inertia 14. The infrared and Raman spectra of sulphur dioxide are
and bond length of HCl molecule. given below:
•Hint: Compare the given expression with ΔnR =2B
(2J + 3).
[Ans: I = 1.626 amuÅ2, r = 1.293Å] Frequency (cm−1) IR Raman
3. The Raman spectrum of PCl3 is excited by the mercury
519 s, || type band p
line 4358 Å. Spectrum shows Raman lines for ΔnR = 184,
256, 482 and 514 cm−1. At what wavelengths will their 1151 s, || type band p
Raman lines appear if PCl3 is irradiated with monochro-
matic light of 5461 Å? 1361 s, ⊥ type band dp
[Ans: 18128, 18056, 17830, 17798 cm−1 ]
4. A substance shows a Raman line at 4567 Å when an excit-
ing line of 4358 Å is used. At what positions Stokes and Determine the structure of sulphur dioxide. Assign the
anti-Stokes lines for the same substance will be observed bands on the basis of polarization and depolarization
when the position of Rayleigh line is at 4047 Å? data.
[Ans: 4227 Å (Stokes), 3882 Å (anti-Stokes)] • Hint: All the three fundamental modes of vibra-
5. The values of the we and xe are 2358.03 cm−1 and 6.00 × 10−3 tion are IR and Raman active. So molecule has no
respectively for the ground state of N2. Compute the zero- centre of symmetry and hence has bent structure.
point energy and the expected Raman displacement. The absence of complicated rotational structure and
simple PR contour also indicates that the molecule is
• Hint: For ground state ν = 0 and the selection rule is
nonlinear.
Δn = ± 1; ΔnR = (EV = 1.− EV = 0)/hc
[Ans: SO2, nonlinear]
[Ans: zero point energy = 8282.07 × 10−16 erg;
15. The infrared and Raman spectra of nitrous oxide are
ΔnR = 2329.74 cm−1] given in the tabular form as under.
6. The Rayleigh line in the Raman experiment is 4358 Å and
the spectrum shows the Stokes line at 6433 Å. Find the
wavelength of the anti-Stokes line. [Ans: 3295 Å] Frequency (cm−1) IR Raman
7. The bond lengths of N2, and O2 are 1.09 and 12 1Å respec-
tively. Their force constants are 22.6 and 11.4 millidynes/Å 589 s, PQR (⊥) contour _
respectively. Draw schematic diagrams for the Raman spec-
1285 vs, PR (||) contour vs, p
tra of N2 and O2 gases excited by 4358 Å line of mercury.
8. The centre of vibrational band of O2 lies at 1556 cm−1 2224 vs, PR (||) contour s, dp
while that of HCl vapour at 2886 cm−1. Calculate the rela-
tive intensities of Stokes and anti-Stokes lines for both
the cases at room temperature. Determine the structure of the compound.
9. When acetylene is irradiated with 4358 Å line of mer- • Hint: The presence of PR (||) and PQR (⊥) structures
cury, a Raman line attributed to symmetrical stretching reveal that the molecule is linear. Since all the funda-
vibration is observed at 4768 Å. Compute the fundamen- mental bands are IR active, the molecule does not pos-
tal frequency for the said vibration. [Ans: 1973 cm−1] sess centre of symmetry. Consequently the structure of
10. The Raman spectrum of CO2 is recorded using a 4358 the molecule is N—N—O but not N—O—N. Two bend-
Å Hg line. The fundamental frequency for CO2 symmet- ing (⊥) modes will be degenerate, which are weak and
ric stretch deduced from the Raman spectrum has been unobserable in Raman spectra. 1285 being vs and p
found to be 1330 cm−1. Determine the position of Raman is NO symmetric stretching band while 2224 being dp
line corresponding to this fundamental mode of vibration. and s, is N — O asymmetric stretching band.
[Ans: 4119 Å (anti-Stokes)] [Ans: N—N—O]
CHAPTER 5 ELECTRONIC
SPECTROSCOPY
Change and development is a never–ending process and so is man’s cognition of truth through practice.
⎯Unknown
The beauty of a law of nature lies in the dimensions of physical quantities involved.
⎯H S Randhawa
5.1 INTRODUCTION
The electronic states of molecules originate from the levels of the atoms that constitute the molecule as a result
of Stark splitting in the electric field of the molecule. Electronic spectroscopy deals with the study of absorption
of visible (400–800 nm), the near ultraviolet (200−400 nm), the far, or, vacuum ultraviolet (200 nm or lower) and
sometimes even in the far infrared (4000−12000 cm−1) region of radiation by inorganic and organic molecules,
namely CH3NO2, C6H6, C2H4, CO, H2, AgCl, HgCl, HCl, N2, I2; molecular ions, namely H 2+ , N 2+ , CH + ; stable
and unstable (lifetime of the order of a few microseconds or less) molecular radicals,viz. OH, NH, CN, C2, etc., in
the liquid/solution, solid and vapour phases. The absorption of radiation causes the electrons to cross over from a
lower-energy electronic state, say Ψelec ′ , to a higher-energy electronic state, also called the excited electronic state,
say Ψelec
′′ . AS a result of this transition, the electronic configuration of the molecule changes and hence the excited
state becomes polarised. Accordingly, the electronic transition is accompanied by a change of dipole moment,
also known as instantaneous dipole moment, and the corresponding signal will be intense.
Further, since each electronic state has vibrational and rotational energy states associated with it (Fig. 5.1),
the electronic spectra of heteronuclear and homonuclear diatomics in the vapour phase will exhibit rotational
fine and vibrational gross structures; but the rotational fine structure is not normally observed in the ultraviolet
visible absorption spectrum because of small energy difference between the adjacent rotational energy states as
compared to that of the vibrational energy states. The rotational fine structure is useful to gather information about
the electronic states of molecules, rotational constants and hence the moments of inertia and bond lengths for
both the lower and excited electronic states of the molecules. Further, since the rotational constants of isotopes
are different, they can be detected and their per cent abundances can be measured by comparing the intensity of
the respective bands. On the other hand, the vibrational parameters, viz. dissociation energy, the frequency of
the fundamental vibrational mode, and hence the force constant, can be deduced for both the ground and excited
electronic states from the vibrational coarse structure. Electronic spectroscopy is also used to study the kinetics
of production and destruction of free radicals from the variation of band intensity with short period of time as in
the technique of flash photolysis.
The technique of electronic spectroscopy applies equally to polyatomic molecules as well. In polyatomic mol-
ecules, the rotational energy levels are more closely spaced because of larger moments of inertia. Consequently,
the rotational fine structure, in general, cannot be observed and the vibrational structures are usually present only
as broad bands. In spite of this, electronic spectra of molecules are useful to gather information about the elec-
tronic structure of a large number of polyatomics.
Further, it is to be noted that the electron spectra of molecules are more complex as compared to rotational and
vibrational spectra because a large number of transitions are simultaneously involved in it. This, and translational
motion of molecules in the vapour and dilute solution, tends to blur out the absorption into a broad band. In pure
liquids, molecular rotation is hindered because of molecular collisions and hence the rotational fine structure is
blurred. The blurring out is enhanced in solution by the solvent interactions. The blurring of the rotational fine
structure often masks the vibrational coarse structure in the spectrum. Due to this very reason, the rotational fine
and vibrational coarse structures are generally missing in the low-resolution electronic spectra of molecules in
the solution phase. Consequently, the electronic spectrum in the solution phase is usually broad and characterless,
but still the intensity and frequency of the broad band in the electronic spectrum is a characteristic of a particular
Electronic Spectroscopy 257
v″ = 4
v″ = 3
v″ = 2
v″ = 1
3
2
1
J″ = 0
v″ = 0 ψ′′elec(S1)
v′ = 4
v′ = 3
v′ = 2
v′ = 1
2 3
J′ = 0 1
v′ = u ψ′elec(S0)
Fig. 5.1 Schematic diagram showing the vibrational and rotational energy levels of two electronic states, Ψelec
′ and Ψelec
′′ , of a diatomic
molecule.
functional group present in the molecule. Thus, the tool of electronic spectroscopy in the solution phase is
extensively used for routine quantitative analysis of mixtures, identification of large organic molecules, structural
diagnosis, determination of stability constants of molecular complexes, kinetics of formation of molecular
complexes, etc.
The electronic spectra are not only studied in the absorption mode but can also be studied in the emission
mode as well. The molecule with an electron in the upper energy electronic state loses the absorbed energy and
returns back directly to the lower state. This results in the electronic absorption band spectrum. The molecule in
the excited state may lose excess energy in order to return to the ground energy state via radiative and nonradiative
intramolecular processes also. In the former, excess energy is lost by emission of radiation which results in elec-
tronic emission spectra, whereas in the latter the excess energy is lost without emission of radiation and is of little
interest in the present context. The radiation emitted during a radiative transition is known as fluorescence (decay
time ∼10−10s – 10−4s) when the transition is between states of same multiplicity, i.e. singlet (S1) → (S0) transition,
and phosphorescence (decay time seconds or even hours) when transition is between states of different multiplic-
ity, i.e. triplet (T1) → singlet (S0) transition. For most solutions and liquids at room temperature, the nonradiative
processes are so rapid that fluorescence cannot be observed. However in some solutions, it can be seen even with
the naked eye, e.g. in fluorescein. Since the nonradiative processes are temperature dependent, the fluorescence
can be observed by cooling the solutions to supercooled liquids or glasses. Aromatic impurities in a crystal result
in strong fluorescence. The phenomena of absorption and emission has been sketched in Fig. 5.2(a). The 0–0 band
of fluorescence and absorption do not always occur at the same frequency because of solvent effects. Absorp-
tion takes place with the arrangement of solvent molecules suitable for the ground state and fluorescence with
different arrangement of molecules appropriate for the excited state. Since absorption is at higher frequencies,
ultraviolet lamps are required to excite the visible fluorescence.
When the molecule is in the gas phase, the emission may occur from an excited vibrational state. This is called
resonance fluorescence. Resonance fluorescence also occurs if the molecule is in a solid state and is unable to
rotate. The emission in such a case may be partially polarised and is distributed non-uniformly at different angles
with respect to the direction and polarisation of the incident light. There is thus a close connection with Raman
4 Inter system crossing

3
2
1
V ′′ = 0 T1
S=1
V ′′ = 0
Non-equilibrium
Excited state
S1
4 S1 Excited state in
3 Equilibrium with
2 Solvent cage
1
v′= 0 5−0
0 0 0 0
Fluorescence Phosphorescence Absorption Fluorescence
Absorption or emission intensity
Absorption S0 Non-equilibrium
S0
Ground state
Ground state in
Equilibrium with
Solvent cage
(b)
Wave number (cm−1) (a)
Fig. 5.2 Schematic energy-level diagram showing the (a) phenomena of absorption, fluorescence and phosphorescence, and (b) 0–0
bands of absorption and emission. The wavy arrows indicate radiationless vibrational relaxation. The diagrams for fluorescence and phos-
phorescence phenomena are called Jablonski diagrams.
spectra and this connection has been discussed in detail in Chapter 4, on Raman spectroscopy under the title
resonance Raman effect/scattering.
The phenomena of phosphorescence and fluorescence will be described in detail later on under the title
‘fluorescence and phosphorescence’.
We have discussed here the subject of electronic spectroscopy in two parts, viz. high-resolution and low-resolution
electronic spectroscopy of diatomic and polyatomics in the gas phase. The former deals with the high resolution
(in gas phase) while the latter deals with the low-resolution electronic spectra of diatomics and polyatomics in the
solution phase.
5.1.1 Multiplicity of Electronic States

It is defined by the expression 2S + 1 where S is the algebraic sum of the spin quantum numbers of the electrons,
1 1
i.e. + a d − in the system. A molecule with two unpaired electrons (↑↑ or ↓↓) is said to be in (2 × 1 + 1 = 3) a
2 2
triplet state and is designated by the symbol T. Ground state of a triplet state is designated as T0. A molecule with
two paired electrons (↑↓ or ↓↑) is said to be in (2 × 0 + 1 = 0) a singlet state and is designated by the symbol S.
Ground state is a singlet state and is denoted by S0. The higher-energy singlet and triplet states are designated as
S1, S2, S3 … and T1, T2, T3 ... respectively. Singlet states have invariably higher energy than the corresponding triplet
states because of greater electron repulsion in the singlet state. In absorption molecular electronic spectroscopy,
transition always occurs between electronic states of the same multiplicity, i.e. S0 → S1, S0 → S2, are allowed but
S0 → T0 are forbidden transitions.
5.1.2 Electronic Transition Energy

′ be the energy of the ground electronic state, say ψ elec
Let E elec ′ , of the molecule before absorption of radiation;
hc
′′ be that of the excited state, say ψ elec
and E elec ′′ , of the molecule after absorption of radiation. The energy E = of
radiation required to bring about the transition is given by λ
hc
E E elec ′ =ΔE=
′′ − E elec (5.1)
λ
Here, the quantity ΔE is the energy difference between the electronic states y ″elec and y ′elec and equals E. Thus,
the energy required to produce one excited-state molecule is given by
hc 6.626 × 10 −34 (J s) × 2.998 × 108 ( m/s) 1.987 × 10 −16

Δ = = = J/molecule (5.2)
λ λ × 10 −9 ( ) λ
Here l is in nm.
The energy required to excite 6.023 × 1023 molecules is known as an einstein and the energy of an einstein of
radiation of wavelength l can be computed from the expression
1.196 × 108
Δ = J mole −1 (5.3)
λ (in units of nm )
Problem 5.1: The band in the electronic spectrum of a molecule appears at 250 nm. Calculate (a) the energy
required to produce one excited state molecule, and (b) the energy of an einstein of radiation.
Solution (a) In order to determine the energy required to produce one excited-state molecule we make use of
the expression.
1.987 × 10 −16
Δ = J/molecule
λ ( nm )
Substituting the given value of l, we get
1.987 × 10 −16
Δ = =7.958 × 10 −19 J/molecule
250
(b) The energy of an einstein of radiation (l = 250 nm) is given by
1.196 × 108
Δ = J mole −1
λ (in units of nm)
1.196 × 108
= J mole −1
250
= 4.784 × 105 J mole−1
5.2 HIGH–RESOLUTION ULTRAVIOLET/VISIBLE SPECTROSCOPY

AND SELECTION RULES FOR ELECTRONIC TRANSITIONS
IN LIGHT DIATOMICS
The high-resolution UV visible spectroscopy deals with the electronic spectra mainly of homo- and hetero-
diatomics in the gas phase.
The selection rules for electronic transitions have been worked out quantum mechanically from the solution of
the transition moment integral,
M t ( err ) ∫Ψ i
*
Mˆ tΨ j d τ (Eq. 2.37)
The square of the integral is a measure of transition probability which is proportional to integrated intensity of
the absorption band, i.e.
+∞ 2
Ia ∫ Ψi Mˆ tΨ j d τ
*
−∞
Here, M̂ t (= er12) is the integral dipole moment operator and is related to the difference in the electronic dipole
moments of the ground and excited states. The difference in the electronic dipole moments results because of
a different electron distribution in the two states, i.e. due to charge migration during transition. Thus when the
transition moment integral becomes zero, the transition is forbidden. yi and yj are the wave functions of the
energy states ‘i’ and ‘j’ respectively and are orthogonal to each other. dτ is the volume element. The components
of Mt along the three axes are
M M Ψd (Eq. 2.38)
M M Ψ dτ (Eq. 2.39)
M M d (Eq. 2.40)
Further, note that the dipole moment operator operates only on the space part of the wave function. Accord-
ingly, Eq. (2.37) is written as
M τ pin (2.37*)
Since the spin integral involves the dot product of the spin functions, the integral will be zero until the wave
functions for the spin belong to the same state, i.e. ΔS = 0. The components of Mt expressed by Eq. (2.37*) along
the three axes can also be written in the same fashion as that for Mt expressed by Eq. (2.37). Thus, for an allowed
transition, one of these integrals must be nonzero. This is possible only when the transition takes place between
states of same symmetry, i.e. reversal of spin is not allowed. The transitions are observed even for ΔS ≠ 0. For
example, the strongest transition in the Hg atom is observed at 253 nm. This is due to the fact that in heavy atoms
and molecules, the interaction between spin and space (i.e. spin-orbital interaction) is infinite and hence the spin
and space functions cannot be considered independent of each other. The mixed orbital wave function of the trip-
let state is given by
where lTS is the mixing coefficient; lTS is inversely proportional to the energy difference between the triplet
and the singlet states. The smaller the value of (ES − ET), greater is the mixing and stronger the singlet triplet
transition.
(i) Let us now prove that the transition between singlet and triplet states is forbidden. Let the wave functions
for the spin part of the singlet and triplet states be
yS = [a(1) b(2) − a(2) b(1)]
and yT = a(1) a(2)
= b(1) b(2)
or = [a(l) b(2) + a(2) b(l)]

For simplicity, the normalisation constants are not considered. For one of the triplet wave functions, we
write
′ = − ⋅ + (2 (1) spin
+ 2
−∫ −∫ (2)d ∫ d 1
Since the wave functions are normalised, the first and fourth integrals will be unity, whereas the second
and third integrals will be zero.
Therefore, ∫ Ψ Ψ′ d spin = + − 0 − 1 = 0. Thus, this transition is forbidden. It can also be shown sim-
ilarly that the integrals ∫ β ( )d spin and ∫ ( )d spin
are zero, indicating thereby that singlet-to-triplet transition is not allowed.

(ii) Molecules having a centre of symmetry exhibit g → u and u → g transitions but not g ↔ g and u ↔ u. In
general, transitions which occur without change in the ‘l’ value of the orbital are forbidden, i.e. Δl/ΔL
= ±1. This is called parity rule.
Parity = (−1)l, where l = orbital quantum number.
For parity = +1, the function has even parity.
For parity = −1, the function has odd parity.
Therefore, a state with even ‘l’ has even parity and odd ‘l’ has odd parity. Thus, for an allowed
transition, the parity of a system should change. The parity of a closed system cannot change. For
C, the electronic configuration is 1s22s22p2 ; Hence L = Sli = 2. ∴ Parity =(−1)2 = 1, even parity. For
N(1s22s22p3),
L = Sli = 3
∴parity = (−1)3= −1, odd parity.

For O(ls2 2s2 2p4), L = Sli = 4, Parity = (−1)4, even parity.
(iii) Simultaneous excitation of more than one electron is not permissible.
The selection rules for various electronic transitions in light diatomics have been worked out in Chapter 2
on microwave spectroscopy. Some of the allowed electronic transitions for homonuclear- and heteronuclear-
diatomics are listed here for ready reference.
Homonucler Heteronuclear
∑ ← →∑ u
+
∑ ← →∑
+ +
g
∑ ← →∑ u
+ −
∑ ← →∑
− −
g
→ ∑ u ∏ ← →∑ g
+ +
← → ∑ ∏ ← →∑
+ −
∏g ←
←⎯ ∏ ←⎯
→ ∑ u ∏ ← →∑ g
− −
∏g ←
←⎯ ∏ ←⎯
← →∏
∏g ←
←⎯→ ∏u ∏ ←⎯
→Δ
∏g ←
←⎯→ u , ∏u ← → Δ g Δ← →Δ
Δ g ← → Δu Δ← → F
⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅ F ←→F
The symbols in the notations of an electronic state of a molecule are explained here by taking the example of
+
2
∑ (i.e. H
g
+
2 , N+ )
(i) ∑ is the electronic state for which electronic angular momentum quantum number Λ is zero.
1
(ii) 2 represents the multiplicity of the state, (i.e. 2S 1 = 2 for S ), the resultant electronic spin of the
1 2
molecule in this state is + .
2
(iii) The superscript plus sign on the right–hand side indicates that the electronic eigen function corresponding
to this state does not change sign on reflection at a plane through the internuclear axis.
(iv) The subscript ‘g’ on the right–hand side represents the parity of the eigen function of the state which for
the given state is even i.e. the wave function of a molecule does not change sign on reflection through the
centre of symmetry.
5.2.1 Electronic Spectra of Diatomic Molecules

The electronic configuration of a molecule changes during the course of an electronic transition. Thus, each elec-
tronic transition is accompanied by an instantaneous dipole moment. Accordingly, the total electronic spectra, i.e.
composite of electronic, vibrational and rotational, of homo- and hetero-diatomics will be active in the ultraviolet
and visible region of radiation. Note that a number of electronically excited states are possible and in general, all
electronic absorption transitions are permissible between states of same multiplicity, i.e. singlets, doublets, trip-
lets, etc., and are weakly or strongly forbidden if multiplicity is different. Here, for convenience, we shall be treat-
ing the transition between ground electronic state as our lower electronic state and the lowest excited electronic
state as the upper state of the molecule. The total wave function of the diatomic molecule may be written as
ψ ψ elecψ ψ rot (5.4)
and based on Born–Oppenheimer approximation, the total energy of a diatomic molecule is expressed as
E = E elec + E vib E rot (5.5)
The energy levels of a rotating diatomic molecule are represented as

EJ = B hc J(J + 1), J = 0,1,2,3,… (Eq. 2.33)
while that of a vibrating molecule is represented by
EV = h c we [V + 1/2] – h c we xe [V + 1/2]2 + ……, V = 0, 1, 2, 3 … (Eq. 3.33)
The frequencies of separate lines in the electronic spectrum as worked out from Eqs. (2.33) and (3.33) are given
by the formula
′′
( E elec E ′′vib E ′′rot ) − ( E elec
′ E ′vib + E r′ot )
v =
hc
( ′′ − ′′ ) ( ′′ − ′ ) ( ′′ − ′ )
= + +
hc hc hc
Δ ΔE vib ΔE
ΔE rot
= elec
+ +
hc hc hc
or v = v elec + v vib + v rot (5.6)
Here, single and double primes stand for the lower and upper states of the molecule.
The quantity v vib characterises the vibrational gross or coarse structure of the band, and v rot , the rotational fine
structure of the band in the electronic spectra of diatomics. The term v elec determines the change in the electronic
energy of the molecule when it crosses over from the lower-energy electronic state (ψ ′ ) to the upper-energy
′′ ) . The quantity v elec exceeds v rot and v vib by a factor of 106 and 103 cm–1 respectively, i.e.
electronic state ( Ψelec
v elec ≈103 v vib ≈ 106 v rot. This shows that (i) the electronic spectrum of molecule appears in the visible or ultravio-
let regions of the spectrum, i.e. electronic spectrum is displaced with respect to its rotation–vibration spectrum
towards short wavelengths, and (ii) the electronic spectra of molecules will exhibit a vibrational coarse- and
rotational fine-structures as well.
The terms v vib and v rot as derived from Eqs (2.33) and (3.33) are expressed as
⎡ ⎛ 1⎞ ⎛ 1⎞ ⎤ ⎡ ⎛
2
1⎞ ⎛ 1⎞ ⎤
2
v vib ⎢w e′′ ⎜⎝V ′′ w e′′ x e′′ V ′′ + ⎟ ⎥ − ⎢w e′ ⎜V ′ + ⎟ − w e′ x e′ ⎜V ′ + ⎟ ⎥ (5.7)

⎢⎣ 2⎠ ⎝ 2 ⎠ ⎥⎦ ⎣⎢ ⎝ 2⎠ ⎝ 2 ⎠ ⎥⎦
and
v rot Bv ′′ J ′′(J ′′ + ) − Bv J ′(J

(J + ) (5.8)
The two different frequencies, w e′ and w e′′ and two different rotational constants Bv′ and Bv″ correspond to two
different electronic states Y ′elec and Y ″elec respectively (contrary to infrared spectra where they belong to the same
′ ). Each of these stats are characterised by a definite frequency

electronic state, i.e. the ground electronic state Y elec
we and also unharmonicity constant xe and definite equilibrium bond length ‘re’, i.e. definite value of rotational
constant Be and, hence, moment of inertia Ie.
5.3 VIBRATIONAL COARSE STRUCTURE OF ELECTRONIC

SPECTRA OF DIATOMICS
Disregarding the rotational fine structure, let us discuss the vibrational coarse structure of the electronic spectra
of diatomic molecules. Under this condition, we write from Eqs (5.6) and (5.7)
v v elec − vib = v elec + v vib
⎡ ⎛ 1⎞ ⎛ 1⎞ ⎤ ⎡ ⎛
2
1⎞ ⎛ 1⎞ ⎤
2
= v elec + ⎢w e′′ ⎜V ′′ + ⎟ − w e′′ x e′′ ⎜V ′′ + ⎟ ⎥ − ⎢w e′ ⎜V ′ w e′ x e′ V ′ + ⎟ ⎥ (5.9)

⎢⎣ ⎝ 2⎠ ⎝ 2 ⎠ ⎥⎦ ⎢⎣ ⎝ 2⎠ ⎝ 2 ⎠ ⎥⎦
Here, the quantity v elec − vib defines the centres or origins of a series of vibrational bands building the electronic
spectrum.

Energy transition V′ → V″ associated with an electronic transition Ψ′elec → Ψ″elec is permissible.
5.3.2 Sequence and Progression

According to the selection rule (5.3.1), a large number of vibrational lines are expected in the electronic vibra-
tional spectra of the molecule due to which the analysis of the observed spectrum becomes cumbersome. This
problem is solved to a large extent by considering the ground electronic state as the lower energy electronic state
Ψelec
′ . Accordingly, the allowed transitions are
Ψelec
′ : V ′ (0, 1, 2, 3, …) → Ψelec
′′ : V ″ (0, 1, 2, 3,…)
0 → 0, 0→1, 0→2, 0→ 3,…;
1 → 0, 1→1, 1→2, 1→ 3,…;
2 → 0, 2 → 1, 2 → 2, 2 → 3,…;
3 → 0, 3 → 1, 3 → 2, 3 → 3,…; etc.
Such a set of transitions is termed as band and a series of bands having a constant value of (V″ – V′) is known
as sequence, e.g. the following sets of transitions
(V′− V″): V″− V′ = 0: 0 → 0 1 → 1, 2 → 2,…
= + 1: 0 → 1, 1 → 2, 2 → 3, …
= − 1: 1 → 0, 2 → 1, 3 → 2,…
= + 2: 0 → 2, 1→ 3, 2 → 4,…
= −2: 2 → 0, 3 → 1, 4 → 2…
constitute sequences. The transitions 1→ 0, 2 → 0, 3 → 0 do not form a sequence since V″ – V ′≠ constant. On the
other hand, a set of bands having a definite value of V′ but varying values of V″ and vice versa is called a progres-
sion. For example, the following set of transitions in bands,
i.e. V′ (0) = → V″ (0, 1, 2, 3…),
V′ (=1) → V″ (0, 1, 2, 3…),
V′ (=2) → V″ (0, 1, 2, 3…), etc,
6
5
4
ψ′′elec 3
2
1
v ′′ = 0
ψ′elec 2
1
v′ = 0
0 0 2 0 4 0 6 0
1 0 3 0 5 0
Wave number (cm-1)
Fig. 5.3 Schematic energy-level diagram showing the vibra-
tional coarse structure of the band in an electronic transition.
constitute V′ progressions while constitute V ′ progressions for a fixed value of V″ and varying values of V′. The
vibrational coarse structure of a band V′ (= 0) → V″ (= 1, 2, 3, 4, 5, 6, 7) is presented in Fig. 5.3. The coarse structure
reveals that the spacing between the adjacent lines in a band decreases and is crowded together at higher frequen-
cies. This is because of anharmonicity factor which increases as the vibrational quantum number increases.
5.3.3 Deslandres System of Bands of a Diatomic Molecule

An empirical formula corresponding to Eq. (5.9) derived by Deslandres is known as Deslandres formula and is
expressed as
w 0′′ V ′′ − w 0′′ x 0′′ V ′′ 2 ]

v = v 0 + [w [w 0′ V ′ w ′ 0 x ′0 V ′ 2 ] (5.10)
Compare Eqs (5.9) and (5.10) in order to obtain the relations between the constants involved in them, i.e.
⎡ w ′′ ⎛ x ′′ ⎞ w ′ ⎛ x ′ ⎞ ⎤
v v elec − vib = ⎢v elec + ⎜1 − ⎟⎠ − ⎜1 − ⎟⎠ ⎥
⎣ 2 ⎝ 2 2 ⎝ 2 ⎦
+ [ ′′(1 − x ′′ ) ′′ − ′′ ′′ ′′ 2 ] − [ ′ (1 − x ′ ) ′ − ′ ′ ′2 ] (Eq. 5.9)
Thus, the relations between the constants are

w ′′ ⎛ x ′′ ⎞ w ′ ⎛ x ′ ⎞
v 0 = v elec + ⎜1 − ⎟⎠ − ⎜1 − ⎟⎠
2 ⎝ 2 2 ⎝ 2
w 0′′ = w ′′(1 − x ′′ )
w 0′′ x 0′′ = w ′′x ′′
w 0′ = w ′(1 − x ′ )
w 0′ x 0′ = w ′x ′ (5.11)
Further, if each frequency of the band is designated as νV ′V ′′ then Eq. (5.10) can be rewritten as
vV V = v 0 + (w 0′′V ′′ − w 0′′ x 0′′ V ′′ 2 ) (w 0′ V ′ w 0′ x 0′ V ′ 2 ) (5.12)

Table 5.1 Deslandres system of bands of diatomics.

V ≤/V 0 1 2 3 4 5 …
0 v 00 v 10 v 20 v 30 v 40 v 50 …
1 v 01 v 11 v 21 v 31 v 41 v 51 …
2 v 02 v 12 v 22 v 32 v 42 v 52 …
3 v 03 v 13 v 23 v 33 v 43 v 53 …
4 v 04 v 14 v 24 v 34 v 44 v 54 …
5 v 05 v 15 v 25 v 35 v 45 v 55 …
From this expression, the entire system of bands expressed by the Deslandres formula are deduced and are
listed in Table 5.1.
The V″ progressions (correspond to V′ = constant, i.e. 0, 1...) as deduced from Eq. (5.12) can be expressed as
v = v 0V + w 0′′V ′′ − w 0′′ x 0′′ V ′′ 2 (5.13)
They are due to transitions from a definite vibrational state (V′ = 0, 1...) of the ground electronic state Ψelec ′ to
all possible vibrational levels ( V ′′ = 0, 1, 2, 3...) of the upper electronic state Ψelec
′′ (or vice versa). The V ′′ progres-
sions are a group of bands that converge to a certain limit at high frequency because of anharmonicity parameters
and are characteristic of the absorption spectra of molecules such as I2, Br2, N2O2, NO, O3, SO3, H2CO, C6H6, etc.
CO (in the Schumann region, < 2000 Å).
The V ′ progressions (correspond to V″ = constant, i.e. 0, 1, 2....) are expressed as
v v 0V ′ − w 0′ V ′ + w 0′ x 0′ V ′ 2 (5.14)
The V ′ progressions are due to the transitions from a definite vibrational state (V ′′ = 0, 1,...) of an upper elec-
tronic state Ψelec
′′ to all possible vibrational levels (V ′ = 0, 1, 2, 3, 4...) in the ground electronic state Ψelec ′ of the
molecule (or vice versa), i.e. 0 ↔ 0, 0 ↔ 1, 0 ↔ 2, 0 ↔3... and 1 ↔ 0, 1 ↔ 1, 1 ↔ 2, 1 ↔3.... These progressions
are characteristics of the emission spectrum particularly in the fluorescence spectrum, e.g. fluorescence spectra
of I2 (excited by green line of Hg 5461 Å), O2, S2, NO, SO2, etc. The progressions observed in these systems cor-
respond to V″ = 26 → V ′ = 1, 2, 3, 4, 5...
It is to be noted that emission spectra of a previously excited molecule by electric discharge, shock wave, temper-
ature and also in chemical (chemiluminescence) excitation will consist of a large number of progressions because
of all possible transitions from vibrational states ( V ′′ = 1, 2, 3, 4, 5...) in the upper electronic state Ψelec ′′ to the vibra-
tional states (V ′ = 0, 1, 2, 3...) in the ground electronic state Ψelec′ of Fig. 5.4. In such a situation, the various pro-
gressions will overlap and the result and pattern of the emission spectrum becomes very complicated. Many times,
the overlapping of different progressions results in segregation of separate groups of bands in the spectrum. These
correspond to the diagonal progressions of Deslandres Table, i.e. v 00 ′ , v 11′ , v 22
′ , ...v 10′ , v 21
′ , v 32
′ ,...v 20
′ ,vv 31
′ , v 42
′ , ...
etc. Such progressions are observed in the emission spectra of C2, N2, CO, CN, etc.
The characteristic constant of each system of bands is the band v 00 ′ , i.e the zero line of the band, which
corresponds to the transition ( V ′′ = 0 ↔ V ′ = 0) and is known as the zero band. The frequency of the zero line
of the zero band does not coincide with the frequency velec because of zero vibrational (or point) energy of the
molecule, i.e. v0 ≠ velec. In other words, we can say that the pure electronic transition is strictly forbidden. This
is often called the 0–0 band and is devoid of vibrational quanta, both in the ground and excited states. The band
corresponding to a transition from a vibrationally excited state of a ground electronic state (V ′, Ψelec ′ ) to the lower
vibrational state of the upper electronic state ( ′′,Ψelec ′′ ) is called the hot band.
4
3
2
1 Ψ′′elec
V ′′= 0
4
3
2 Ψ′elec
1
V ′= 0
ΔV = + 1 ΔV = 0 ΔV = −1 ΔV = −2
Wave number (cm−1)
Fig. 5.4 Progressions in the emission spectrum of a pre-excited diatomic molecule.
Energy Difference Between Two Electronic States The ground ( ′ ) and excited ( ′′ ) electronic states
of a molecule are usually represented, by the different Morse potential curves. The energy difference, v elec,
between two electronic states will thus correspond to the energy difference between the minima of two Morse
curves. The energy difference between the two minima as deduced from Eq. (5.9) is given by
⎛ w ′′ w ′′ x ′′ ⎞ ⎛ w ′ w ′ x′ ⎞
v elec −vib(V
vib(V = 0 V = 0) = v elec + ⎜ e − e e ⎟ − ⎜ e − e e ⎟ cm −1
⎝ 2 4 ⎠ ⎝ 2 4 ⎠
⎛ w ′′ w ′′ x ′′ ⎞ ⎛ w ′ w ′ x ′ ⎞
or v elec v elec
l − vib(
ib(V ′ = 0 V 0) − ⎜ e − e e ⎟ + ⎜ e − e e ⎟ cm −1
⎝ 2 4 ⎠ ⎝ 2 4 ⎠
or v elec v elec − vib(V ′ = 0→V ′′ 0) – zero point energy of the upper state + zero point energy of the ground state. (5.15)
[ 1 cm–1 ≡ 11.958 J mole–1]
Problem 5.2: The zero-point energy of the ground state of C2 is 817.8 cm–1 and that of its lowest excited state is
890 cm–1. The energy difference between the two minima of the electronic potential energy curves is 3.832 ×
10–19 J. Determine the energy of V' = 0 → V″ = 0 transition in cm–1. What is the corresponding wavelength?
Solution
ΔE
E 3.832 × 10 −19 (J)
v elec = = = 19295.8 cm −1
hc 6.62 × 10 −34 (J s) × 3 1010 (cm/s)
Substituting the values of zero-point energies of the ground and excited states together with the value of v elec into
Eq. (5.15), we get
v elec − vib(V ′ V ′′ ) = (19295.8 + 890 − 817.8)cm −1
= 19368.0 cm–1
The corresponding wavelength is 5163.1 Å.
Problem 5.3: The analysis of the vibrational structure of a band in the ultraviolet spectrum of CO leads to the
following data:
Ground state Excited state
(1∑ ) +
(1 )
w e′ ( −1
) x e′ ×103 w e″ ( 1
) x e″ ×103
2170.21 6.20 1515.61 11.38
The ν 00 transition is observed at 64746.55 cm–1. Calculate (a) the zero-point energies for the upper and lower
states, and (b) the energy difference between the two electronic states. What is the corresponding wavelength?
Solution (a) From Eq. (5.15), we write the expression for the zero-point energies of the upper and lower states as
⎛ w e′′ w e′′ x e′′ ⎞ −1

Zero-point energy of the upper state = ⎜ − cm
⎝ 2 4 ⎟⎠
⎛ w e′ w e′ x e′ ⎞ −1
Zero-point energy of the lower state = ⎜ − cm
⎝ 2 4 ⎟⎠
Substituting the values of the respective constants from the problem, we get the zero-point energy of the upper
state
⎛ 1515.61 1515.61 × 11.38 × 10 −3 ⎞ −1
= ⎜ − ⎟⎠ cm
⎝ 2 4
= (757.80 − 4.31) cm–1 = 753.49 cm–1

and zero-point energy of the ground state
⎛ 2170.21 2170.21 × 6 20 10 −3 ⎞ −1
= ⎜ − ⎟⎠ cm
⎝ 2 4
= (1085.1 − 3.36) cm–1 = 1081.74 cm–1

(b) Substituting these values of energies together with the value of v00 from the problem into Eq. (5.15), we get
the energy difference between the two electronic states:
v elec = ( .55 − 753.49 + 1081.74)cm −1
= 65074.80 cm −1 = 778.1 kJ mole −1 [1 cm–1 = 11.958 J mole–1]
The corresponding wavelength is 1536.69 Å.
Problem 5.4: The 0→0 transition in the electronic vibrational spectrum of C2 arising from the ground (π)
3
u
( )
and excited π g electronic states of C2 is observed at 19378 cm . The values of we and xe for ground and
3 –1
excited electronic states of C2 are 1641.4 cm−1, 7.11 × l0−3 and 1788.2 cm−1, 9.19 × l0−3 respectively. Calculate
(a) the energy difference between the two minima of the Morse curves corresponding to the electronic states of the
molecule, and (b) the force constants for the normal modes of vibration of C2 in the ground and excited electronic
states of the molecule.
Solution (a) In order to determine the energy difference between the two electronic states of C2, we are in need
of the quantities w e′ x e′ and w e′′ x e′′ , from the data reported in the problem, we get
−1
w e′ x e′ = 1641.4 ( ) × 7.11 × 10 −3 = 11.67 cm −1
−1
w e′′ x e′′ = 1788.2( ) × 9.19 × 10 −3 = 16.43 cm −1
Substituting the values of w e′ x e′ and w e′′ x e′′ and also the wave number of 0–0 band into Eq. (5.15), we obtain
⎡ ⎛ w ′′ w ′′ x ′′ ⎞ ⎛ w ′ w x ′ ⎞ ⎤
v elec ⎢v elec
l − vib
ib (V ′ = 0 →V 0) − ⎜ e − e e ⎟ + ⎜ e − e e ⎟ ⎥ cm −1
⎣ ⎝ 2 4 ⎠ ⎝ 2 4 ⎠⎦
⎡ ⎛ 1788.2 16.43 ⎞ ⎛ 1641.4 11.67 ⎞ ⎤ −1

= ⎢19378 − ⎜ − ⎟+ − ⎟ cm
⎣ ⎝ 2 4 ⎠ ⎝ 2 4 ⎠ ⎥⎦
=( .99 + 817. ) cm −11 = 19306.79 cm

.99 1
= 230.85 kJ mole−1
(b) The fundamental mode of vibration of C2 in the ground and excited states is determined from
Eq. (3.57), i.e.
v w e ( − x e )cm −1
Thus,
v for ground state = (1641.4 − 11.67) cm−1 = 1629.7 cm−1
v for excited state = (1788.2 − 16.43) cm−1 = 1771.7 cm−1
12 × 12
μm = = 6 amu
12 + 12
From Eq. (3.19), we now obtain the C–C force constant in the ground and excited states of the molecule.
2
⎡ 1629.7 ⎤
k for the ground state = ⎢ ⎥ ×6 9.38 millidynes/Å
⎣1303.16 ⎦
2
⎡1771 7 7 ⎤
k for the excited state = ⎢ ⎥ × 6 11 09 millidynes/Å
⎣ 1303.16 ⎦
Problem 5.5: (a) The frequencies assigned to the vibrational bands in the absorption spectrum of AgI near
3200 Å are listed here.
Frequencies (cm−1) V′ → V″
31642 0→5
31393 0→2
30956 1→0
31577 0→4
31281 0→1
30753 2→0
31491 0→3
31163 0→0
30549 3→0
31666 0→6
(i) Suggest assignment for the bands at 31256, 31086, 30769, 30868, 30656, 31157, 30953 cm−1.
(ii) Assign frequencies corresponding to the
2 → 2, 2 → 3, 2 → 1, 1 → 3 bands
(b) The frequencies assigned to the vibrational bands in the electronic absorption spectrum of nitrogen are
listed below.
V′ → V″ Frequencies(cm−1)
0→0 29662
0→1 31652
1→0 27957
0→3 35463
2→0 26281
3→0 24634
0→2 33593
4→0 23022
5→0 21425
(i) Suggest assignment for the bands at 28271, 32082, 26624, 27226 cm−1 and (ii) assign frequencies corre-
sponding to the 2 → 2, 3 → 2, 1 → 3 and 4 → 2 bands.
Solution (a) Based on the data in the problem, construct the Deslandres Table. The missing elements in the
Table are generated on the basis of vibrational intervals between the adjacent elements in the first row and first
column of the Table. The numbers in parenthesis represent the vibrational intervals.
V″ / V′ 0 1 2 3
0 31163 (207) 30956 (203) 30753 (204) 30549

(118)
1 31281 31074 30871 30667
(112)
2 31393 31186 30983 30779
(98)
3 31491 31284 31081 30877
(86)
4 31577 31370 31167 30963
(65)
5 31642 31435 31232 31028
(24)
6 31666 31459 31256 31052
(i) From the Table, the following bands are assigned to the frequencies in question:
2 → 6, 2 → 3, 3 → 2, 3 → 3, 3 → 1, 2 → 4, 3 → 4.
(ii) From the Table, the frequencies assigned to the bands in question are 30983, 31081, 30871, 31284.
(b) Proceeding as in the part (a) of this problem, we arrive at the following assignments:
(i) 2 → l, 2 → 3, 3 → l, 5 → 3
(ii) 30212, 28565, 33758, 26953 cm−1
5.3.4 Intensity Distribution in Vibrational Electronic Spectra

From Franck–Condon Principle
According to the Franck–Condon principle, also sometimes called the vertical principle, the internuclear configura-
tion of a diatomic molecule does not change appreciably during an electronic transition since the time scale of com-
plete electron transition is of the order of ~10–15 s and that of complete vibrational transition is of the order of ~10−13
sec, i.e. the frequency of electronic transition is hundredfolds faster than that of vibrational transition. The molecule
spends most of the time at the extreme ends of its vibration motion when the vibrational quantum number V is large
and very little time in between since the vibrational kinetic energy is zero at the extreme ends. Quantum mechanically,
the probability of transition V′ → V″ is determined from the quasi-moment that corresponds to the given transition.
v ′ = 10
M t (V V ) ∫ψ V ( r ′′ )M t ψ V ( r ′ )dτ (5.16)
Here, ψ v ′ and ψ v ′′ are the vibrational wave functions cor-

responding to the ground and excited electronic states of the v′ = 4
diatomic molecule.
The form of the wave functions has been shown in
Fig. 5.5. The nuclei are most likely to be found at distances
apart given by the maxima of curve for each vibrational state.
The vibrational wave functions have a perceptible magnitude v′ = 3
in a certain interval of values of r. Due to this diffuseness, they

overlap in the interval Δr and this, according to Eq. (5.16),
leads to the possibility of a transition from a given level V′ not
only to the upper lying level V″ but also to the adjacent levels
as shown in Fig. 5.5, i.e. V′ (= 0) → V″ (0, 1, 2, 3…), 0 → 0, v′ = 2
0 → 1, 0 → 2, 0 → 3,..., etc.
Quantum mechanically, the absorption or emission is an
instantaneous process and occurs without any change in elec-
tronic configuration. Such transitions, are termed vertical and
v′ = 1
correspond to the arrows of Fig. 5.6. No change in bond length
occurs in vertical transitions while in adiabatic transitions,
length changes. Adiabatic transitions are strong in photoioni- ΨΨ′′ Ψ
sation processes. The two Morse curves shown in Fig. 5.6 have
v ′= 0
different curvatures, different vibrational intervals and differ-
ent equilibrium distances. The square of the vibrational wave
function is greatest at extreme ends of vibrational levels for 0
−0.2 −0.1 0 0.1 0.2A
larger V′, suggesting that the probability of finding the nuclei
is greatest there. The most probable transition is the vertical
Fig. 5.5 Eigen functions (dashed curves) and probability
arrowed line which starts at the point corresponding to the
internuclear distance in the ground state and terminates at the densities (solid lines) of a harmonic oscillator for V = 0,1,
2, 3... and 10.
point where it meets the excited state curve at a vibrational
level. Thus, the most probable internuclear distance for the
vibrational levels other than V′ = 0 corresponds to the extreme
ends and to the mid-point for V′ = 0. The relative intensity
of the bands in the absorption or emission spectrum depends
on the relative horizontal separations of the potential energy S1
curves of the ground and excited-state molecules. Thus, the
band associated with quantum transitions in which there is no 3
change either in internuclear distance or the nuclear motion 2
U (r)
1
V ′′= 0
will have the highest intensity. Intensity distribution among
the bands of an electronic transition not only differs from mol-
ecule to molecule but also from one progression to another of S0
the same molecule. There are three possible main situations:
(i) Internuclear distances are equal in the lower and upper
states, i.e. rv rv ′′, (ii) upper state internuclear distance is
slightly greater than that in lower state, i.e. rV ′′ > rV ′′, and (iii) 3
2
1
upper-state internuclear bond distance is appreciably larger as V ′= 0
compared to that in the lower state, i.e rV ′′ >> rV ′ involving the r(Å)
ground state and the first excited singlet states as shown in Fig. 5.6 Schematic diagram showing the probabilities
Fig. 5.7(a) – (c). of transition after the Franck–Condon principle.
(a) CASE (I) When re′ = re′′, the transition V′ = 0 → V″ = 0 is the most probable and the internuclear distance does
not change during the transition. Consequently, an intense band is expected corresponding to this transition as shown
in Fig. 5.7 (a). For other transitions V′ (= 0) → V″ (1, 2, 3,...), the probability of the successive transitions decreases
sharply and the bond distance also changes which is contrary to the Franck–Condon principle. Consequently, weak
bands with decreasing intensity are expected corresponding to these transitions as shown in Fig. 5.7 (a).
S1 S1 X Y
m
m−1
4
3 S1
2
1
U(r)
V⬙ = 0
S0 S0 S0
V⬘ = 0
r(A)
Spectra 0 0
0 2
Continuum of absorption
0 3
0 1
0 1 0 m
0 2 0 4 0 m−1
0 3 0 0
0 4
Wave number (cm−1)

(a) (b) (c)
Fig. 5.7 Intensity distribution in absorption bands from Franck-Condon principle (a) internuclear distances are equal in upper and lower
states (b) inter nuclear distance in the upper state is a little greater than in lower state (c) inter nuclear distance is considerable greater than
that in the lower state.
(b) CASE (II) In the second case, where re′′ > re′ , the probability corresponding to the vertical transition
V′ (= 0) → V″ (= 2) is maximum and also the difference between internuclear distance in the ground and excited
states corresponding to this transition is expected to be minimum. Consequently, the intensity of the band cor-
responding to this transition will be the maximum. The transitions from V′ (= 0) to lower than V″ (= 2), i.e. V″
(= 0, l) states are (i) more probable, and (ii) differences in internuclear bond distances are small as compared
to those from V″ (= 0) to higher than V″ (= 2), i.e. V″ (= 3, 4) states. Thus, the bands on the lower wave-number
side of the v20 band will be more intense than their counterparts on the higher wave-number side of the v20 band.
In general, the intensity of the bands corresponding to the V′ (= 0) → V″. (0, 1) successive transitions increases
and becomes maximum for 0 → 2 and then decreases for V′ (= 0) → V″ (3, 4...) successive transitions as shown
in Fig. 5.7(b).
(c) CASE (III) When re′′ >> re′ , the transition probability increases and the difference between the internuclear
bond distance in the ground and excited states increases as the vibrational quantum number grows. Consequently,
the intensity of the respective bands in the progression V ′ (= 0) → V ′′ (0, 1, 2, 3...) will increase and becomes
maximum at very high values of V″ = m (Fig. 5.7c). The transition finishes at a state represented by the dotted line
XY where the energy of the excited molecule exceeds its own dissociation energy. The molecule in this state will
dissociate into atoms without any vibrations and atoms may take up any value of kinetic energy. All transitions
from the ground state to the points above this level will lead to a continuum of absorption since such transitions
are not quantised. The intensity distribution in such a case has been sketched in Fig. 5.7 (c).
Further, in case when V ′ ≠ 0 in V ′ progressions, there are two bands with maximum intensity that correspond
to the maxima of the wavefunction Ψv'' at the two turning points of the nuclei (Fig. 5.5). That is why we will
have two intensity maxima in these series. The same is true when the initial level is V″ level (emission). For each
value of the number of V″ (for V″ > 0), there will be two value of V ′ that correspond to two of the intense bands
of the V′ progressions in the series. The two maxima will coincide only if the equilibrium separations are the
same in both the electronic states. At V″ = 0, we will have a series with one intensity maximum just as in case
of absorption when the zero vibrational level (V′ = 0) is the initial state of the molecule, and there is only one
intensity maximum in the series.
5.4 ROTATIONAL FINE STRUCTURE OF ELECTRONIC-VIBRATION

STRUCTURE OF A DIATOMIC MOLECULE
We know that the electronic spectra of diatomics consist of one or more series of convergent lines. These series of
convergent lines constitute the vibrational coarse structure associated with each electronic transition. Normally,
each of these lines is broad and diffuse, and under high resolution appears as a group of a large number of very
close lines. Such a cluster of lines is termed fine structure.
Without taking into consideration the centrifugal distortion constant, let us determine the peculiarities of the
rotational fine structure of the electronic spectrum of a diatomic molecule.

The molecular electronic terms and selection rules for rotational fine structure have been discussed in detail in
Chapter 2 on microwave spectroscopy under the title hyperfine structure. The selection rules for J depend upon
the molecular electronic transition. For lower and upper electronic states, in which there is no electronic angular
momentum about the bond axis, the selection rule is
1 1
ΔJ ly f ∑→
lower
∑
upper
ii , ΔΣ
Δ =0 (5.17)
Thus, for such 1 Σ → 1Σ transitions, only P − (ΔJ = −1) and R − (ΔJ = +1) branches will appear in the electronic
spectrum.
For all other transitions in which either the upper or the lower electronic states (or both) have angular momen-
tum about the inter nuclear axis, the selection rule is
Δ J = 0 or ±1 and for both the cases J = 0 ↔ J = 0 is forbidden. (5.18)
Thus, for such transitions, in addition to P− and R–branches, Q (ΔJ = 0) branch will also appear in the elec-
tronic spectrum. Note that for L ≠ 0, the selection rule Δ J = 0 or ±1 and J′ = 0 ↔ J″ = 0 is also allowed.
5.4.2 Transition Between Rotational Levels Associated with a Particular

Vibrational Level in the Electronic Spectrum of a Diatomic Molecule
We have already proved that the frequencies of separate lines in the electronic spectrum of a diatomic molecule
are represented by
v = v elec + v vib + v rot (Eq. 5.6)
The frequencies of the individual rotational lines of a particular vibrational band characterised by vibrational
quantum numbers V′ and V″, thus, can be represented by the formula
v rot elec = v eelec
lec + v vvib
ib BV ′′ J ′′(J ′′ + ) − BV J ′(J
(J ′ + )
lec vib(V ′V ′′ ) + BV J ′′(J ′′ ) BV ′ J ′(J ′ + ) (5.19)
Here BV ′′ and J″ refer to the upper electronic state and BV′ and J′, to the lower electronic state of the diatomic
molecule contrary to infrared spectroscopy where they belong to the same electronic state. Very often, when
the electron is excited from the bonding orbital of the molecule in the ground state, the bond in the higher
electronic state becomes weaker and longer. Further, since the quantity B, the rotational constant, is inversely
proportional to the equilibrium moment of inertia and hence the square of the equilibrium bond distance of the
molecule, it decreases during the electronic transition. Thus, in such a case, BV″ < BV′. Sometimes, when the
electron excitation is from the antibonding orbital of the molecule, the reverse happens, i.e. BV″ > BV′. v elec − vib(V ′V ′′)
is the wave number of an electronic vibrational transition. This corresponds to any of the transitions represented
by V′ (= 0 1, 2...) → V″ (0, 1, 2, 3...), i.e. 0 → 0, 0 → 1…, 1 → 0, 1 → 2, etc. Thus, it represents the origin
of a particular band. Applying the selection rules ΔJ = ±1, ΔJ = 0, ±1; J′ (= −0) � J″(= 0) to Eq. (5.19) for
the spectral lines, we arrive at the expressions for the frequencies of P-, Q-, and R-branches of the electronic
spectrum of the diatomic molecule
P:ΔJ = −l, J″ − J′ = – l or J ′ = J ″ +l
vP =v (V V ) + BV J ′′ (J ′′ (J ′′
) BV ′ (J )(J ′′ + )
= v elec − vib(V V ) + (J ′′ + ) [ ′′ + − − ′′ + − ]
or vP v (V ′ V ) − ( BV BV ′ ) J ′ ( BV − BV )J ′ 2 (5.20)
where J ′ = 1,2,3… J ′ ≠ 0 otherwise J ″ = −1 which has no meaning.
Equation (5.20) reveals that no rotational line of the P-branch appears at the band origin v P v l − (V ′V ′′ )
R: ΔJ = + 1, J″ − J′ = + 1,J″ = J′ + 1
v R = v elec vib(V V )
vib(V + ( BV BV ′ ) (J ′ ) ( BV ′′ − BV )(J
(J ′ + ) 2
where J′ = 0, 1, 2,… (5.21)

Here too, no rotational line of the R-branch will appear at the band origin.
Q:ΔJ = 0, J″ − J′ = 0, J ′′ = J ′, J ′ = 0 J ′′ = 0
vQ = v V V ) + ( BV BV ′ ) J ′ ( BV − BV )J ′ 2
where J′ = 1,2,3... (5.22)

since J′ = 0 � J″ = 0, no line from the Q-branch will appear at the origin. However, for L ≠ 0, the rule
J′ = 0 ↔ J ″= 0 is also allowed and thus J′ = 0, 1, 2, 3,..., the lines of Q-branch will start from the origin.
Equations (5.20) and (5.21) are clubbed together into
v PR = v elec − vib(V
vib(V V ) + ( BV BV )m ( BV ′′ − BV )m 2
where m = ± l, ±2, ± 3… (5.23)
Here, m is negative for P-branch (Δ J = −1) and positive for R-branch (Δ J = + 1). m can’t be zero otherwise for
P-branch J ″ = −1 which is not possible.
The wave numbers of the rotational fine lines in the P- R- and Q-branches as worked out from Eqs (5.23)
and (5.22) respectively for BV″ (= 10 cm−1) < BV′ ( = 12 cm−1). 20 per cent difference between the two in
the electronic-vibrational spectrum of the diatomic molecule have been listed in Table 5.2, and the corres-
ponding spectrum is sketched in Fig. 5.8. We already know that even the higher rotational levels are
Table 5.2 Position of rotational fine lines relative to v elec − vibi for BV ′′ < BV ′ and 20 per cent
difference between the two in the electronic vibrational structure of diatomic molecule,
BV 10 1
BV ′ = 12 cm−1 , v PR = v elec − vibi + 22 m − 2 m2, m is positive for R-branch and negative for
P– branch. v Q v elec − vibi − 2 J′( J + 1) .
m/J ′ v R ( cm 1 ) −v P ( cm 1 ) −v(QQ ) ( cm 1 )
1 20 24 4
2 36 52 12
3 48 84 24
4 56 120 40
5 60 160 60
6 60 204 84
7 56 252 112
8 48 304 144
9 36 360 180
10 20 420 220
Q-Branch
300 200 100

nQ (cm−1)
R-Branch
0 10 20 30 40 50 60
nR (cm−1)
P-Branch
500 400 300 200 100

nP (cm−1)
Fig. 5.8 Electronic vibrational spectrum of a diatomic molecule.

populated even at room temperature, consequently a large number of rotational lines appear in the P- and
R-branches and wherever possible in the Q-branch also in the spectrum. The lines in the R branch become closely
spaced on the high wave-number side and crowd together and eventually a maximum value of v R is obtained at
a particular value of m. This limiting position of the band corresponding to maximum value of v R is called band
head, and the lines become widely spaced on the low wave-number side of the band head called the tail of the
band. This decline in total intensity away from the band head forms a line distribution in the band called shading
or degrading of the band. The shading of a band towards longer wavelengths or low frequencies is termed red
shading and the band is said to red-shaded as in the present case and violet-shaded if the shading is towards the
lower wavelengths. The red shading is due to the dominance of m2 over the 22m term. Further looking at the curve,
it is found that the lines spread out in the P branch as m decreases. The Q-branch also follows the direction of
the P-branch. However, the spacing between adjacent lines is much closer since (BV″ − BV′) is numerically smaller
than (BV″. + BV′) and also there is no change in J′. The spacing between adjacent lines grows with J′. It may be
noted that in many cases there is no band head. This occurs when the difference between BV″ , and BV′ , is slight.
The characteristic band constant whose value corresponds to wave number of the band origin also called the zero
line, i.e. v elec − vib(00) is absent from the spectra of diatomics. The position of zero line in a band can be determined
from the characteristic gap that disrupts the line sequence. The exact value of v elec − vib(00) band can be determined
from a detailed analysis of the rotational structure of the bands. The set of values of v elec − vib so obtained aids in
establishing the Deslandres formula for a given system of bands, i.e. determining the constant velec and also the
vibrational constants we and xe for both the combining electronic terms.
The foregoing arguments are true for BV″ < BV′. In case BV″ > BV′, all the discussion is reversed and (i) the band
head appears in the P-branch towards lower wave numbers or higher wavelengths away from the band origin
v elec − vib(00) and is shaded towards longer wave numbers, i.e. lower wavelengths away from the band head and is said
to be violet shaded, (ii) the lines in the R-branch still appears on the higher wave-number side of the band origin
and show an intense separation with m and (iii) the lines of the Q-branch spreads out to lower wavelengths or
higher wave-number side of the band origin and the separation between adjacent lines increases with J′.
Problem 5.6: The fine structure line of C2 band at 19378 cm−1 fits well into the equation
v rot elec = ⎡⎣19378 + 3.3853 m + 0.1201m 2 ⎤⎦ cm −1
(a) Find the value of the band head and position of the band head. (b) Calculate the separation between zero line
and band head and state the direction of shading of the band. (c) Determine moments of inertia and internuclear
distances in the two electronic states of the molecule.
Solution (a) The rotational fine structure of lines of a band is expressed by the relation
v rot elec = v rrotot − elec + ( BV BV ′ ) m ( BV ′′ − BV )m 2 , cm −1 (Eq. 5.23)

Comparing Eq. (5.23) with the given relation, we obtain
v elec − vib(V V =0 ) or v 00 = 19378 cm −1
BV″ + BV′ = 3.3853 cm−1
BV″ − BV′ = 0.1201 cm−1
Solving these for BV d BV ′, we get

BV″ = 1.7527 cm−1
BV′ = 1.6321 cm−1

The value of m at band head is given by
( BV BV ′ ) 3. (cm −1 )
m=− =− = −14
2( BV BV ′ ) 2 × 0. (cm −1 )
The negative value of m reveals that the band head lies in the low wave-number side of the band origin or null
line, i.e. P-branch of the spectrum. The band is shaded towards the high wave-number side or towards violet.
The position of the head, i.e. maximum value of v rot elec corresponding to m = −14 is determined from the equa-
tion.
v head = [19378 + 3.3853 (−14) + 0.1201 (−14)2]cm−1
= (19378 − 47.39 + 23.53) cm−1

= (19378 − 23.86) cm−1
= 19354.14 cm−1
(b) The separation between band head and the null line is (19354.14 − 19378) cm−1 = −23.86 cm−1. Thus the
head is shaded towards violet.
(c) The moments of inertia and bond distances in both the states of the molecule are determined from
BV d BV ′
h 16.8575
For the ground state, I′ I′ = = = 10.328 au.
8π BV ′ c 1.6321
2
12 × 12
Since mm for C2 = = 6.0 amu, the bond length for C2 in the ground state is
12 + 12
I′ 10.328
r′ = = = 1.7213 = 1.311Å
μm 6
Similarly, for the excited state,
16.8575
I ′′ = = 9.618 au
1.7527
and
I ′′ 9.618
r ′′ = = = 1.603 = 1.266Å
μm 6
5.4.3 The Fortrat Diagram

It has already been stated that frequencies of the individual rotational lines of a band that is characterised by defi-
nite values of vibrational quantum numbers V′ and V″ (that result because of electronic transitions between two
electronic states of a diatomic molecule) are given by
v = v elec − vib(V V ) + BV J ′′(J ′′ ) BV ′ J ′(J ′ + ) (Eq. 5.19)
Here, BV″ and BV′ correspond to two different electronic states, say Ψelec′′ and Ψelec
′ , of a diatomic molecule. It
can be easily proved from Eq. (5.19) that the rotational frequencies of the P-, Q-, and R-branches in the electronic
spectrum are given by
P :v P v ( BV ′′ + BV )J ′ ( BV BV ′ )J ′ 2 ( ΔJ = −1) (Eq. 5.20)
Q :vQ v ( BV ′′ − BV )J ′ ( BV BV ′ )J ′ 2 ( ΔJJ = 0) (Eq. 5.21)
R :v R v ( BV ′′ + BV )(J ′ + 1) ( BV BV ′ )(J ′ 1) 2 ( ΔJJ = +1) (Eq. 5.22)
and v PR = v elec − vib + ( BV BV ) m ( BV ′′ − BV )m 2 (Eq. 5.23)
where m = ±1, ±2, ±3…, m is negative for P-branch and positive for R-branch. These three quadratics in J′s,
represent three parabolas, i.e P, Q and R and are usually termed Fortrat parabolas. Thus, the Fortrat diagram
comprises parabolic curves corresponding to P-, Q- and R-branches of an electronic-vibrational band that is
characterised by definite values of vibrational quantum numbers V′ and V″, and are constructed by plotting the
frequencies of the rotational lines of V′→ V″ band against the rotational quantum numbers in the lower energy
electronic state of the molecule. The Fortrat diagram for BV″ (= 10 cm−1) < BV′ (12 cm−1) and a 20 per cent difference
12
10 R-Branch
m/J′
Null line
6
Q-Branch
4
Band head
2
0
−60 −40 −20 0 20 40 60
−2 −
ν P, Q, R(cm−1)
P-Branch Band origin
m
−4
−6
Fig. 5.9 Fortrat diagram of Bv’′ (=12 cm ) < Bv′ (=10 cm–1); 20 per cent difference between BV′′ and BV′.
–1
between the two is shown in Fig. 5.9. The position of the band head depends upon the difference between the
values of BV ″ and BV ′.
The Fortrat diagram can be explained on the basis of arguments on the rotational fine structure of a vibrational-
electronic transition of a diatomic molecule reported in Section 5.2. A usual property of the Fortrat diagram is
that the band head is due to the turning point in the PR parabola. The problem arises as to which of the branches
(R or P) has a turning point. This problem has been solved by the sign of quadratic term in the expression (5.23).
The sign is determined as follows.
Differentiating Eq. (5.23) we get
d v PR
= ( BV BV ′ ) + 2m ( BV − BV ′ )
dm
d v PR
At the tuming point, =0
dm
∴( ) + 2m(( )=0
or
( BV BV ′ )
m=− at the band head.
2( BV BV ′ )
Thus, the position of the band head depends on the difference between the values of BV ′ and BV ′′ ,
(i) For BV BV ′ m is positive. Thus, the band head occurs in the R-branch.
(ii) For BV BV ′ m is negative. Thus, the band head lies in the region of m negative, i.e. in the P- branch.
Substituting the value of m in Eq. (5.23), we obtain the expression for the wave-number separation between the
(B BV ′ ) 2
band head v PR and the band origin v elec − vib , i.e. v PR v elec − vib = − V (5.24)
4( BV BV ′ )
v PR v elec − vib is positive for a band shaded to red ( BV BV ′ ), while it is negative for a band shaded to violet
( BV B V ′ ).
The value of J′ at which the vertex on the band head of the Q-parabola lies is obtained by differentiating
Eq. (5.22) with respect to J′, i.e.
dv Q
= (B V B V ′ ) + 2 (B V B V ′ ) J ′
dJ ′
dv Q
At the turning point of the parabola, =0
dJ ′
1
∴ ( BV BV ′ ) + 2 ( BV BV ′ ) J ′ 0 J′ = − (5.25)
2
Thus, unlike the PR band head, J ′ , i.e. the Q-band is independent of the values of BV ′′ and BV ′ . However, if
the ( BV BV ′ ) difference is large, the spacing between the adjacent Q-lines increases rapidly, consequently the
Q-head is not strongly expressed.
5.4.4 Simplification of the Analysis of Rotational Fine Structure

in the Vibrational Electronic Spectrum
The analysis of the rotational spectrum of diatomics is simplified by setting up definite relations between frequen-
cies of the rotational lines, in order to isolate individual rotational terms and rotational constants of the molecule.
This is achieved by denoting the rotational term by F(J), i.e. F(J) = BV J(J + 1). Thus, in terms of F(J), Eq. (5.19)
can be expressed as
v (J ) = v elec − vib F ′′ (J ) F ′ (J ) (5.26)
The expressions for P Q Q-, and R-branches in terms of F(J) will then appear as follows:
For P-branch: ΔJ = −1, J ′′ − J ′ = −1, J ′′ = J ′ − 1
vP =v + BV ′ (J ′ − 1) J ′ BV ′ J ′ (J ′ + 1) .
or in terms of F(J)
v P (J ′ ) = v elec − vib F ′′ (J ′ 1) F ′ (J ′ ) (5.27)
For R-branch: ΔJ = +1, J ′′ = J ′ + 1,
vR =v + BV ′ (J ′ + 1)(
) (J ′ 2) BV ′ J ′ (J ′ + 1)
or in terms of F(J)
v R (J ′ ) = v elec − vib F ′′ (J ′ 1) F ′ (J ′ ) (5.28)
For Q-branch: ΔJ = 0, J ′′ = J ′, J ′′ = 0 ↔
/ J′ = 0
vQ = v + BV ′ J ′ (J ′ 1) BV ′ J ′ (J ′ + 1)
or in terms of F(J)
v Q (J ′ ) = v elec − vib F ′′ (J ′ ) F ′ (J ′ ) (5.29)
From the above expressions (5.27 – 5.29), the following relations are obtained for separate lines.
v Q (J ′ ) − v P (J ′ ) = v R (J ′ − 1) − v Q (J ′ − 1)
⎫
⎪
= F ′′(J ′ ) − F ′′(J ′ ) ⎪
⎪
= BV (J ′ + 1) ′ ′(J ′ 1)⎪
⎪
= 2BV ′′ J ′ ⎪
⎪
⎬ (5.30)
v Q ( ′ 1) v P ( ′ ) v R ( ′ 1) − v Q ( ′ ) ⎪
⎪
= F (J ′ ) − F′′(J ′ − ) ⎪
⎪
= BV ′ J ′(J ′ + ) − BV ′ (J ′ − )J ′ ⎪
⎪
= 2 BV ′ J ′ ⎪⎭
Equation (5.30) reveals that the frequency difference between two lines in a band is repeated twice and is a
quantity that is a multiple of 2BV′ or 2BV″ . Thus, the starting point in the analysis of rotational structure of a band
is to determine the difference that satisfies the relations expressed by Eq. (5.30). Hence, we can directly find the
values of the constants BV′′ and BV ′ of the molecule and its rotational terms, e.g. for J′ =1, we obtain: F″ (1) − F″
(0) = F″ (1) = 2 BV″ and F′ (1) − F′ (0) = F′ (1) = 2BV′.
These relations straight away give the constants BV″ and BV′ and the terms F′ (1) and F″(1). For J′ = 2, we
have:
F″(2) − F″(1) = 4 BV″ and F′ (2) − F′ (1) = F (1) = 4 BV′
As has already been stated in the text that for 1 ∑ ↔ 1∑ transition, due to Δ J = ± 1, only P- and R-branches
appear in the spectrum and the Q-branch is absent. However, in the 1 ∑↔ 1∏ transition, in addition to P- and
R-branches, the Q-branch also appears in the spectrum. But here the L- type doubling of ∏ term should also be
taken into condsideration. Since change in energy associated with L-doubling in Eqs.(5.20–5.22) is not taken
into account, these formulae and also the relations in terms of F(J) (5.27 –5.30) derived from them are not very
accurate. However, in spite of this, it in no way affects the number of branches in the spectrum corresponding to
the combination 1 ∑ − 1∏. The rotational ∑ − terms are alternately positive and negative and the components of
L-doubling of Π-term, i.e. ∏a and ∏b also have different signs. Based on the selection rules Δ J = 0, ±1 and
+ ↔ − only three combinations of each 1∑ term, i.e. three simple branches are possible. These combinations can
schematically be represented as follows.
J 0 1 2 3 4
− + − +
1Π
u + − + −
+ − + − +
1Σ +
g
The rotational structure of bands in the combination of other terms is beyond the scope of the book. However,
the results of some of the combinations are stated here.
Combination Branches
∏↔ ∏ ⎫
3 1
⎪ six (two R-, two Q- and two P-branches)
⎬
∏↔ Δ
1
1
⎪⎭
∑↔2∑
2
Four (two R-, and two P-branches).
∑↔2∏
2
Twelve
5.5 ELECTRONIC ABSORPTION SPECTRA OF POLYATOMICS

The vibrational and rotational structures in the electronic spectra of polyatomics is more complex than that of
diatomics because a large number of constants, viz., vibrational and rotational, are involved in them. However, the
rotational structure of the bands of symmetric linear molecules of the type CO 2 is close to that of diatomics. In
spite of the complexities involved in vibrational and rotational structures in the electronic spectra of polyatomics,
the spectra of polyatomics aid in assigning the fundamental modes of vibrations both in the ground and excited
electronic states of the molecule as evidenced by the following examples. The ultraviolet-spectral data on methyl
nitrite (I), nitromethane (II) and dimethyl nitrosoamine (III) as deduced from their near ultraviolet electronic
absorption spectra in the vapour phase have been listed in Table 5.3.
O O O
CH3
O N CH3 N N N
CH3 O CH3
(I) (II) (III)
Differences from 0−0 band are given within a parentheses. The intensities of bands decline progressively
downwards.
Table 5.3 UV spectral data of methyl nitrite, nitromethane and dimethylnirosoamine in the vapour phase.
Methyl nitrite Nitromethane Dimethylnitrosamine

1 1 1
n( ) Assignment n( ) Assignment n( ) Assignment
32777(1343) v′ (NO2)
27432 (1080) v′ (C−O)
33213 (907) v′ (C −N)
28512 0-0 34120 0−0 26485 0-0
29481 (969) v″ (C - O) 34955 (835) v″(C− N) 27469 (984) v″(N-N)
30465 (1953) 2 × v″(C-O) 35704(1584) 2 × v″ (C−N) 28400 (1915) 2 × v″(N-N)
31367 (2885) 3 × v″(C-O) 36619(2499) 3 × v″ (C−N) 29518(3033) 3 × v″(N-N)
32212 (3700) 4 × v″(C-O)
Methyl nitrite weak band on the long-side wavelength side is comparable to the ν ′ (C – O) fundamental in the
infrared spectra. The other fundamentals are 1600 ν ′ (N = O), 800 ν ′ (N − O) and 600 cm−1 δ (ONO). Thus, it
is surmised that the bands in the progression are most likely to arise from the ν ′′ (C – O) in the first excited state.
As expected, the ν ′′ (C – O) is lower than in the ground state. The combination tones with progressively decreas-
ing intensities follow this band.
In case of nitromethane, the bands at 1377, 658 and 918 cm−1 in its vapour phase spectrum is assigned to NO2
symmetric stretching, NO2 symmetric bending and C – N stretching modes respectively. The NO2 bending mode
is not involved with the progression. The progression is because of C – N stretching mode since the difference
between the ground and excited state frequencies (907- 835 = 72 cm−1) is of right order.
The 1023 cm−1 IR band of dimethylnitrosamine increases in frequency on changing from non-polar to polar
solvents and is assigned to the N – N stretching mode. Bands at 1489 and 1016 cm−1 in vapour phase IR spectrum
are due to N = O and N – N stretching modes. Thus, the observed interval of 984 cm−1 in UV absorption spectrum
of dimethylnitrosamine can only be due to the ν (N – N) stretchning mode.
Thus, it follows that electronic absorption spectroscopy is complementary to infrared- and Raman- spectros-
copy or vice versa
5.6 THE ISOTOPE EFFECT IN MOLECULAR ELECTRONIC SPECTRA

This effect applies equally to molecular spectra with electronic excitation and to infrared-, and Raman-, spectra.
The vibrational isotope effect is characterised by the difference in natural frequencies of the isotope molecules
whereas the rotational effect, from their differences in moments of intertia. The isotope effect in molecular spec-
tra is a sensitive and precise method of identifying isotopes and determining their relative abundances from their
relative intensities in the spectrum, e.g. a number of isotopes H2 (D), C13, N15,O17, O18, etc., have been discovered
by molecular spectroscopy.
5.6.1 Vibrational Isotope Effect

We know that we 303. 6 k / μ m
hw e 1303.16 h k
and xe = =
4 De 4 De μm
Accordingly, the natural frequency and anharmonicity constant are inversely proportional to the square root of
1
the of the reduced mass ( μ m ) of the diatomic molecule, i.e. w e or x e α while we xe to the reduced mass
μm
1
of molecule, i.e. w e x e α .
μm
m1m 2
Here, mm = and m1 and m2 are the masses of the atoms comprising the diatomic molecule.
m1 + m 2
In order to study this effect, a quantity P rho is introduced.
∗
m
(5.31)
m
where m and m are the reduced masses of the deutrated and undeuetrated isotopes of the molecule.
The magnitude shift of zero line ( Δv )vib in the spectra of diatomics can be determined from the Deslandres
Table as follows.
2
1 1
v − ′′ e′′ V +
2
⎡ 1 1
2
⎤
−⎢ − ′ e′ V ′ + (5.32)
2 2
⎡ 1 1
2
⎤
v ⎢ − 2
′′ e′′ V ′′ +
⎣ 2 ⎦
1 1
2
⎤
− ′ e′ V ′ + (5.33)
2 ⎦
Subtracting Eq. (5.32) from Eq. (5.33), we get
(Δ = − lec vib
⎡ 1 1 ⎤⎞
( ) ′′ − +
2 2
1 1 ⎤
− 2
− 1) ′′ ′′ V + w ′+ (5.34)
2 2 ⎦
Disregarding the anharmonicity constant, Eq. (5.43) reduces to
1 1 ⎤
= w e′′ ′
e ′+ (5.35)
2 2
∗
In order to determine Δv0 for different transitions (r − 1) for isotopes of the type X X m1 2
(e.g. 16O2,
16
O 18O, 12C2, 12C 13C; 14N2, 14N l5N; etc.) is expressed as
m +m m + −
= = =
m 1 +m
∗
m m2 ( )
On rationalising and expressing = Δm we obtain
Δm
− = (5.36)
2 m +
Approximately, Δm is small compared to m2 or m1 (since m2 = m1), consequently (5.36) reduces to
Δ Δm
or (5.37)
With the aid of Eqs (5.35) and (5.37), one can study the vibrational isotope effect in the isotopic molecules of
*
the type, X 2 , 2
. For more accurate results, the anharmonicity factor should be retained in Eq. (5.34).
The second term of Eq. (5.34) then needs be multiplied by the quantity (r2 −1) which is derived as follows.
We know that
= m
= ∗
×
m
or
2 2m
)
( )
2 Δm
Thus,
m − +
Δm being small compared to 2m1 can be disregarded in the denominator, consequently.
2 Δm
(5.38)
2 m1
*
Problem 5.7: Prove that for molecular isotopes of the type, X Y m , and 2 2
(e.g. H35CI, H37CI),
the term
m m
wher − m 1
2m 1 + m2
m + −
− − = ×
1 2
∗
m m
On rationalising, we get
m −
− = 1 2
m 1 + + + m2
m 2 m2
=
m m +m m
or
m m
− =
m 2 + + +m m
Disregarding Δm relative to m2, we get
m m m
− = (5.39)
m m + m2 2
2m 1 + m2
Problem 5.8 The values of we and xe in the ground state (3pu) and a particular excited state (3pg) of C2 are
1641.4 cm−1, 7.11 × 10−3 respectively. Calculate the vibrational isotope shift of zero line for the transitions V ′→ V ″;
0 → 0, 0 → 1, 1 → 1, and 1 → 2 in the electronic spectra of the C2 molecule.
Solution The isotopes of C2 molecule are 12C 12 C and 12C 13C. In order to determine the vibrational isotope
shift (Δν0)vib, the quantity r −1 needs be determined first.

Δm 1
− = =
4 m1 4 × 12
Substituting this value of (r −1) together with the values of natural frequencies into Eq. (5.35), we obtain
⎡ ⎛ 1⎞ ⎛ 1⎞ ⎤
( ) vib = 0.02 ⎢1 88.2 ⎜V ′′ + ⎟ − 1641.4 ⎜V ′ + ⎟ ⎥
⎣ ⎝ 2 ⎠ ⎝ 2⎠ ⎦
The values of (Δv0)vib in the asked transitions as deduced from the above expression are as follows:
Transition (Δv0 )vib(cm−1)
V→V″
0→0 1.5
0→1 37.5
1→1 4.4
1→2 40.1
Problem 5.9: The values of we in the ground state (1∑) and a particular excited state (3∑) of the terrestrial
atmospheric O2 are 1580.3 and 1432.6 cm−1respectively.Calculate the isotope shift of zero line in a single progres-
sion of bands V′(= 0) →V″(= 0,1,2,3 ) in the electronic absorption or emission spectra of terrestrial atmospheric
oxygen.
Solution The two isotopes of O2 are 16O16O and 16O 18O. Proceeding as in the above problem, we write
(18 − 16)
ρ −1 = = 0.031
4 × 16
Transition (Δν0)vib(cm−1)
and
V′→V″
⎡ ⎛ 1⎞ ⎛ 1⎞ ⎤
( ) vib = 0.031 ⎢1432.6 ⎜V ′′ + ⎟ − 1580.3 ⎜V ′ + ⎟ ⎥ cm −1
⎣ ⎝ 2 ⎠ ⎝ 2⎠ ⎦ 0→0 −2.229
0→1 42.12
The magnitude of (Δν0)vib for the asked bands of the series as deter-
mined from the above expression are listed below. 0→2 86.53
0→3 130.94
5.6.2 Rotational Isotope Effect

As already has been stated in Section 5.6, the rotational isotope effect is characterised by the difference in the
moments of intertia of the isotopic molecules. Assuming the Franck–Condon principle holds good, we write
1
B∝ (5.40)
μm
∗
Thus, in order to study this effect in homonuclear diatomics of the type X m X m2 ; X m X m 2 , the quantity (r−1)
defined by Eq. (5.38) is used here, i.e.
Δm
ρ2 − 1 =
2m1
The frequencies of the individual rotation lines in the electronic−vibrational spectra of diatomics are given by
lec vib + BV J ″(J ″ ) BV ′ J ′(J ′ + ) (Eq. 5.19)
Thus, the magnitude of shift of rotation line (Δν0)rot in electronic spectra of diatomics can be determined as
follows.
∗ ∗ 2 ∗
v rot − elec = v elec − vib + ρ BV J ′′ J ρ 2 BV∗ J ′(J
(J ′ + 1) (5.41)
Subtracting (5.19) from (5.41) and applying the selection rule ΔJ = ±l, i.e. J″−J′ = + 1 or J′ = J + 1, we obtain
(Δv0)rot = (Δv0)vib + r2 [(B*V″ − BV″)(J′ + 1) (J′ + 2) −(B*V′ − BV′) J′ (J′ + l)]
or
⎡ ⎛ BV∗ ′′ ⎞ ⎛ B∗ ⎞ ⎤
( ) ( ) vib + ρ 2 ( ′ ) ⎢ BV 1⎟ ( ′ ) BV ′ ⎜ V − 1⎟ J ′ ⎥ (5.42)
⎣ ⎝ BV ′′ ⎠ ⎝ BV ′ ⎠ ⎦
1
Since B ∝
μm
BV∗ ′′ BV∗ ′ μ m 1
∴ or = ∗ = 2
BV ′′ BV ′ μ m ρ
Consequently, Eq. (5.42) yields
(Δv0)rot = (Δv0)vib − (r2−1) (J′ + 1) [BV″(J′ + 2) − BV ′ J′] (5.43)
Thus, the superposition of the rotational isotope effect on the vibrational isotope effect imparts an additional
shift to the rotational lines in the spectrum. The rotational isotope shift, though measurable, is small compared to
vibrational isotope shift because the difference between the rotational terms is small compared to the vibrational
terms.
The isotope effect, both vibrational and rotational, has also been observed in polyatomics. However, an iso-
tope shift of only those lines which are associated with the vibrations or rotations involving an isotope has been
observed as expected theoretically also.
Problem 5.10: Using the required data from Problem 5.6 and Problem 5.8, calculate the rotational isomer
shifts for the rotational transitions J′ → J′′ : 0 → 1, 1 → 2, 2 → 3 with V′→V″: 0 → 0 transition in the electronic
spectrum of C2.
Solution The data acquired from Problem 5.6 are BV′′ = 1.7527cm−1 and BV′ = 1.6321cm−1 and from Problem
5.8 are (Δν0)vib(0→0) = 1.5 cm−1
Δm 1
Here ρ2 − 1 = = = 0.041
2m1 24
Substituting the value of (r2 − 1) and acquired data into
Eq. (5.43), we obtain J′ Transition (Δv0)rot (in cm–1)
(Δν0)rot= 1.5 − 0.041 (J′+ 1) [1.7527 (J′+2)−1.6321 J′] cm–1
0 0→1 1.35
The values of (Δν0)rot determined from the above expression
for the asked transitions are listed here. 1 1→2 1.20
The rotational isotope shift decreases as the-rotational quan- 2 2→3 1.04
tum number J′ increases.
5.7 NUCLEAR SPIN AND INTENSITY ALTERNATION IN ELECTRONIC

BAND STRUCTURE
Particles with odd half-integral spin quantum numbers are called fermions, e.g. protons and electrons. According
to the Pauli exclusion principle, they are symmetric with respect to interchange of pair of particles, i.e.
Ψelec exchange →Ψelec

(x, y, z) (x, y, z)
On the other hand, the particles with integral spin quantum numbers, e.g. a particle (s = 0), photons (s = 1) are
called bosons and they are anti-symmetric with respect to exchange of a pair of particles, i.e.
Ψphoton exchange
→Ψphoton
(x, y, z) (−x, −y, − z)
The total wave function of a homonuclear diatomic molecule with nonzero spin such as H2, can be represented as
Ψ ΨelecΨvibΨnucΨrot
y should be either symmetric or asymmetric with respect to exchange of nuclei, i.e.
Ψ exchange
→ Ψ
(x, y, z) (−x, − y, − z)
Ψ exchange
→ Ψ
(x, y, z) (x, y, z)
ψelec and ψvib are symmetric with the interchange of pair of nuclei in the homonuclear diatomic molecule since
ψelec is dependent of the coordinates of the nuclei and ψvib depends on the internuclear distance and is independent
of the nuclear coordinates.
Ψelec ⎫⎪ Ψelec ⎫⎪
⎬⎯ →
Interchange
⎬
Ψvib ⎪⎭ Ψvib ⎪⎭
(x, y, z) x, y, z)
Thus, the wave function Ψrot can either be symmetric or asymmetric with the interchange of pair of nuclei in
the homonuclear diatomic molecule and it is found that
Yrot → Yrot for even values of J, i.e. J =0, 2, 4
(x, y, z) (x, y, z)
Yrot → Yrot for odd values of J = 1, 3, 5.
(x, y, z) (–x, −y −z)
Let us now construct the nuclear spin wave functions for a homonuclear diatomic molecule with nonzero spin.
If atoms are designated as 1 and 2 then the nuclear spin wave functions are
and a(1) a(2), b(1) b(2), a(1) b(2), and b(1) a(2)
The wave functions a(l) a(2) and b(l) b(2) are symmetric while a(l) b(2) and b(l) a(2) are neither symmetric
nor asymmetric with respect to the interchange of pair of nuciei (i.e. 1 and 2) in the molecule. The linear combina-
tion of wave functions gives
(1)α (2) ⎫
⎪
α ( ) β ( 2) + α ( ) β (1)⎪⎬ symmetric nuclearr wave functionΨnnucs
⎪
β ( ) β ( 2) ⎪⎭
and a(1) b(2) − a(2) b(1) asymmetric nuclear wave function o Ψnnuc
a
Thus, in view of the above arguments, the rotation levels of a given electronic state are not only affected by the
spin of the electron but by the spin of the nucleus also. The total wave function Ψ is thus expressed as
Y = Yelec Ynuc
where Ynuc and Yelec are the nuclear-spin and electron-spin wave functions of the nucleus and electron respectively.
According to the Pauli exclusion principle, Y must be symmetric with respect to exchange of two nuclei if they
are bosons while asymmetric if the nuclei are fermions.
Let us first discuss the effect of two wave functions independent of each other on the rotational levels of a
given electronic state. So the product of Yelec and Yrot, i.e. Y ′ = Yelec Yrot will characterise the rotation term belong-
ing to the corresponding electronic state of the molecule. Each of the functions may be positive or negative
depending on whether the sign of the function changes with a change in sign of all coordinates of electron and
nuclei (Yelec) and with change of sign of coordinates of the nuclei abone, i.e. when the sense of rotation is reversed
(Yrot). It then turns out as shown in the Figure given below:
In case of Π −, Δ −, etc., terms, one of the components of L− type doubling of each rotational terms will be
positive and other negative.
ψelec
ψelecsym ψelecasym
When ψelecs is When ψelecs is When ψeleca is When ψeleca is

even and positive odd and negative even and negative odd and positive
(Σ +g), J values (Σ −u), J values (Σ −g), J values (Σ +u), J values
are even. are odd. are even. are odd.
Further, when we consider the effect of Ψnuc on the rotational levels of a given electronic state, the product of Ynuc
and Yrot will give a wave function Y ′′ = Ynuc Yrot which characterises the rotation levels of a given electronic state.
If the spin quantum number of each nuclei is I then the total number of linearly independent spin functions
corresponding to each nuclei = 2 I + 1, since MI = I, I −1, …. − I. Therefore, in a molecule the total number of
wave functions Ψnuc s
= (2I + 1) (2I + 1), the number of Ψnuc
s
with respect to exchange of nuclei = (2I + 1) (I + 1)
and Ψnuc = (2I + 1) I,
a
Accordingly,
( ) = (2I + 1)( I + 1) = I + 1
Statistical weight of symmetric spin states, N s Ψsnuc
Statiistical weight of asymmetric spin states N ( Ψ ) (2 I 1)( I )
a a
nuc I
Thus, if
(i) Ψ = Ψselec Ψsnuc , the symmetric rotational levels belonging to Ψselec is stronger for bosons
(ii) Ψ = Ψaelec Ψsnuc , the asymmetric rotational levels belonging to Ψaelec are stronger in case of fermions
Let us now extend the treatment to homonuclear diatomics in which both the nuclei are either fermions or
bosons.
In case of H2, where both the nuclei are fermions,
Ψ = Ψselec Ψanuc and Ψ = Ψaelec Ψsnuc
Accordingly, Ψarot corresponding to Ψaelec will be stronger as compared to Ψsrot since Ψsnuc has higher statistical
weight than that of Ψanuc . Thus, in Σ +g and Σ −u states, odd J levels are asymmetric and hence stronger while in Σ g−
and Σ +u states, even J levels are asymmetric and hence stronger. Thus, in case of H2 (fermion), we write,
Nuclear spin statistical weight of asymmetric levels, N a Ψanuc

u ( ) = Odd J values (or even J ) = I + 1 ,
s N
Nuclear spin statistical weight symmetrical of levels, s
(Ψ ) Even J values (or odd J ) I
s
nuc
1
+1 3
= 2 =
1 1
2
i.e. odd number rotational levels in ∑ +

g and ∑ −1
states are three times stronger than the even number rotational
u
even J 3
levels, while the reverse is true for ∑−
g and ∑
+
u states, i.e. = .
odd J 1
Further, in case of diatomics with integral spins, i.e. bosons, say N2, D2, and O2,
Ψ Ψselec Ψsnuc and Ψ = Ψaelec Ψanuc
Accordingly, Ψsrot corresponding to Ψselec will be stronger in comparison to Ψarot since Ψsnuc has higher statisti-
cal weight as compared to Ψanuc . Thus, in Σ g+ and Σ −u states, even J values are symmetric and hence stronger while
in Σ g− and Σ +u states, odd J values are symmetric and hence stronger.
Hence, in case of bosons, i.e. N2, D2, and O2, we write
ger )
Nuclear spin statistical weight of symmetrical levels (strong dd J )
Even J (or odd
=
l ( weaker )
Nuclear spin statistical weight of asymmetrical leve ls Odd J (or even J )
I +1 1+1 2
= = =
I 1 1
i.e. even-number rotational levels in Σ g+ and Σ u− states are two times stronger than odd-number rotational levels
odd J 2
while the reverse is true for Σ g− are Σ +u states, i.e. = .
even J 1
In case of O2, since I = 0 (MI = 0), so asymmetric rotational levels are of weaker intensity or even missing in
the spectrum.
The intensity alteration has also been observed in the rotational Raman spectra of bosons and fermions. The
intensity alteration in these spectrum can also be explained on the aforesaid arguments.
In case of H2, the ground state 1 Σ g+ is even and positive. So, even J values are symmetric, i.e. Ψ Ψselec Ψanuc
while odd J values are asymmetric i.e. Ψ Ψaelec Ψsnuc consequently.
odd J I + 1 3
= =
even J 1 1
The Stokes lines in the rotational Raman spectrum of H2 follows this pattern of intensity alternation.
In case of N2 and D2, I = 1, so the complete molecule wave function Ψ must be symmetric. Hence, symmetric
rotational lines corresponding to s
elec Ψsnuc will be stronger. The electronic state 1
∑ +
g is even and positive. So
even J levels will correspond to Ψselec and odd J levels will correspond to Ψaelec consequently
even J I + 1 1 + 1 2
= = =
odd J I 1 1
i.e. even J lines will be two times stronger than the odd J lines.
In case of oxygen, since I = 0 (MI = 0), so only symmetric nuclear spin function snuc is possible. Thus, the com-
plete molecular wave function Ψ Ψselec Ψsnuc is possible while Ψ Ψaelec Ψanuc is not possible. For O2, the ground
state is 3 ∑ g− . It is even and negative, hence even J values are asymmetric and odd J values are symmetric. Con-
sequently, rotational lines corresponding to even J values will be missing in the rotational Raman spectrum. The
absence of these lines reveals that the spin of each of the nuclei of oxygen molecule is zero.
Now we know that in addition to isotope shift, the spectra of isotopic molecules are also influenced by nuclear
spin of diatomics. The spins of diatomics with identical nuclei do not interact with one another or with the elec-
trons of the molecule. Consequently, during a radiative transition of the molecule, the nuclear spin symmetry of
the molecule does not change and the selection rule for space symmetry of rotational levels is s ↔ / a, i.e. mol-
ecules in even J state tend to remain in even J state while in odd J state, tend to be in odd J state. Thus, the rule
is absolutely true for molecules with zero spin, e.g. O2, C2, He2, etc. On the other hand, this rule is not true abso-
lutely in diatomics with nonzero spins, e.g. H2, N2, D2, etc. Such types of molecules, i.e. molecules with nonzero
spins, exist in two modifications, the characteristics of which are listed in Table 5.4. The statistical weight of the
para modification being larger than the ortho, the spectral lines corresponding to the former will be more intense
compared to the latter. However, such an alteration in intensity is missing in heterodiatomics. This difference is
particularly strongly expressed when nuclear spin is zero. In this case, the para modification does not exist and
hence in the absence of para term, every other line due to para form will be missing in the spectrum, e.g. 16 O 16O .
On the other hand, the spectra of 16 O 17O and 16 O 18O molecules show all lines permitted by the selection rule for
the rotational quantum number J. This is illustrated in Fig. 5.10 showing the 0′ → 1″ band of isotopic molecules
Table 5.4 Characteristic parameters for the ortho- and para- modifications of homonuclear
diatomics with nonzero spin.
Modification Nuclear spin J levels occupied Statistical weight
Ortho ↑↑ Odd Large
Para ↑↓ Even Small

8
10 4
12 10
6
2
8 6
14 14 12 4
2
20 16 16 0
20
18 16
3 11
11
11 11 9 3 56 8 8 10 12 12
1313 8 43 3 45
12 12
10 10
8 7 7 6 6
5 54 2 2 1
1 2 2 4 7 6 7 9 910 1313
14 14 9 1 0
1
16
Fig. 5.10 Band (0′ l″) of isotopic molecules O16O and 16
O18O in the absorption spectrum of the terrestrial atmosphere. The lines of the
16
O18O molecule are denoted by arrows. Intense doublets represent the P-branch of the band of 16
O18O.
16
O16 O and 16
O18O in the absorption spectrum of terrestrial atmosphere (light source is sun). The two sets of dou-
blets with different intensities are because of isotopic molecules 16 O16 Oand 16 O18O. The intensity difference is due
to different relative abundance of the isotopic species.
The intense doublets with even values of J correspond to the P-branch of 0′ 1″ band of 16 O16 O molecule. The
doublet structure of the lines is due to the interaction of nuclear spin with the rotation of the molecule. This struc-
ture results provided the nonpermissible 2 ∑ g 1 ∑ +g transition is considered and the selection rule ΔJ = 0, ±1
rather than Δ J = 0, ±1, ±2 is valid since the latter rule results for quadrupole radiation. This shows that atmo-
spheric bands of O2 are associated with magnetic dipole radiation but not with quadrupole radiation. The weak
lines, both with even and with odd values of J in the second set of doublets, correspond to the 16 O18O molecule.
The asymmetric distribution of intensity in the bands of both the isotopes and also the presence of weak interme-
diate lines are because of the superposition of O2 spectrum by the spectral lines of iron and water vapour.
Further, it is found that the ratio of the statistical weights for the ortho and para modifications and hence the
ratio of their line intensities in the spectrum is given by
I ortho I + 1
= (5.44)
I para I
Thus, from the intensity measurements, it is possible to determine the nuclear spin and isotopic abundance in
the molecule. The intensity ratio and nuclear spin of some homonuclear diatomics that exhibit ortho- and para-
modification are listed in Table 5.5.
Table 5.5 Nuclear spin and ratio of line intensities of the ortho- and
para- modifications in the spectra of homonuclear diatomics.
Molecule Intensity ratio Nuclear spin
1
H2, P2 3:1
2
D2, 14
N2 2:1 1
7 3
Li 2 1.67:1
2
35 127
Cl 2 I2 1.4:1 5/2
He 2 C 2 ,⎫
12
16 32
⎪
O2, S2 ⎬ ∞ 0
80 ⎪
Se 2 ⎭
Problem 5.11: The spin of a hydrogen molecule is 1 . Determine the per cent abundance of ortho- and para-
modification of H2. 2
1
I ortho 1 + 2 3
= =
I para 1 1
2
3 × 100
Amount of ortho modification = = 75 per cent
4
1 × 100
And amount of para modification = = 25 per cent
4
Exercise: Draw the J levels for the following cases: (i) Σ + ( Λ = ) and Π( Λ ) , (ii) Σ +g Σ u− and Σ g− Σ u+
(bosons), and (iii) Σ +g Σ u− and Σ g− Σ u+ (fermions).
5.8 CONTINUOUS ABSORPTION AND EMISSION SPECTRA

5.8.1 Continuous Absorption Spectra
Molecular absorption electronic spectra of diatomic gases exhibit marked discrete linear band structure and in addi-
tion, a continuum is also observed particularly in polyatomics. Broadening of bands takes place at high temperature
2v obs 2RT ln 2 v 2RT
and pressure.The former is called Doppler broadening and is expressed by Γ D = = 1 67 obs
c M c M
1 8RT
and the latter is known as collision or impact broadening and is given by Γ p = Here, l0 is the mean
π l0 M
free path, M is the molecular mass, v obs is the observed frequency and T is the absolute temperature. Thus, it is
possible to determine T from the former expression if GD and νobs are known. For example, if 57Fe line emitted
from the surface of the sun is observed at 677.4 nm and has a bandwidth of 5.3 pm, the temperature of the sun’s
surface will be 6.3 × 107 K. It follows that when the pressure or temperature is increased, the separate lines in the
rotational fine structure begin to broaden and at a certain stage, the width of rotational lines exceeds the interval
between adjacent lines in the rotational fine structure. Consequently, the rotational fine structure starts disappear-
ing and ultimately disappears completely. At sufficiently high pressure, the width of the separate rotational lines
even exceeds the interval between the bands in the spectrum and a complete continuum results. Such a conver-
sion of discrete linear band structure into a continuum occurs at high temperatures and pressures. However, a
continuum is also observed at low temperatures and pressures. Thus, it follows that a continuum is not due to
external parameters such as temperature and pressure but due to some internal causes also. One of the causes for
continuum is that the rotational and vibrational motion is not quantised at least in one of the combining electronic
states of the molecule in an electronic transition. Thus, a complete continuous absorption spectrum results when
an electronic transition occurs from a ground electronic state to an unstable upper electronic state (i.e. a state
without discrete rotational and vibrational levels) of the molecule as shown in Fig. 5.11(a). The molecule in the
unstable state dissociates into atoms and the atoms fly away with kinetic energy. Since this kinetic energy of atoms
is not quantised, a complete continuum results in the electronic absorption spectrum of the molecule. The lower
limit of continuum will have energy 0′ = Eex, where Eex is the total excitation energy of the products and D 0′ is
the dissociation energy of the molecule in the ground state. The continuum also appears when transition takes
place from a ground electronic state to an upper stable electronic state of the molecule as shown in Fig. 5.11(b).
When the molecule crosses over from ground curve to the point ‘X ’ in the upper curve by electronic excitation,
it dissociates into separate atoms that fly apart with kinetic energy. Since the kinetic motion is not quantised,
a complete continuum results. The lower limit of the continuum is Eex + 0″ = D 0′ (asymptote dashed line).
Transitions to points below the asymptote, i.e. dissociation levels result into a successively deconverging discrete
X A+B
Ekinetic V ′′ = 3 D′′0
V ′′ = 2
A+B
V ′′ = 1
U (r) V ′′ = 0 AB
U (r)
Eex Eex
A+B A+B
V′ = 3 D′0 D ′0
V′ = 2
V′ = 1
V′ = 0 AB AB
r (Å) r (Å)
(a) (b)
−
νconv
(c)
Fig. 5.11 Schematic diagram showing the photodissociation by excitation of a diatomic molecule from the ground stable state into (a) an
upper unstable state, (b) an upper stable state, and (c) an absorption spectrum containing discrete and continuous portions.
bands. Such a spectrum in which contiunnum adjoins the convergence limit of discrete bands (V″ progression) has
been observed in Cl2, Br2, I2, O2, C1O2, etc., and forms the basis of band convergence method for determination
of dissociation energy of diatomics. [(Fig. 5.11(c)].
(i) When re′′ > re′, there is one discrete region. This is due to the reason that the V′ progression breaks off before
the convergence limit is attained since the probability of transition of molecule to the region above the dissocia-
tion level is negligibly small. (ii) When re′′ (upper curve is repulsion curve and there is no minima in the unstable
upper state) >> re′ all transitions from all the V′ levels of ground state to the region above the dissociation level
have maximum probability and complete continuum results since this region behaves like a repulsion curve, e.g.
diatomics such HX, MeX, where X = Cl, Br, I, F and polyatomics such as N2O, C12O, COS, XCN, H2O, H2S,
H2O2, SO3, saturated hydrocarbons, alcohols, ethers, acids, etc.
5.8.2 Continuous Emission Spectra

The continuous emission spectra are the inversion of continuous absorption spectra of diatomics (A2). They origi-
nate from the recombination of atoms, i.e. A + A*→ A2+ hv, e.g. temperature luminescence of halogens. The H2
molecule when bombarded with electrons also exhibits continuous emission spectra but not when exposed to light
(fluorescence). Continuous emission spectrum is due to the transition from the stable triplet state 1 s s 2s σ 3 ∑ g
+
2 p σ 3 ∑ u and in this transition, the hydrogen molecule dissociates

+
(excited by electron) to the unstable state 1s 2p
into two normal hydrogen atoms that fly away with unquantised kinetic energy. The absence of continuous emis-
∑
1 +
sion spectrum when hydrogen is excited by light (fluorescence) is due to the fact that the transition g
(ground
stable of H2) →3 ∑ g is forbidden. However, when excitation is by a free electron, the exchange of electron takes
+
place according to the scheme:
1 2 3 1 3 2
↑ + ↓ ↑ → ↑ ↑ + ↓
e H2 (1 ) H2 (3 ) e
2 σ 3 ∑ g is excited and a continuous emission spectrum in the 500−160 nm range

+
Consequently, the level 1sσ 2s
results due to the transition,
1sσ 22s σ 3 ∑ g → 1s 2 p σ 3 ∑ u + hv
+ +
2p
The origin of continuous emission spectra of diatomics such as Xe2, He2, etc., is similar to that of the continu-
ous emission spectrum of H2. Some diatomics, namely Hg2, Cd2, Zn2, exhibit continuous spectra with alternating
maxima and minima of intensity, both in the absorption and emission mode. Vapours of MX (where M = alkali
metals, X = halogens) also exhibit such a peculiar continuous absorption spectra. The continuous spectra of MX
diatomics could be interpreted on the basis of transition from a ground stable state to the upper unstable state of
the molecule represented by a horizontal line. All the transitions from different vibrational levels of the ground
state to the upper state will have frequencies
2
⎛ 1⎞ ⎛ 1⎞
v v elec-vib = v 0 w e ⎜V ′ w e xe V ′ + ⎟ (5.45)
⎝ 2⎠ ⎝ 2⎠
The upper state being unstable, the transition to it results in the dissociation of molecules into separate atoms
that fly apart with unquantised kinetic energy and a broad maximum results. According to Eq. (5.45),the maxima
should coverage to a certain limit lying in the region of long wavelengths (V″ progressions). The distance between
two maxima should approximately be equal to we. Note that if the repulsive curve is not horizontal but a steep
one, the separate maxima will merge into a continuum with uniform distribution of intensity, e.g. NaCl, NaBr
and KC1.
5.9 PREDISSOCIATION/DIFFUSE SPECTRA

The phenomenon of predissociation occurs due to the transfer of a molecule from a stable excited electronic state
to an unstable electronic state of the molecule in an electronic absorption transition; the unstable state is the cause
of predissociation. This process causes anomalous broadening of rotational lines which, in turn, results in diffuse
bands within a progression in the electronic absorption spectra of gases. The diffuseness may build up slowly with
diminishing wavelength in some cases and suddenly in other cases.
5.9.1 Mechanism of Predissociation in Diatomics

The electronic states of molecules are characterised by Morse poten-
tial energy curves. For predissociation to take place, the Morse curves
corresponding to the stable excited electron state and unstable state of
the molecule should intersect as shown in Fig. 5.12. Thus, the transi- A + B∗
tion of a molecule from excited curve to an unstable curve will be the Excited
cause of pre-dissociation. Note that the repulsion curve of the excited
C
state has no minimum or the minimum lies at a much greater inter-
U (r)
Unstable
nuclear distance than that of the ground state, i.e. r″ >> r'.
A+B
According to the Franck–Condon principle which holds good
for radiationless transitions also, the predissociation transition from
excited curve to the repulsion curve is possible only near the points of Ground
intersection of the curves, since the vibrating molecule spends most
of its time at the extreme ends of its vibrational motion, especially
when the vibrational quantum number is large. The point/points of r ′AB (Å)
intersection must lie below the initial vibrational level of excitation Fig. 5.12 Schematic diagram showing the pre-
and the lifetime of the excited molecule in vibrational states below and dissociation of a diatomic molecule.
above the point of intersection is of the order of 10−8 s but the lifetime
ofthe excited molecule which is τ = a τ1, where τ1 is the lifetime of
the ordinary excited state, 1 − a is the probability of predissociation, in the vibrational levels immediately near
the point of intersection, i.e. the region of predissociation decreases and the probability of transfer of molecule
from stable to unstable state and hence predissociation increases. The transfer rate being faster than the rota-
tional period (∼ 10−11s) but slower than the vibrational period (10−13s), the predissociation will occur before the
⎛ h ⎞
molecule rotates. Consequently, Γ rot Γ rot = ⎟ width of rotational lines will increase and the fine structure of
⎝ 2πτ ⎠
Region of Predissociation
V′ → V′′ 0 → 6 0→7 0→8 0→9 0 → 10 0 → 11

− −1
ν (cm )
Fig. 5.13 Schematic diagram showing the predissociation spectrum. Rotation fine structure is seen both on
the left-hand and right-hand sides of the predissociation region.
bands will disappear and the bands become diffuse but the vibrational coarse structure will not be affected. How-
ever, if the rate of transfer is faster than the vibrational period, the vibrational coarse structure will also be absent
and a complete continuum will occur in the electronic absorption spectrum of the molecule. On the other hand,
when the excited molecule is present in the vibrational levels away from the predissociation region, i.e. point of
intersection of curves, the number of vibrations and rotations in the lifespan of the excited molecule will be 105
and 103 respectively as expected; since when excited molecule is present in these levels, it spends insufficient time
near the point of intersection for appreciable dissociation to take place and a normal rotational fine structure will
be observed for vibrational transitions both on the right- and left-hand sides of the predissociation region. The
transitions below the point of intersection appear on the higher wavelength side of the predissociation region and
lower wavelength side above the point of intersection as shown in Fig. 5.13.
Predissociation in diatomics also results when diserete vibrational or rotational levels of a certain upper elec-
tronic state of the molecule are overlapped by the dissociation continuum of lower electronic state, and radia-
tionless transition occurs from discrete rotational or vibrational levels to dissociation continuum of the lower
electronic state.
Further, the diffuseness of bands in a progression may set in suddenly or grow gradually with diminishing
wavelength. The gradual increase in diffuseness of bands with diminishing wavelength in the absorption spectra
of some diatomics, e.g. S2 may be due to tunnelling effect, the probability of which increases as the distance of
the initial vibrational level from the point of intersection of Morse curves increases. In such cases, no sharp pre-
dissociation limit, i.e. definite wavelength, is observed. However, when the diffuseness of bands in a progression
sets in suddenly, a sharp predissociation limit (i.e. with a certain definite wavelength) is observed and the predis-
sociation in such cases is termed sharp predissociation. The predissociation transitions take place only at Δ J = 0,
when ΔL= 0, ±1, and ΔS = 0. The positive levels combine with positive levels and negative levels combine with
negative levels. Symbolically, + ↔ +; − ↔ −; + � −. In case of diatomics having the same nuclei, the symmetric
levels combine with symmetric levels and asymmetric levels with asymmetric levels, i.e. s ↔ s; a ↔ a; s � a.
When transition from the stable curve to the repulsive curve obeys the selection rules, spontaneous predissocia-
tion, i.e. spontaneous decay of the excited molecule is observed. Forbidden transitions are also possible under the
influence of an external field and predissociation in such cases is called induced predissociation, e.g. predissocia-
tion induced by a magnetic field (e.g, in I2) or induced by molecular collisions (e.g. in I2, Br2 , N2, NO, S2, Se2, Te2).
The predissociation caused by molecular collisions is useful in understanding the mechanism of photochemical
reactions.
5.9.2 Predissociation in Polyatomic Molecules

Predissociation in polyatomics occurs due to overlapping of higher vibrational levels of an electronic state by
dissociation continuum adjoined to a lower dissociation limit of the same electronic state and the radiationless
splitting of a particular atom or group of atoms takes place. This kind of predissociation is called predissociation
by vibration.
The polyatomics dissociation differs from diatomics in that the interaction between atoms in polyatomics is
characterised by potential-energy surfaces rather than the potential-energy curves. Thus, the region of predisso-
ciation in polyatomics is determined by the point of intersection of surfaces. However, the location of molecule in
the predissociation region is not always a sufficient condition for predissociation. Moreover, the atoms must be in
the phase of the motion that can cause the predissociation decomposition of the molecule. Both these conditions
are to be satisfied simultaneously by the molecule for sharp predissociation which, of course, are not very fre-
quently satisfied. However, sharp predissociation is also observed in the ultraviolet absorption spectrum of some
polyatomics, e.g. NO2. The dissociation energy of NO2 (NO2 → NO + O) as determined from predissociation limit
is 70.8 kcal mole−1. The value obtained from thermochemical data is 71.8 kcal mole−1.
Predissociation in Emission Spectra The phenomenon of predissociation in emission spectra differs

from predissociation in absorption spectra in that in some cases the bands are totally missing in the predis-
sociation region and there is abrupt breaking off of the rotational lines in several bands. In the former, the
predissociation limit is characterised by the abrupt breaking in the emission spectrum, e.g. S2, NO, N2, etc.,
for certain values of V and J determined for each series of bands (V ) and for each band (J ). Note that in some
cases, e.g. S2, the bands absent in the predissociation region at low pressure appear at high pressure. This
happens only in those cases where the mean lifetime of predissociating molecule becomes comparable with
the time interval between two quenching collisions of an excited molecule with other molecules. The abrupt
breaking off the rotational structure in the 0′0″ emission band of CaH λ = 3533.6 Å. The sudden breaking
off the rotational band structure observed in the emission spectrum of HgH and AlH is due to the splitting
of the molecule under the interaction of the centrifugal force produced by the rotation of the molecule. This
type of predissociation is termed predissociation by rotation. In predissociation by rotation, the higher rota-
tional levels of a given vibrational level of a diatomic molecule are overlapped by the dissociation continuum
belonging to the same electronic state, due to which radiationless decomposition of the molecule with no
change of electronic state occurs.
5.10 DISSOCIATION ENERGY AND ITS DETERMINATION

The energy required to split a diatomic molecule into separate atoms comprising it is called the dissociation
energy or heat of dissociation. It is designated as D and at absolute zero as D0/D0. The heats of dissociation of
diatomics are used for calculating the energies of splitting of complex molecules into atoms (heats of atomisation)
which are of prime importance in determining the bond energies. The heat of dissociation of AB may be obtained
from the heat of dissociation of A2 and B2 and the heat of formation of AB from the formula.
1
AB ( D A 2 + D B 2 ) ΔH
H AB (5.46)
2
where ΔHAB is the heat of formation of AB, e.g. D N0 2 = 225.09 kcal mole−1, DO0 2 = 117.96 kcal mole−1, and
ΔHNO= 19.22 kcal mole−1.
1
Thus, 0
D NO 225.09 + 117.96) − 19.22 = 152.3 kcal mole −1
= ((225
2
Heat of dissociation can be measured experimentally by thermal and spectroscopic methods; the former gives
an approximate while the latter gives an accurate value of dissociation energy. Thermal methods are based on the
P2
measurement of thermodynamic equilibrium (i.e., X2 � 2X, KP = X ) as a function of temperature, e.g., Cl2, Br2, I2.
PX 2
0 −1
DI 2 determined by this method is 35.514 kcal mole . The other method based on thermodynamic equilibrium
is the shock-wave method and is used for measurement of large values of heats of dissociation. This method is
based on the relation between the rate of propagation of a shock wave through a gas and its state, i.e. temperature,
density and pressure. The heat of dissociation measured by this method at absolute zero in case of N2 is 225.09
kcal mole−1. The other thermal methods, based on the measurement of rates of chemical processes at different
temperatures, are called kinetic thermal methods. One of the examples of such methods is the pyrolytic method
which deals with the rate of decomposition of substances.
The dissociation energy of a bond at a certain single temperature is given by
1013
D R T ln (5.47)
k
where k is the specific rate constant. Dissociation energy of the C−I bond in various organic compounds was first
of all measured by this method.
The spectroscopic methods differ from thermal methods in that the dissociation is achieved through electronic
excitation of the molecule and its dissociation energy can be measured both in the ground and excited states. The
two main spectroscopic methods employed to determine the dissociation energy of diatomics are band conver-
gence and pre-dissociation limit methods. The former is most precise and gives accurate values of dissociation
energy of diatomics.
5.10.1 Band-Convergence Method

Basically, the precision of the band-convergence method is based on the accurate determination of a vibration
level of the excited molecule in an electronic transition in which the molecule breaks up into separate atoms with
almost zero kinetic energy. This level is known as the dissociation vibrational level of the molecule in an elec-
tronic transition. Below this level, discrete but successively deconverging, bands appear while in the upper adjoin-
ing level, the molecule splits up into separate atoms and fly away with kinetic energy and a continuum results, the
lower limit of which is the dissociation level. The continuum in the electronic absorption spectrum may appear
due to the electronic transition from the ground electronic state either to stable or to an unstable upper electronic
state of the molecule as shown in Figs 5.11 (b) and (a). In the former, the molecule splits up into separate atoms
and flys away with kinetic energy and only a continuum results while in the latter, a convergent discrete- band
structure followed by a continuum appears. This has been schematically shown at the bottom of Figs 5.11(a) and
(b), i.e. Fig. 5.11(c). This method is based on the fact that the maximum vibrational energy of a molecule equals
its dissociation energy, Only two vibrational levels are sufficient to prove this fact. We know that
2
⎛ 1⎞ ⎛ 1⎞
EV hcw
h e V hcw e x e V + ⎟
⎝ 2⎠ ⎝ 2⎠
2
⎛ 3⎞ ⎛ 3⎞
EV h e V
hcw hcw e x e V + ⎟
⎝ 2⎠ ⎝ 2⎠
Thus EV EV = hcw e hcw e x e (V + 1)
hcw
When (EV+1 − EV) → 0, V → V max
1 − 2x e hcw e w
Thus, V max = and De EV ,max = J = e , cm −1
2x e 4x e 4x e
This convergence limit of bands in a progression (V′) in the electronic absorption spectrum is related to the
excitation energy Eex of the products and heat of dissociation of the excited molecule D0″ by the relation
v conv = v elec + D 0′′ = E + D 0′′, cm −1
or
D 0′′ = v conv − E ex , cm −1 (5.48)
Thus, D0″ can be determined if Eex can be measured accurately from atomic spectroscopy. The convergent
limit of bands is observed only in rare cases, e.g. diatomics Cl2, Br2, I2, O2, and ClO2, etc., otherwise it is deter-
mined by the Birge–Sponer method of extrapolation. The band convergence method and Birge–Sponer methods
have already been discussed in detail in Chapter 3 on infrared spectroscopy. Note that the extrapolation method
is applicable in case of vibrational structure of the spectrum and linear extrapolation always overestimates the
dissociation energy. This method is applicable for both the ground and excited states of the molecule. When no
vibrational structure is observed and only a continuum results as in Fig. 5.11 (a), the lower limit of the continuum
is taken as the convergent limit of the spectrum, e.g. HX (X = C1, Br, I, F) and a number of polyatomics: N2O;
C12O, COS, XCN, H2O, H2S, H2O2, SO3, saturated hydrocarbons, alcohols, ethers, acids, etc.
Thus, v conv = D 0′′ = E = D 0′. Now the problem arises how to determine the dissociation energy of the molecule
in the ground state. It can be determined provided the origin of the terms corresponding to the ground and excited
states of the molecule are known. A number of cases are possible and only three of them are discussed here:
(i) The ground and excited terms of the AB molecule originate from ground terms of atoms A and B. The
excited molecule in such a case splits up into two atoms in the ground state as shown in Fig. 5.14(a). The
dissociation energy of the molecule in the ground state D0′ is then given by
D0′ = Eex + D0″, cm−1 (5.49)
(ii) The ground term of the AB molecule originates from the ground terms of atoms A and B while the excited term
of AB originates from the ground term of the atom A and excited term of the atom B. In such a case, the dis-
sociation of the excited molecule will result into one normal atom A and other excited B* while the dissociation
of AB in the ground state will give rise to normal atoms A and B as shown in Fig. 5.14 (b). In this case,
Eex+ D0″ = D0′ + Eex(B*)
or D0′ = Eex + D0″ − Eex(B*)

Using Eq. (5.48),
D 0′ = ν con − E ex ( B * ), cm −1 (5.50)
Here, Eex (B*) is the excitation energy of B*.

Thus, in order to determine the dissociation energy of the molecule in the ground state in such case, the
excitation energy of one of the atoms must be known, from atomic spectroscopy. This case is encountered
in Cl2, Br2, I2, O2, etc.
(iii) The ground term of the AB molecule originates from the ground term of the atom A and excited term of
the atom B while the excited term of AB originates from the normal terms of atoms A and B. In such a case
the excited molecule will dissociate to yield normal atoms A and B while the ground state molecule AB*
will split into one normal atom A and another excited atom B* as shown in Fig. 5.14(c). In this case,
A+B A + B∗
V ′′ = 3 D ′′0 V ′′ = 3 D ′′
0
V ′′ = 2 V ′′ = 2
V ′′ = 1 V ′′ = 1
AB AB ∗
V ′′ = 0 V ′′ = 0
Ecx(B ∗)
U (r)
U (r)
Eex Eex
A+B A+B
V′ = 3 D ′0 V′ = 3
V′ = 2 D′0
V′ = 2
V′ = 1 V′ = 1
V′ = 0 AB AB
V′=0
rAB(Å) rAB(A)
(a) (b)
A+B
V ′′ = 3 D ′′0
V ′′ = 2
V ′′ = 1
AB
V ′′ = 0
Eex(B ∗)
U (r)
Eex
A + B∗
V′ = 3 D ′0
V′ = 2
V′ = 1
AB ∗
V′ = 0
rAB(Å)
(c)
Fig. 5.14 Different cases of the photodissociation of diatomic molecules.
D 0′ − Eex (B*) = D0″+ Eex
or D 0′ = D 0′′ + E ex + E ex ( B ∗ ) = v conv + E ex ( B ∗ ) (5.51)
Thus, D 0′ can be determined provided Eex (B*) is known from atomic spectroscopy.
Further, it is to be noted that the dissociation energy of diatomics can also be measured from the band progres-
sions (V″) in the emission spectrum. In case of diatomics, the convergence limit directly gives the dissociation
energy of the molecule in the ground state, Do′ provided the ground term of the molecule originates from the
normal terms of constituent atoms, i.e.
D 0′ = v elec − v conv , cm −1 (5.52)
e.g. D0′ for hydrogen has been found to be 103.24 kcal mole−1.
Problem 5.12: In the electronic absorption of iodine vapour, the vibrational bands converge at 4995 Å. The
excited state of iodine molecule dissociates into one normal atom (2P3/2) and other excited atom (2P1/2). The energy
separation between the doublet (2P1/2 − 2P3/2) as determined from atomic spectroscopy is 22 kcal g atom−1. Calcu-
late the dissociation energy of iodine molecule.
Solution According to band convergence method, the dissociation energy of a diatomic molecule is given by
D 0′′ = v conv − E ex , cm −1 (Eq. 5.48)
In case of I2, Eex = 2P1/2 − 2P3/2 = 22 kcal g atom−1 = 92.4 kJ g atom−1

Substituting the values of Eex and v conv in Eq. (5.48), we get
6.62 × 10 −34
34
(J ) × 3 × 1010 (cm/s) × 6.023 × 1023 − 92.4
D 0′′ =
4995 × 10 −8 (cm )
( )
= (239.47−92.4) kJ mole−1
= 147 kJ mole−1 = 35.01 kcal mole−1
The literature value is 36.46 kcal mole−1.
Problem 5.13: The dissociation energy of O2 determined by the band-convergence method at 0 K has been
found to be 117.96 kcal mole−1. The ground state ( 3 g ) of the oxygen molecule originates from 3P ground terms
3
of the oxygen atom while that of the excited state ( ) originates from the lD excited state of the oxygen atom
and 3P ground state of the oxygen atom. The energy difference between the doublet, as determined from atomic
spectroscopy, is 45 kcal mole−1. Calculate the convergence limit of progression (V′) in the electronic absorption
spectrum of oxygen gas in Å, and cm−1.
Solution Given D0″ = 117.96 kcal mole−1

Eex = 1D − 3P = 45 kcal mole−1
Substituting these values into Eq., (5.48) and rearranging, we get
(117.96 + 45) × 4.2 × 103 (J mole )

v conv =
6.627 × 10 −34 (J s ) × (6.023 × 10
1 23 ) 3 1010 (cm/s)
=
684.43 × 103 ( ) = 57226.58 cm−1
11.96 ( )
or l =1747 Å
The experimentally observed value is 1758 Å.
Problem 5.14: The vibrational energy levels for the first eleven vibrational quantum numbers in the ground
state as deduced from the gross structure of a band in the ultraviolet spectrum of CO can be represented by the
expression
2
⎛ 3⎞ ⎛ 3⎞ −1
v 2170 21 V ′ + ⎟⎠ − 13.46 ⎜⎝V ′ + ⎟⎠ ,cm
⎝ 2 2
Calculate (a) the dissociation energy De′ of the molecule in kJ mole−1 and kcal mole−1, and (b) force constant for
the fundamental mode of vibrational of CO.
Solution (a) The dissociation energy can be determined by the Birge–Sponer method reported in Chapter 3 on
infrared spectroscopy. Accordingly,
2
⎛ 1⎞ ⎛ 1⎞
vV 2170 21 V ′ + ⎟ − 13.46 ⎜⎝V ′ + ⎟⎠
⎝ 2⎠ 2
2
⎛ 3⎞ ⎛ 3⎞
vV +1 21 0 21 V ′ + ⎟ − 13.46 ⎜⎝V ′ + ⎟⎠
⎝ 2⎠ 2
Thus, ΔG ′ = v V′′+1 − v V′
Generate the values of ΔG ′ = 0, 1, 2, 3, 4, 5,… 8, 9 and 10 and plot a graph between ΔG' and V′. From the
graph, read off the value of the vibrational quantum number corresponding to ΔG' = 0. Note that V′ = V′max for
ΔG' = 0. Our value of V′max has been found to be 79. Substituting the value of V′max = 79 into the given equation,
we obtain
2
⎛ 1⎞ ⎛ 1⎞
vV 3.46 79 + ⎟ , cm −1
De′ = 2170.21 79 + ⎟ − 13
⎝ 2⎠ ⎝ 2⎠
= 2170.21 × 79.5 −13.46 × (79.5)2, cm−1 = 87461.13cm−1 = 1045.86 k J mole−1

= 249 kcal mole−1 (literature value is 257 kcal mole −1).
(b) In order to determine the force constant, we must first determine the frequency of the fundamental mode
of vibration of CO, i.e. v ′ .
From Eq. (3.57),
v ′ = w e′ ( − x e′ ), cm −1
Substituting the values of we and we xe into the given equation, we obtain
v ′ =2170.21−13.46=2156.75 cm−1
16 × 12 192
For CO, the reduced mass μ m′ = amu = amu. Substituting the values of v ′ and m′m and into
Eq. (3.19), we get 16 + 12 28
2
⎡ 2156.75 ⎤ 192
k′= ⎢ ⎥ × 28 millidynes /Å = 18.78 millidynes/Å
⎣ 1303.16 ⎦
Problem 5.15: The vibrational energy levels in the excited state, as deduced from the coarse structure of a
band in the electronic absorption spectrum of CO, fits into the expression
⎛ 1⎞ ⎛ 1⎞ −1
v ′′ = 1 1 61 ⎜V ′′ + ⎟ − 17.25 ⎜⎝V ′′ + ⎟⎠ , cm
⎝ 2⎠ 2
Calculate (a) the dissociation energy De of the molecule in kJ mole−1 and kcal mole-1, (b) force constant for the
fundamental mode of vibration of CO, and (c) the energy corresponding to the electronic transition if the 0–0 band
is observed at 64746.5 cm−1. Use the required data from Problem 5.15 also.
Solution Proceeding as in Problem 5.15, we get;

(a) ΔG″ = 1515.61 − 34.50 (V″+ 1);
′′ x = 42 and De″ = 33255.61 cm−1 = 397.670 kJ mole−1 = 94.681 kcal mole−1

Vma
(b) ν ′′ = (w ″− w ″ x ″) cm−1 = (1515.61 −17.25) cm−1 = 1498.36cm−1

e e e
2
⎛ 1498.36 ⎞ 192
Therefore, k ′′ = ⎜ × millidynes/Å = 9.06 millidynes/Å
⎝ 1303.16 ⎟⎠ 28
(c) In order to determine the energy corresponding to the electronic transition, we need the values of we′ and
we′ xe′. From Problem 5.15, we write we′ =2170.21 cm−1 and w′e x′e = 13.46 cm−1.
Substituting these values together with we″ and we″ xg″ and ν 00 into Eq. (5.15), we obtain
⎛ 1515.61 17.25 ⎞ ⎛ 2170.21 13.46 ⎞ −1

v elec = 64746.5 − ⎜ − ⎟ + − ⎟ cm
⎝ 2 4 ⎠ ⎝ 2 4 ⎠
= (64746.5 − 753.49 + 1081.74) cm−1
= 65074.75 cm−1.
Problem 5.16: The analysis of the vibrational coarse−rotational fine- structure of a band in the electronic
spectrum of N +2 corresponding to the transition 2 ∑ u+ → ∑ g+ fits into the following expressions:

2
⎡ ⎛ 1⎞ ⎛ 1⎞ ⎤
2
v elec − vib 2 461 ⎢ 2419 84 V ′′ + ⎟ − 23.19 ⎜V ′′ + ⎟ ⎥

⎢⎣ ⎝ 2⎠ ⎝ 2 ⎠ ⎥⎦
⎡ ⎛ 1⎞ ⎛ 1⎞ ⎤
2
− ⎢ 2207.19 ⎜V ′ + ⎟ − 16.136 ⎜V ′ + ⎟ ⎥ cm −1 (i)

⎢⎣ ⎝ 2⎠ ⎝ 2 ⎠ ⎥⎦
v elec −rot
rot ⎡⎣v eelec
lec vib + 4.015 m + 0.151 m 2 ⎤⎦ cm −1 (ii)
Calculate (a) the zero-point energies of the ground and excited states of the molecule, (b) the 0−0 band, (c) the
frequency and force constant for the fundamental modes of vibration in the lower and upper electronic states of
the molecule, (d) the bond lengths of N +2 in the ground and excited states, (e) the value of the band head and its
position, (f) separation between band head and zero line and the direction of shading, and (g) dissociation energies
in the two electronic states of the molecule.
Solution (a) Zero-point energy of the upper state
⎛ w ′′ w ′′ x ′′⎞ ⎛ 2419.84 23.19 ⎞

⎟⎠ cm = 1204.12 cm
−1 −1
=⎜ e − e e⎟ = −
⎝ 2 4 ⎠ ⎝ 2 4
Zero-point energy of the lower state
⎛ w ′ w ′ x ′ ⎞ ⎛ 2207.19 16.136 ⎞ −1
=⎜ e − e e⎟ = − ⎟ cm = 1099.56 cm−1
⎝ 2 4 ⎠ ⎝ 2 4 ⎠
(b) 0−0 band is determined from the given Eq. (i)
⎡ ⎛ 2419.84 23.19 ⎞ ⎛ 2207.19 16.136 ⎞ ⎤

v elec-vib(V ′ ′ 0)
0 →V ′′= = ⎢ 25461.5 + ⎜ − ⎟ −⎜ − ⎟ ⎥ cm −1
⎣ ⎝ 2 2 ⎠ ⎝ 2 2 ⎠⎦
= (25461.5 + 1204.12 − 1099.56) cm−1 = 25566.06 cm−1
(c) We know that v ′ = w′e − w′e x′e

Thus, v ′ = (2207.19 − 16.136) cm−1 = 2191.05 cm−1
2
⎡ 2191.05(cm −1 ) ⎤ 196
and k′ ⎢ ⎥ × ( ) = 19.78 y s / A°
⎣ 1303.16 ⎦ 28
Similarly,
v ′′ = (2419.84 − 23.19) cm−1 = 2396.65 cm−1 and
( ) ⎤⎥
2
⎡ 2396.65 196
k ′′ = ⎢ × (amu) = 23.67 millidynes/Å.
⎢⎣ 1303.16 ⎥⎦ 28
(d) Comparing Eq. (ii) with Eq. (5.23), i.e.
rot = v eelec
v elec −rot lec vib + ( BV BV ′ ) m (BV ′′ − BV ) m 2 ; cm −1
we get BV ″ + BV ′ = 4.015 cm−1
BV ″ − BV ′ = 0.151 cm−1
Solving these equations for BV ″ and BV ′ , ,we get
BV ″ = 2.083 cm−1 and
BV ′ = 1.932 cm−1
16.8575
∴ I′ = (au) = 8.725 au
1.0932
16.8575
and I ′′ = (au) = 8.092 au
2.083
14 × 14
Since mm for N +2 = amu = 7 amu, the bond length for N +2 in the lower state is
14 + 14
I′ 8.725 (au ) o o
r′ = = = 1.246( A 2 ) = .116 A
μm 7 (amu )
and for the upper state, it is
I ′′ 8.092 (au ) o o
r ′′ = = = .156( A 2 ) = .075 A
μm 7 (amu )
(e) The value of m at the band head is given by
m=
− ( BV BV ′ )
=
(
− 4.015 ) = −13.29 ≈ −13.
2 ( BV BV ′ ) 2 × 0.151( )
Thus, the frequency at the band head:
v head = [25566.06 + 4.015 (−13) + 0.151 × (−13)2]cm−1
= [25566.60 − 52.195 + 25.519] cm−1 = 25539.384 cm−1

(f) The separation between the band head and null line is (25539.384 − 25566.06) cm−1 = − 26.676 cm−1. The
negative value of m and that of separation between band head and null line show that the band is shaded towards
violet.
(g) The dissociation energies in the lower and upper states can be determined with the aid of Eq. (3.44), i.e.
ΔG ′ = w e′ − 2w e′ x e′ (V ′ + 1)
Thus, for the lower level,

⎛ 1⎞
ΔG ′ = − ⎜⎝V ′ + ⎟⎠ ;
2
V′max = 67 and
⎡ ⎛ 1⎞ ⎛ 1⎞ ⎤
2
De′ = ⎢ 2207.19 67 + ⎟ − 16.136 67 + ⎟ ⎥ cm −1

⎢⎣ ⎝ 2⎠ ⎝ 2 ⎠ ⎥⎦
= 75465.675 cm−1 = 9.34 eV = 902.41 kJ mole−1
Similarly, DG″ = 2419.84 − 46.38 (V″ + 1);
V″max = 51 and
⎡ ⎛ 1⎞ ⎛ 1⎞ ⎤
2
De′′ = ⎢ 2419.84 51 + ⎟ − 23.19 51 + ⎟ ⎥ cm −1

⎢⎣ ⎝ 2⎠ ⎝ 2 ⎠ ⎥⎦
= 63116.083 cm−1 = 7.81 eV= 754.742 kJ mole−1
Problem 5.17: Calculate the heat of dissociation of HX where X = F, Cl, Br, and I from the following data:
AB Do (A2 ) Do(B2 ) −ΔHAB
HF 103.24 38.5 63.23
The values of D°s HC1 103.24 57.10 22.03 are in kcal mole−1.
HBr 103.24 45.44 12.16
HI 103.24 35.57 1.09
Solution Substituting the values of respective parameters into Eq. (5.46), we get DoHF = 134.1 kcal mole−1,
DoHC1 = 102.2 kcal mole−1, DoHBr = 86.5 kcal mole−1 and DoHI = 70.5 kcal mole−1.
5.10.2 Predissociation Limit Method

As has already been stated in Section 5.9, the process of predissociation occurs due to transfer of a molecule
from a stable excited electronic state to an unstable electronic state of the molecule in an electronic absorp-
tion spectrum.The transfer of molecule to the latter results into diffuse bands due to anomalous broadening of
rotation lines in a progression in the electronic absorption spectra of diatomic gases. The observed frequency
corresponding to the predissociation limit yields the upper limit of the dissociation energy of the state causing
the predissociation, i.e.
D 0″ v pred (5.53)
D0″ can be very precisely determined provided it is possible to observe the break off of the rotational structure
in several bands, i.e. vibrational levels of the molecule, e.g. H2, N2, P2, SO, etc. However, the state of the predis-
sociation products of the molecule must be known and taken into consideration. The dissociation energy of P2 by
this method has been found to be 116.03 kcal mole−1 on the assumption that the P2 molecule dissociates in the
predissociation region into atoms P (4S) and P (2D). The sharp limit of predissociation that coincides with the true
dissociation limit is observed in the absorption spectrum of polyatomic NO2 which dissociates as NO2 → NO + O.
The dissociation energy from the predissociation method comes out to be 70.8 kcal mole−1. Thermochemical data
yields a value of 69.6 kcal mole−1. The predissociation limit in the spectra of a majority of polyatomics is highly
diffuse and in such cases only the upper limit of dissociation energy is obtained.
It is to be noted that predissociation limit, i.e. when predissociation begins, does not always give an upper
limiting value for the corresponding dissociation limit of the state causing predissociation. Under certain circum-
stances, the predissociation limit may be appreciably higher than the dissociation limit belonging to it.
5.11 LOW RESOLUTION UV VISIBLE SPECTROSCOPY

5.11.1 Introduction
Historically, the simplest form of colourimetry, namely colour matching, as in the case of pH indicators has
been known for a long time. However, colourimetry, or spectrophotometry, is comparatively a new technique.
The invention of the spectroscope, an instrument which separates visible radiations/radiations of different wave-
lengths, i.e. disperses them into a spectrum for visual observation, in the closing decades of the nineteenth cen-
taury brought with it an analytical approach which proved to be fruitful qualitatively in the beginning and both
quantitatively and qualitatively later on. Gradually, few colourimetric methods were introduced for substances
for which other techniques such as volumetric and gravimetric were unknown and unreliable. The rapid development
of electrical measurement techniques in the years around 1930 resulted in the improvement of colourimetric
methods. However, the use of spectral region on either side of the visible spectrum for analysis began only after
1930 in case of UV and after 1940 in case of IR radiations. Existence of IR radiation was demonstrated by W
Herschel in 1800 and Julius used it for qualitative analysis in 1892. WW Coblentz in 1905 observed that certain
atomic groups in molecules always gave characteristic absorption bands in definite positions that varied but little
with the structure of the rest of the molecule. In 1913, Eva von Bahr resolved the rotational structure of HC1.
There followed the demonstration of quantised rotational energy levels in molecules, which led to rapid progress
in the analysis of their IR spectra and the determination, from
the data, of molecular structure and moments of inertia. The wide
spread of IR region for analytical purposes began around 1943. It
was mainly developed for the analysis of polybutadiene during the
Second-World War (1939–45). During this period, SBR (synthetic
styrenebutadiene rubber) was the most important of the synthetic
rubbers developed to replace unavailable natural rubber because
Absorbance
of its structural resemblance with natural rubber, i.e. cis-poly

2- methyl-, 3-butadiene. In the case of UV region, though the
discovery of UV radiation dates back to 1880, but its use for
analytical purposes began only after 1930 and a reliable absorption
spectrum in this region was obtained conveniently within a
reasonable time in 1941.
As has already been stated that the electronic state of a mol-
ecule is associated with vibrational and rotational energy states,
230 250 270
consequently, the electronic spectra of molecules in the vapour
Wavelength (in nrn)
phase show simultaneously rotational and vibrational fine struc-
Fig. 5.15 Ultraviolet absorption spectrum of ben-
tures. Due to this, the electronic absorption line becomes a broad zene. The dashed lines represent the vibrational
peak, i.e. a band containing rotational and vibrational fine struc- coarse structure visible under high resolution.
tures. However, when a spectrum is recorded in solution phase,
rotational and vibrational features in the spectra disappear due
to solute–solvent interactions and a smooth curve is obtained. The vibrational fine structure is also observed
in certain electronic transitions when the spectra of dilute solutions in nonpolar solvents or of vapours are
recorded under high resolution, e.g. 260 nm transition of benzene shows vibrational fine structure as shown in
Fig. 5.15. However, low-resolution electronic spectra of benzene does not show such a vibrational fine struc-
ture. Thus, the low-resolution spectrum of molecules in solution or vapour phase will consist of a smooth curve
free of rotational and vibrational fine structures and is of interest to analysts for qualitative and quantitative
analysis. Such a band is characterised by its intensity (i.e. absorbance) and position (i.e. wavelength). Most
of the applications of low-resolution UV-visible spectroscopy are based on these two parameters and will be
discussed later on.
5.11.2 Terms and Symbols Associated with UV/V Absorption

Measurements of Molecules in Solution/Vapour Phase
Most of the terms and symbols and formulae involved in low-resolution spectroscopy of diatomics and poly-
atomic molecules in solution/gas phase have already been described in Chapter 1 on introduction. In addition, the
terms which are routinely used in UV/V spectroscopy are described here.
1.0
10
0.9
9
0.8
8
0.7
0.6
Absorbance
0.5 6
7
0.4 6.5
6.8
0.3 7
8
0.2 9 6.8
0.1 6.5
6
5
0
300 400 500 600 700 800
Wavelength (nm)
Fig. 5.16 The absorption spectra of phenolred at various pH values.
(a) Isosbestic Point pH affects the absorption spectrum of the reagents where the protonation of a functional
group takes place. However, there is a point at which the absorption curves at different pHs intersect and at this
point absorbance is indepdent of pH and degree of dissociation of the reagent (Fig. 5.16). This point at which all the
curves at different pHs interesect is called the isosbestic or isoabsorptive point and is a characteristic of a system
consisting of two chromophores which are interconvertible so that the total quantity is constant, i.e. the existence of
an isosbestic point indicates that a chemical equilibrium exists between two species. The wavelength corresponding
to the isosbestic point is known as isosbestic wavelength. The absence of an isosbestic point indicates a more com-
plex system, e.g. Pyridonine, a molecule related to vitamin B6 undergoes the following reversible reaction at pH < 7.
UV absorption spectra of this reagent in the 220–350 nm range at various pHs exhibits three isosbestic points at
l = 236, 266 and 305 nm. Similarly, visible absorption spectra of phenolred in the 350–800 nm range at various
pHs exhibits an isosbestic point at l = 495 nm. Absorbance curves at various pHs (pH = 5–10) reveal that as the
pH increases, the absorption at l = 610 nm increases while at l = 430 nm, it decreases. The curves at different pHs
intersect at l = 495 nm, the isosbestic point. The graph between absorbance and pH at the isosbestic wavelength
will be S-shaped from which the pK value of the phenol can be determined. The pH value at which one half of the
reagent is present in the acid and one half in the basic forms is the pK value of the reagent.
CH2OH CH2OH
–
HO CH2OH O CH2OH
+ H+
H3C N+ H3C N+
H H
(b) Band Shifts and Intensity Changes The position and the intensity of the absorption band is not only af-
fected by pH but also depends upon the nature of the solvent. The substituents present in the molecule also affect
the electronic spectrum of the parent molecule via inductive effect, resonance effect, etc. Since inductive effect
influences the charge densities on the atoms of the parent molecule (i.e. it affects the Coulomb integral only), either
all the molecular orbitals are raised or lowered in energy relative to the parent molecule containing no substitu-
ent. However, the difference between HOMO and LUMO remains unchanged. Consequently, the band maximum
of the transition is hardly influenced but the intensity of the band grows because of increase in the dipole length.
On the other hand, resonance or mesomeric effect shifts the band maximum. In resonance effect, the lone pair of
the substituent participates in the bonding and tries to behave like an extended conjugation. As a result of this in-
teraction, the energies of both the HOMO and LUMO are raised, but the effect is more pronounced in the former,
because of which the energy difference between the HOMO and LUMO decreases. Consequently, the band maxi-
mum shifts to longer wavelength. The interaction of nonbonding electrons with s orbital will be less as compared
to p orbital. Thus, the resonance effect will be more pronounced in p–p * transition than s–s* transition. Greater
the electronegativity of the hetero atom containing the lone pair, the more tightly it will be bound to the atom, and
hence resonance effect will be small and lmax of the original transition will be shifted slightly. The effect of substitu-
tion on the transition follows the order:
–NR2 >–NH2 >– OH > I > Br > Cl > F.
Further, the arrangement of the lone pair in space also affects the interaction. The interaction will be maxi-
mum if the orbitals are properly arranged, e.g if n and p orbitals are parallel to each other, their interaction will
be maximum and minimum if perpendicularly arranged. Due to this reason, the energies of p and p* orbitals will
be affected accordingly. The shift of the p-p* absorption band will be maximum in the former case, i.e. parallel
arrangment and minimum in the latter case, i.e. perpendicular arrangement of orbitals. Further, it is to be noted
that the inductive and resonance effects due to substituents operate simultaneously and may act in the same direc-
tion or in the opposite direction. For example, both –NH2 and –OH groups decrease the energy of nonbonding
electrons by inductive effect. The resonance of a lone pair of electrons in both these groups will increase the
energy of the p* anti-bonding orbital. Both the effects are acting in the same direction, i.e. increase the energy
between the nonbonding and p* anti-bonding molecular orbitals.
The shifts of absorption bands and their intensity changes are generally described in the literature using the
terminology shown in Fig 5.17.
(c) Red Shift The shift of absorption maximum to longer wavelength due to substitution or solvent effect is called
red shift. p-p* transition of mesityloxide undergoes red shift as the polarity of the solvent increases, i.e. lmax= 230.6 in
iso-octanol and lmax = 237.6 in chloroform. lmax for thiophene in hexane is observed at 231 nm (emax = 7100) while for
2-bromothiophene, it is observed at 236 nm (emax = 9100). n-p* transition of cyclohexanone in hexane is observed at
285 nm, (emax = 14) while in o-chloro, -bromo, -hydroxy cyclohexanones lmax is observed at +307, +313, +302 nm,
respectively in the axial conformation.
(d) Blue Shift The shift of absorption maximum to shorter wavelength due to substitution or solvent effect is called
blue shift. n–p* transition of mesityloxide undergoes blue shift as the polarity of the solvent increases, i.e. lmax = 321 nm
in iso-octanol and lmax = 314 nm chloroform, n–p* transition of cyclohexanone in hexane at 285 nm (emax =14) is ob-
served at 278, 280, 273 nm in o-chloro, -bromo, -hydroxycyclohexanones respectively in the equatorial conformation.
(e) Hyperchromic Effect The increase in absorption intensity due to solvent effect or substitution is called
hyperchromic effect. For example, K (p-p*) and B (Benzenoid) bands of aniline are observed at 230 (emax = 8600)
Hyperchomic effect
λ max
Blue shift Red shift
Absorbance
or or
Hypsochromic shift Bathochromic shift
Hypochromic
effect
Wavelength (nm)
Fig. 5.17 Hypothetical plot between absorbance and wavelength describing the terminology related to the shifts of absorption maxima
and their intensity changes.
and 280 nm (emax = 1430) respectively in water o-nitroaniline exhibts these bands at 283 (emax = 5400) and 358 nm
(emax = 4500). In case of phenol p–p*: lmax = 210.5 nm (emax = 6200) and B: lmax = 270 nm (emax 1450) in water
while in o-nitrophenol p-p*: lmax = 279 nm (emax = 6600) and B: 351 nm (emax = 3200). The intensity of n-p* band
of azomethine [C3H7N = C(CH3)2] increases as the polarity of the solvent increases, lmax = 246 nm (emax = 140) in
cyclohexane and lmax = 232 nm (emax = 200) in ethanol. Here, the hyperchromic effect is caused by the solvent.
(f) Hypochromic Effect The decrease in absorption intensity due to solvent effect or substitution is called
hypochromic effect. For example, furan shows an intense band at 200 nm (emax = 10,000) and a weak band at
252 nm (emax = 1). In 2-nitrofuran these bands appear 225 nm with emax = 3400 and at 315 nm with emax = 8100.
These bands in 2-methoxyfuran appear at 225 nm (emax = 2300) and 270 nm (emax = 12900). The intensity of n–p*
band of azomethine [C3H7N=C (CH3)2] decreases as the polarity of the solvent decreases, i.e. the n-p* transition
of azomethine exhibits hypochromic effect as the polarity of the solvent decreases.
5.12 INTENSITY OF ABSORPTION BANDS

Recall from Part I that the intensity of an absorption band in the electronic absorption spectrum of a molecule
depends on the transition probability corresponding to the band. The transition with high probability yields an
intense band and the transition is said to be permissible, while the transition with low probability is said to be
forbidden and results in a weak band. The intensity of an absorption band is measured in terms of different
parameters. One such parameter is the molar decadic extinction coefficient. The intensity of the absorption band
is measured in terms of molar extinction coefficient emax at the value of wavelength of maximum absorption. It
is a measure of the strength of the electric dipole transition involved and its value varies from 5 × 105 for the
strongest bands to one or less for very weak bands. The other parameter is the integrated intensity. The integrated
intensity of an absorption band is the molar extinction absorption coefficient integrated over the entire band and
is given by
v2
Integrated absorption intensity = ∫ ε (v ) dv (5.54)

v1
where ν1 and ν 2 are the frequency limits in cm−1. It is determined graphically by plotting e against frequency (in
cm−1) and calculating the area of the band from the relation on
v2
∫ ε ( )d
v1
= ε (v ) × Δv 1/ 2 (5.55)
where Δν 1 is the width of the band in cm−1 and is temperature dependent. Accordingly, the integrated absorption
2
intensity also depends upon temperature.
Another useful dimensionless parameter for expressing the intensity of the absorption band is the oscillator
strength (f) and is linked to the integrated absorption intensity of the band by
v2 v2
2.303 × 103 me c 2
∫ ε (v )dv = 4 33 × 10 ∫ ε (v )dv
−9
f = (5.56)
N A e2 v1 v1
where me is the mass of the electron in grams, e is the charge of the electron in electrostatic unit, ν is the
frequency expressed in wave numbers (cm−1). The quantity 2.303 is the conversion factor from natural logarithm
to the base- ten logarithm while the factor 103 is the number of cubic centimetres in a litre required by introduction
of the molar absorption coefficient.
It is defined as the effective number of dispersion electrons, which are oscillating with a characteristic fre-
quency vi. For example, the spectral lines emitted by an atom can be associated with the corresponding electronic
oscillator strengths. The coefficient of integral is independent of temperature whereas the integrated absorption
intensity is temperature dependent. Consequently the oscillator strength will also depend upon temperature.
For symmetric absorption bands,
f ~ 4.60 × 10−9 (emax Δν 1 ) (5.56a)
2
Otherwise, f ≈ 4.33 × 10−9(emax Δν 1 ) (556b)

2
where the frequencies are measured in cm−1 and the value of emax is expressed in dm3 mole−1 cm−1 (or 103
mole−1 cm2, since 1 dm3 =1000 cm3).
Actually, ‘f ’ is a product of theoretical intensity of transition and integrated absorption intensity. The advan-
tage of introducing oscillator strength is that it can be interpreted theoretically and can be calculated quantum
mechanically. It is found that intense transitions have oscillator strenghts of about unity. Values of ′ f from 0.1 to 1
correspond to absorptivities in the range 103 to 105 depending on the width of the band.
The transition dipole moment is related to the observed oscillator strength by
3 he 2
M t2 =
⋅f (5.57)
8π 2 m e v
Substituting the values of the respective constants into Eq. (5.57), we obtain
(6 626 × 10 ) × (1 602 × 10 C )
34 2 2 19 2
3 f
M =
2
14 )
(3.14 ( )v ( )
t 2
8 × 9.10 10 31 1
=
0.7107 × 10 −42 ( )f
v
21
or Mt 0.843
8 3 10 f / v (Cm) (5.57a)
1/ 2
⎡ ε Δν ⎤
or M t = 0.0958 ⎢ max 1/ 2 ⎥ (Debye) (5.57b)
⎣ νmax ⎦
Here, frequencies are in cm−1 and emax in dm3 mole−1 cm−1;
1D = 3.335641 × 10−31 Cm
ε max Δν 1 1
−30 2 2
or M t 0.31955x10 [ ] Cm (5.57c)
ν max
As has already been defined in chapters 2 and 3,
M t (M M ) err = e ∫ ψ 2r
*
r ψ 1 dr
Accordingly, the effective dipole length is given by

ε max Δν1/ 2
30
0.31955x [ ] Cm
M ν max
r r12 = t =
e 1.602x 19 C
1
⎡ ε Δν ⎤ 2
= 1.99469 × 10 −12 ⎢ max 1/ 2 ⎥ m (5.58)
⎣ ν max ⎦
For example, from the charge-transfer band observed around 300 nm in the case of 2, 4-dinitrophenol-benzene
system, we obtain Δν 1/2 = 614 cm−1, emax = 1960.8 dm3 mole−1 cm−1. Based on this data, we get
f = 4.33 × 10−9 × 1960.8 × 614 = 5.2 × 10−3
1
⎡1960.8 × 614 ⎤ 2
M t = 0.0958 ⎢ ⎥ = 0 57 D
⎣ 333 × 10 ⎦
2
and r=
M t 0.57( D ) 3.335641 × 10
=
−30
( ) = 1.186 × 10 11
m 0.118Å
e 1.602 × 10 −19 (C)
To conclude, we can say that theoretically the oscillatory strength of a fully allowed transition is unity and e is
about 104. SO from the extent of deviation of experimental value of oscillator strength from unity, one can predict
whether a given transition is forbidden or allowed. The values of oscillator strengths of the bands in the absorption
spectrum of an aromatic carbonyl compound are listed in Table 5.6.
Table 5.6 Oscillator strengths for transitions in an aromatic carbonyl compound.
lmax Tansition Oscillator strength f emax(in m2mole-1)

170 p–p* 0.5 allowed 200
180 n–s* 0.02 forbidden 20
280 n–p* 0.0004 forbidden
350 n–p* 10−5 forbidden
Often bands with emax values below 100 m2 mole−1 result from forbidden while those with emax above 1000 m2
mole−1 result from allowed transitions.
5.12.1 Factors Affecting Position and Intensify of Absorption Bands

Both physical and chemical environmental factors influence the position and intensity of the absorption bands.
The study of these effects aids in prescribing best analytical conditions for best optimum precesion and sensitivity.
The two types of effects that influence the wavelength of maximum absorption as well as intensity are intermo-
lecular and intramolecular effects.
(a) Intramolecular Effects They are due to a change in the structure of a molecule that causes a change in
the absorption spectrum. Optical isomers have identical spectra. Structural changes may be brought about by
substitution which acts as mesomeric and inductive effects. Substitution may bring steric effects into operation
which modifies the spectrum, e.g. 0–0 substituted stilbenes. Steric hindrance to coplanarity about a double
bond causes an increase in the ground state energy but the energy of the excited state relatively remains un-
changed. Consequently, a red shift results from the above results. Mild steric hindrace to coplanarity about a
single bond has a small effect on the position of the absorption maxima but mild steric hindrance to coplanar-
ity about a single bond strongly decreases the intensity and may cause either blue or red shift of the absorp-
tion maximum, e.g. p-nitroaniline shows absorption maximum at 375 nm (emax = 16000) while 2, 6-dimethyl-
p-nitroaniline absorbs at 385 nm (emax = 4840). p-methylacetophenone absorbs at 252 nm (emax = 15000) while
2,4, 6-triethylacetophenone absorbs at 242 nm (emax = 3200). Extreme steric hindrance about a single bond
prevents overlapping of the separated chromophores, i.e. prevents conjugation. Consequently, a blue shift is
expected and has been observed indeed in many cases, e.g. dilactone produced from shelloic acid shows no
absorption in UV region but an acid of this lactone absorbs at 227 nm (emax = 5500). The overlapping of double
bond and carbonyl orbitals occurs since a constraint is released. One can make use of intramolecular effect for
specific analytical purposes by modifying the structure of ligands to analyse inorganic ions colourimetrically,
e.g. 4, 7-diphenyl-1, 10-phenanthralene is a better reagent for iron (II) and Cu (II) since it increases the molar
extinction coefficient, thus giving higher sensitivity. Aromatic amines and phenols are determined calorimetri-
cally by diazotisation to give coloured dye. The sensitivity of detection and limit of detection is far better in
modified form.
(b) Intermolecular Effects They are due to solute–solvent interaction in the solution or interaction of other
solutes in the solution with the solute under study. Solvents affect absorption by two or three mechanisms, namely
their dielectric property, solvating power and/or hydrogen bonding. As the dielectric constant, i.e. polarity of the
solvent, increases, the change in the instantaneous dipole moment during transition also increases, and conse-
quently, the transition energy becomes less and absorption shifts to the longer- wavelength side. The molar dec-
adic coefficient also changes to some extent.
Solvent effect is more pronounced in case of transitions involving a lone pair electrons as in pyridine, C = O
compounds, NO2 compounds, etc. Polar solvents or hydroxylic solvent solvate and hence stabilise the lone pair
strongly. Consequently, more energy is required for excitation and the absorption moves to shorter-wavelength
side. This effect is much more pronounced when the solute–solvent interaction is of H-bond type. Apart from
these, the tautomeric equilibrium also depends upon solvent polarity, e.g. in ethylacetoacetate, the keto form has
only a weak n−p* band (lmax = 275 nm, emax = 16) of an isolated carbonyl group, while the ⎯C=CH⎯C=O unit
in the enol form absorbs strongly (lmax = 244 nm, emax = 16000). The percentage of enol form decreases as the
polarity of the solvent increases.
(c) Effect of Temperature Temperature also affects the nature of the absorption spectra. Thionine and meth-
ylene blue show reversible shape of absorption curve in the temperature range 10−70 °C. When acetoacetic ester
in petroleum ether is cooled with liquid air, a pure keto form (m.p. −39°C) crystallises out. At room temperature,
it returns slowly to the equilibrium mixture. On vacuum distillation from a quartz flask, the first fraction distilled
off is the lower- boiling enol form. The distilled off enol form, when allowed to stand, also changes to the equi-
librium mixture of the two forms.
(d) Effect of Irradiation Irradiation can cause reversible changes due to photo-fading, rearrangement, photo-
oxidation, isomerisation. Irradiation causes cis–trans isomerisation of thioindigo or stilbene. Ascorbic acid, when
oxidised to dehydroascorbic acid, does not absorb in the ultraviolet range. Salts of quinoline show reversible effect
on irradiation. Their colour changes in light and they regain their original colour when stored in dark. This effect
is reversible but fading is irreversible. This is strongly shown by 10−5M methylene blue and 0.02M ferrous ion at
pH = 3. The strong blue solution becomes colourless when exposed to light and regains colour again. This is prob-
ably an oxidation–reduction reaction. The spectrophotometers which irradiate the analyte after dispersion have
a specific advantage in this respect because amount of energy falling on the analyte is quite low and the effects
would be reduced to a large extent.
5.13 SYMMETRY SELECTION RULES FOR POLYATOMIC MOLECULES

The probability of an electronic transition in polyatomics depends upon the change in a large number of factors,
namely electron spin, orbital symmetry, momentum of the molecule, and electronic wave function involved in the
transition. The probabilities due to these factors contribute to the total probability of an electronic transition which
is defined by f, the oscillator strength of the transition under study and is given by
f p fa (5.59)
where Ps Po ′ Pp a
and
nd Pm denote the spin-, orbital-, parity-, and momentum probabilities respectively. fa is the oscil-
lator strength of fully allowed p −p* transition and is the order of unity whereas the values of the P factors differ
from unity. This formula forms the basis of selection rules for polyatomics.
(a) Electron Spin Ps When spin of an electron changes in an electronic transition, it is said to be spin forbid-
den. Thus, electronic transitions between states of same multiplicity (2S + 1) are allowed while between states
of different multiplicity are forbidden, i.e. singlet–singlet, doublet–doublet, etc. transitions are highly allowed
while singlet–triplet transitions are highly forbidden. The probability of S–T transitions is of the order of 10−5 for
second row elements. This rule fails, i.e. S–T transitions are enhanced when heavy atoms such as iodine or para-
magnetic molecules such as O2 are added to the system under study, e.g. S–T transition is enhanced in the absorp-
tion spectrum (400–500 nm) of liquid naphthalene when ethyliodide is added to it as a solvent; and 2-naphthyl
phenyl ketone in presence of O2 (100 atm). In both the cases, vibration coarse structures has been observed in the
S–T bands. The 0–0 band appears on the high wavelength side of the S–T band.
(b) Parity Pp We know that if electronic wave functions of a molecule that have the same charge remain un-
changed when inverted at the midpoint of the internuclear axis, i.e. the centre of symmetry of the molecule, it is
said to have an even parity and is designated as g (g stands for gerade) and odd parity u, (u stands for ungerade)
when it changes sign on reflection at a plane through the centre of symmetry, e.g. wave functions corresponding
to bonding p orbitals of symmetric ethylenes are ungerade while those corresponding to anti-bonding p* orbitals
are gerade.
According to parity selection rule (also called the Laporte rule), an electronic transition is permissible between
states of different parity, i.e. the transition g ↔ u is allowed while g ↔ g and u ↔ u are forbidden. In some cases
weak transitions are observed, though they are parity forbidden. This is due to the coupling of vibrational motion
of the nuclei with the electronic motion. Coupling of vibrational motion changes the symmetry of the molecule
and hence the above rule is not vigorously followed. These types of transitions are called vibronic transitions. In
aromatic hydrocarbons, the electronic transition is parity forbidden and the probability Pp is of the order of 10−1
for a condensed systems. The electron spin has no influence on the even–odd symmetry.
(c) Orbital Symmetry P0 When two orbitals involved in an electronic transition simultaneously have large
amplitudes in the same region of space, the transitions are called space allowed or non-overlap, otherwise orbit-
ally forbidden as shown in Fig. 5.20. P0 for n−p* transitions is 10−2 in carbonyl compounds or for second-row
heteroatoms.
(d) Momentum Pm The electronic transition is said to be momentum forbidden when there is a large change in the
linear or angular momentum of the molecule in the electronic transition: Pm for condensed ring compounds falls in the
range 10-1 −10-3.
Thus, it is inferred that the transitions that strictly obey the selection rules are said to be highly allowed and give
rise to intense bands in the electronic absorption spectra of molecules. Otherwise they are highly forbidden, i.e.
the probability of occurrence is very low and results in very weak bands in the observed spectra.
5.14 MECHANISM OF ABSORPTION AND OF COLOUR

We know that the electrons in the atoms of the molecules occupy electronic states of certain energy values and
the UV/V radiations change the electronic conFiguration of the molecules. If E1 and E2 are the energies of the
electronic ground and excited states, the frequency of transitions n between these two levels under resonance
condition is given by ± ΔE = hv = hc v = hc 1 , where ΔE = E2 − El for absorption or ΔE = E1 − E2, for emission.
λ
In organic molecules, the absorption of light causes the electrons to move from a low-energy orbital to a high-
energy orbital whereas in inorganic molecules, partially filled d-orbitals of transition metal ions are involved
in electronic transitions. The latter will be discussed in detail under a separate heading entitled ‘Inorganic
Spectroscopy’.
5.15 ELECTRON TRANSITIONS IN ORGANIC MOLECULES

In organic molecules, transitions occur between bonding and anti-bonding molecular orbitals. Bonding orbitals
result due to the linear combination of atomic orbitals when their wave functions are added, i.e. Ψbonding = Ψa + Ψb,
while the anti-bonding orbitals are formed due to the linear combination of atomic orbitals when their wave func-
tions are subtracted, i.e. Ψanti-bonding = Ψa − Ψb. Ψa and Ψb are the wave functions of the atomic orbitals. Ψσ and Ψ σ *
are the wave functions of the bonding and anti-bonding molecular orbitals. Bonding orbitals have lower energy
(more stable) than the combining atomic orbitals while the reverse is true for the anti-boilding molecular orbitals.
Accordingly, Ebonding < Eanti-bonding.
There are three types of molecular orbitals in organic molecules, namely sigma orbitals, pi-orbitals and
non-bonding orbitals.
(a) Sigma Orbitals These orbitals are formed by the linear combination of 1s [Ψ (1s)] and of 2p [Ψ (2pz)]
atomic orbitals lying along the axis joining the atoms. The resulting molecular orbitals are shown in Fig. 5.18.
Bonding and anti-bonding orbitals are denoted by the symbols s and s* respectively.
(b) Pi-orbitals They are formed by the sidewise overlapping of 2py [Ψ (2py)] atomic orbitals. The overlapping
gives rise to a bonding and anti-bonding orbitals as shown in Fig. 5.19. Bonding and anti-bonding p-orbitals are
denoted by the symbols p and p* respectively.
The six molecular orbitals in benzene are formed by the overlap of the six 2p atomic orbitals, three of these are
p orbitals and three are p* orbitals as shown in Fig. 5.20. The ultraviolet absorption spectrum of benzene in the
range 180–280 nm shows three bands corresponding to three p – p* transitions.
(c) Nonbonding Orbitals Hetero-atoms in molecules have electrons in orbitals associated with them and
these orbitals do not participate in a bonding system of the molecules and hence are known as anti-bonding orbit-
als. They are denoted by the symbol n, e.g. there are two electrons in the nonbonding 2p orbital ( n orbitals) on
the oxygen atom as shown in Fig. 5.21. The absorption of radiation causes one of these electrons to move either
to a s* or p* orbital.
+ − σ∗
ence
Differ
+ +
Sum
Ψ (1s) Ψ (1s) + σ
e − + − + σ∗
nc
e re
if f
D
− + + −
Ψ (2 pz) Ψ (2 pz) Su
m
− + − σ
Fig. 5.18 Schematic diagram for the formation of sigma orbitals from two 1s and two 2pZ atomic orbitals.
+ −
π∗
nce
fere −
Dif +
+ +
− − +
Sum
Y (2 py) Y (2 py)
π
Fig. 5.19 Schematic diagram for the formation of pi-orbitals from two 2py atomic orbitals.
p*
p
Fig. 5.20 Schematic diagram for the p bonding and p* anti-bonding orbits in benzene. Solid contours represent the negative parts of the wave
function while dotted contours are for positive parts of the wave function.
π∗
π
n
Y
X C ⎯
⎯ O
Fig. 5.21 Representation of the n, p and p* orbitals associated with the carbonyl group. p and p* orbitals lie in the YZ plane while n orbital
(2p) points along the X-axis.
From the foregoing discussion, it is evident that four types of electronic transitions in molecules are possible
and these have been classified as follows: s − s*, n − s*, n − p* and p − p*. It can be seen that for a molecule
containing s, p* and n orbitals-, the lowest energy transition is likely to be n − p*. But it is possible that for mol-
ecules with a high degree of conjugation, the p orbital can have higher energy than the n orbital and thus−p − p*
transition would then have one of lowest energy. The energy of these transitions follows the order: s − s* > n −
s* > n − p* > p − p*. The p − p* transitions in conjugated carbon systems are of particular chemical interest. The
wavelength of radiation needed for s − s* and n − s* transitions is often shorter than 200 nm and hence not handy
from a practical point of view. The schematic energy-level diagram showing these transitions is given in Fig. 5.22.
s∗(Anti-bonding)
n s∗
p ∗(Anti-bonding)
n p∗
n(Non-bonding)
p p∗
p(Bonding)
s s∗
s (Bonding)
Fig 5.22 Schematic energy- level diagram showing various electronic transitions.
The Mulliken’s notations corresponding to molecular orbital symbols for electronic transitions in organic
molecules are givn below:
Mulliken’s notations Molecular orbital notations

N→V s →s* , p → p*
N→Q n →p* , n → s*
Photo-ionisation, i.e. the excitation energy is so high
N→R
that the electron leaves the molecule
N−denotes the ground state; V, Q, R, etc. denote the excited states, N–R denotes the Rydberg transition. The group
theory and Platt’s symbols for electronic transitions are not used in the text. Readers wanting to learn about these
symbols may consult any book on electronic spectroscopy.
5.15.1 Characteristics of Various Types of Electronic Transitions

The electronic transitions are characterised by their energies and intensities and the effect of various parameters
such as temperature and nature of solvent on them. The solvent effects are especially useful in identifying the
electronic transitions.
(a) N → V Transitions This type of transition results when an electron moves from a bonding orbital to
an anti-bonding orbital. The excited state is polarised. This means that the transitions are accompanied by a
considerable change of the dipole moment and the corresponding band is of high intensity. The most common
types of N → V transitions are s − s*, and p − p*, transitions. s − s* transitions lie in the vacuum ultraviolet
(< 200 nm) region and generally do not appear when spectra are recorded on commercial spectrophotometers since
absorption of UV radiation by atmospheric oxygen starts at
p∗
≈ 195 nm (Schumann−Runge band system) and most spec-
p∗
trometers are limited to this wavelength limit, e.g. saturated ΔE1
hydrocarbons exhibit s − s* transitions. s − s* transition ΔE2
of CH4 is observed at 122 nm while that of ethane is at 135 p
p
nm. p − p* transitions require a lower transition energy as No Solvation
Solvation
compared to s − s* transitions and lie in the near ultravio-
let (200–400 nm) and sometimes even in the near infrared ΔE1 > ΔE2
(4000–12000 cm ) region of the spectrum. The intensity of
-1 (a)
p − p* transitions is high, the molar decadic extinction coef- p∗
ficient usually exceeds 1000. p∗
In polar solvents, the p − p* absorption shifts to longer ΔE1

wavelengths. The shift increases as the polarity of the solvent ΔE2
increases. The shift is probably because of more stabilisation n

No Solvation n
of excited state than that of the ground state due to solvation
Solvation
and hydrogen bonding. [Fig. 5.23(a)]. The hydrogen bond-
ΔE1 < ΔE2
ing causes much larger spectral shifts than most other sol-
(b)
vent–solute effects, e.g. unsaturated organic molecules with
double and triple bonds exhibit p − p* transitions: p − p* Fig. 5.23 Schematic diagram showing the effect of solvation
transition of ethylene is observed at 165 nm (emax = 10000) on (a) p–p *, and (b) n–p * transitions.
and that of acetylene at 173 nm (emax = 6000). The substitu-
tion of an electron- attracting group in the molecule increases the double-bond character, and consequently the
absorption shifts to longer wavelengths.
It is observed that introduction of one more double-bond conjugated to the first one, shifts the p − p* transi-
tion to longer wavalength by about 30 nm. Moreover, in highly conjugated p-electron systems, the movement of
electron in the conjugated system can be compared to a particle in a one- dimensional box. The energies of the
h2n2
levels are given by E n = , where n = 1, 2, 3…, a is the effective length of the molecule. In lycopene, there
8m e a 2
are eleven conjucted double bonds and lmax = 474 nm. Thus, the total number of p-electrons = 22. Two electrons
with opposite spins can reside in one energy level. Thus, the highest occupied molecular orbital (HOMO) and the
lowest unoccupied molecular orbital (LUMO) are 11 and 12 respectively. Accordingly, the energy of transition is
h2 23h 2
ΔE
Δ E12 − E11 = ⎡122 − 112 ⎤⎦ =
2 ⎣
8m e a 8m e a 2
hc 23h 2
or hv = =
λ 8m e a 2
1/ 2
⎡ 23h λ ⎤
or a=⎢ ⎥
⎣ 8m e c ⎦
Substituting the values of the respective constants and l = 474 nm, we get
1
⎡ 23 × (6.26 × 10 −34 kg m 2 s −2 ss)) × ( 474 × 10 −9 m) ⎤ 2 1
a=⎢ ⎥ = ⎡⎣3.30 × 10 −18 m 2 ⎤⎦ 2
⎣ 8(9.1 × 10 −31 kg) × (3 × 108 ms m −1 ) ⎦
= 18.1 × 10−10 m = 18.1 Å
In case of butadiene, lmax = 217 nm. The number of p electrons = 4. So according to electron-free model
a = 573 pm [i.e. two C = C bonds (= 2 × 135 pm = 270 pm + one C⎯C bond (= 154) pm + the distance of carbon
radius at each end = 2 77.0 pm ( = 154 pm ) = 578 pm.]
(b) N– Q Transitions This type of transition results when an electron moves from a nonbonding orbital (more
polar) to a anti-bonding orbital ( less polar). The most common types of N–Q transitions are n – p* and n – s*
transitions. These transitions are observed at longer wavelengths as compared to p – p* transitions but the cor-
responding molar decadic coefficients in n – p* transitions are usually lower than 100 and > 100 <1000 in case of
n – s* transitions, e.g. unsaturated molecules containing hetero atoms such as O, N or S exhibit n–p* transitions
n – p* transitions in acetone and acetaldehyde are observed at 279 nm (emax = 15) and 290 nm (emax = 17) respec-
tively in n-hexane. On the other hand, saturated molecules containing hetero atoms such as O, N, S with lone-pair
electrons exhibit n – s* and s – s* transitions. [(CH3)3N: shows bands at 227 nm and 99 nm corresponding to
n – s* and s – s* transitions respectively. n – s* transition of methanol is observed at 177 nm (emax = 200) in
hexane while that of 1-hexanethiol is observed at 224 nm (emax = 126) in cyclohexane.
In polar solvents, the absorption shifts to shorter wavelengths in both the n – s* and n – p* transitions
because of more stabilisation of ground state than that of the excited state due to solvation and hydrogen bonding
[Fig. 5.23(b)]. The substitution of an electron-repelling group in the unsaturated molecules a containing hetero
atom, decreases the double- bond, character, and consequently the absorption will shift to shorter wavelengths.
Note that when spectrum of an amine like RN: where R = (CH3)3 −, (C2H5)3 −, for n − s* , C5H5 − (for n −p *)
is recorded in acidic medium, the band → corresponding to n–p* or n − s* ( as the case may be) transition is not
observed due to protonation of the nitrogen atom:
RN: + H+ → RN+H
This has been fixed as one of the criteria to differentiate between n – p * and p – p * transitions.
(c) Rydberg or Atomic Transition When the excitation energy lies in the vacuum ultraviolet region, photo-ioniastion
of the molecule occurs [i.e. X + hv → X + + e ] and intense narrow bands appear in the spectrum.
5.15.2 Colour
Colour is a property of light which reaches our eyes but not a 580 nm 620 nm
property of the substance we see. Since substances absorb cer- Orange
tain wavelengths from the light incident upon them and reflect
others, the colour of an opaque substance depends upon the kind Yellow Red
of light which it reflects to the eye. If a substance reflects all 800 nm
565 nm
400 nm
colours it receives, we say it is white. We call it black if it absorbs Green Violet
all the wavelengths of light that fall upon it. However, the colour
of transparent substances depends upon the colour of light which Blue
they transmit. Any two colours which combine to form white 490 nm 430 nm
light are said to be complementary. The rotating colour wheel of Fig 5.24 Rotating colour wheel of the various comple-
various complementary colours is shown in Fig. 5.24 and visual mentary colours. The apparent colour is across the wheel
responses of colours are given in Table 5.24. from the actual colour absorbed.
Tabel 5.7 Visual responses of colour.
Colour absorbed Wavelength ( in nm) Apparent colour of substance

UV < 300 Colourless
Violet 380–435 Yellowish green
Blue 435–480 Yellow
Greenish blue 480–490 Orange
Bluish Green 490–500 Red
Green 500–560 Purple, rose, magenta
Yellowish green 560–580 Violet
Yellow 580–595 Blue
Orange 595–650 Greenish blue
Red 650–780 Bluish green
Infrared > 780 Colourless
(a) Organic and Inorganic Colours Organic colours are obtained by lowering the transition energies until
the visible region is reached, while inorganic colours are obtained when the transition energies are increased
from zero until the visible energies are reached. The organic chemist makes use of p − p* transitions to explain
the colour of organic compounds while inorganic chemists make use of the partially filled d-orbitals of transition
metal ions to explain the colour of complexes, and will be discussed later on under the heading entitled ‘electronic
absorption spectra of inorganic molecules’.
(b) Chromophores Certain unsaturated chemical groups which when substituted in hydrocarbons impart
them colour, were christened as chromophores by Witt in 1876, i.e. a covalently unsaturated group responsible
for electronic absorption is called a chromophore (or chromophoric group). The chromophores are of interest
to chemists from the point of view of elucidation of structure and related properties of the molecules. Some of
the chemical chromophores along with their respective absorption maxima and molar extinction coefficients are
Table 5.8 Typical chromophore groups.
a ( nm )
λ max 5
Group System Solvent log 10 Transition
C C Ethylene Vapour 165 417609 p–p*
193 400000 p–p*

—C ≡ C— Acetylene Vapour 173 377815 p–p*
C O Acetone Vapour 166 420412 p–p*
n–Hexane 189 295424 p–p*

279 117609 n–p*
C O Acetaldehyde Vapour 160 430103 p–p*
180 400000
n–Hexane 290 123045 n–p*
C S 205 strong
495 weak
S=O Cyclohexylmethylsulphoxide Ethanol 210 317609
O
N Nitromethane Ethanol 271 126951 n–p*
O
Methanol 201, 274 369897, 123045

—ONO2 Butylnitrate Ethanol 270 123044 n–p*
—ONO Butylnitrite n-Hexane 220 416136 p–p*
356 193951 p–p*
C N Acetoxime Water 190 369897 p–p*
—C≡N Acetonitrile Vapour 167 ? n–p*

—N=N— Azomethane Ethanol 338 60206 n–p*
= N2 Diazomethane Vapour 410 47713
—N3 220 217609
287 130103
—COOH Acetic acid Water 204 177816 n–p*
Methyl orange has a very long conjugated system of delocalised p-electrons, and it absorbs in the visible region
at lmax = 460 nm. The conjugation of p-electrons in this molecule is extended by the chromophore —N = N— as
is evident from the structure of methyl orange given here. High conjugation is responsible for the colours of many
other dyestuffs and natural pigments. Lycopene, the red pigment in CH3
HO3S N=N ¨
N
tomatoes, has an absorption maximum at around 485 nm due to the
CH3
presence of eleven conjugated ethylenic bonds. Ultraviolet spectra
Structure of methyl orange
of organic compounds are therefore associated with delocalised
p-electrons.
(c) Auxochromes or Auxochromic Groups Saturated groups with nonbonded electrons, which when
attached to chromophores alter both the wavelength and intensity of the absorption band corresponding to p – p *
transition of chromophores, are termed auxochromes or auxochromic groups. These groups do not show absorp-
tion bands on their own in the 200 – 800 nm range. The most typical auxochromes are NH 2 NR 2 , OH, OR , SH .
The substitution of electron-repelling groups, especially those having unpaired electrons, shifts the wavelength of
p – p* transitions to shorter wavelengths. The effect of some auxochromes on the benzene chromophore is given
in Table 5.9.
Table 5.9 Effect of auxochromes on absorption of benzene.
Compound Solvent Ethylenic Band Benzenoid band
lmax (in nm) Logemax lmax (in nm) Logemax
Benzene Cyclohexane 198 3.90 255 2.36

Diphenylether Cyclohexane 255 4.04 272 3.30
278 3.25
Phenol Water 210.5 3.79 270 3.16
Thiophenol n–Hexane 236 4.00 269 2.85
Aniline Water 230 3.93 280 3.16
Anilinium cation Aq. acid 203 3.87 254 2.20
Chlorobenzene Ethanol 210 3.88 257 2.23

Phenolate ion Aq. base 235 3.97 287 3.42
The number of isolated double bonds in organic molecules does not practically influence the position of the
band but intensity of the band increases proportionately. However, when the number of conjugate double bonds
increases, the band corresponding to p – p* transition shifts to longer wavelength side of the spectrum and ulti-
mately at some limiting value of the conjugation chain length, the absorption maxima shifts to the visible region
and the compounds become coloured. The effect of conjugation chain length on lmax in case of organic molecules
is shown in Tables 5.10 and 5.11.
Table 5.10 Values of λmax molecules with different conjugation chain lengths.
Compound lmax (in nm)
n=1 n=2 n=3 n=4 n=5

CH3 ⎯(CH=CH)n- COOH
204 254 294 327 —
(in alcohol)
CH3 ⎯(CH=CH)n⎯CHO
220 271 315 353 393
(in alcohol)
CH3 ⎯(CH=CH)n ⎯C6H5 306 334 358 384 420
(in benzene) (24) (40) (75) (86) (114)
*
The quantities within parenthesis are values of emax × 10−3.
Table 5.11 Values of lmax and emax for a series of hydrocarbons with number of conjugated bonds increasing from compound to compound.
Compound Number of conjugated bonds lmax(in nm) emax × 10

−3 Colour
Butadiene 2 217 21 Colourless

Hexatriene 3 258 35 Colourless
Decatetraene 4 310 42 Colourless
Vitamin A1 5 326 4.66* Colourless
Tetradecahexaene 6 360 70 Colourless
a-Carotene 10 445 145 Orrange
Lycopene 11 476 191 Red
*log emax
The absorption band due to p – p * transition in such C C C C C C C C

conjugated systems, i.e. enes, is called a K (Konjugierte) π ∗ 4
band and in these conjugated systems there is one common π∗ π∗
K-band instead of several bands. Thus, the length of the π ∗ 3
conjugated system can be deduced from the position and

intensity of the K-band as is evident from Tables 5.10 and
5.11. The K bands of enes are independent of the polarity π 2
of the solvent. π π
The electronic transition is accompanied by a change Isolated
π
Isolated
1
of dipole moment, and thus depending on the polarity system
Conjugated
system
system
induced in the excited state, the abosprtion band may have (a)
high or low intensity. In long conjugated systems, the
exited dipolar states have large dipole moments and hence
C C C C C C C O
intense K-bands are observed in these sytems. Values of π ∗ 4
emax for K-bands lie in the range of 104 – 105. On the other π∗ π∗
hand, less energy is required to excite p electrons in con- π ∗ 3
jugated molecules, and as the conjugation increases, the
excitation energy decreases, i.e. the absorption shifts to n
longer wavelength side of the spectrum (Fig. 5.25a).

Carbonyl compounds such as ketones, aldehydes, π 2
esters, etc., not only show intense K- bands but also a low- π π
intensity band called R (Radikalartig) band due to n–p* Isolated π Isolated

1
system system
transition, (Fig. 5.25b). The former moves to the longer Conjugated
system
wavelength side, while the latter to shorter wavelength (b)
with the polarity of the solvent. In general, chromophores Fig 5.25 Molecular orbital energy relationships between iso-
having one electron pair on the hetero atom give rise to lated and conjugated. (a ) C = C group in dienes, and (b) C = C
R-bands. Since R-bands are of low intensity they are called and C=O groups in a, b-unsaturated ketones.
forbidden bands. emax of R-bands is <100. K-bands of carbonyl compounds, differ from those of enes in that the
former depend on the polarity of the solvent (undergo red shift), while the latter are independent of it. Absorption
data for some conjugated enes are carbonyl compounds are compared here in Table 5.12.
Table 5.12 Absorption data for some conjugated enes and carbonyls.
Systetm K-band ( p − p*) R-band ( n − p*)

lmax (in nm) emax lmax(innm) emax
1,3 –Butadiene ( in hexane) 217 21000 — —
1,3, –Cycolohexadiene ( in hexane) 256 8000 — —
Methylvinylketone ( in ethanol) 212.5 7080 320 20.9
Acrolein ( in water) 210 11482 315 25.7
Crotonoaldehyde ( in ethanol) 220 14792 322 28.2
Acetone ( in hexane) 188 900 279 15
Aromatics or hetero-aromatics show three absorption bands, say at wavelengths λ E1 λ E 2 / K and lB, due to p – p*
transitions. The absorptions and intensities corresponding to these bands follow the order.
λ B λ E2 / K > λ E1
and ε ma
B
x(
2 3
) ε mE2ax/ K ( 3 4
) < ε max
E1
( 4 5
)
The band at lB is called the benzenoid band and is regarded as a representative of the ‘forbidden’ biradical
transition. B-bands generally show a vibrational fine structure which often disappears or is reduced in polar sol-
vents. The other two bands at wavelengths λ E2 / K and λ E1 are called ethylenic (E) bands and like the B-band they
may also show vibrational fine structure. Any substituent capable of conjugation causes a red shift of the K-band,
B-band and an increase in the intensity of the bands of the parent molecule. These effects vary from substituent
to substituent. When a chromophore, i.e. NO2, C = O, etc., is attached to an aromatic ring, the B-bands undergo
more red shift as compared to E2/K bands. When an R-band due to n –p* transition is also observed along with the
p–p* transitions of the aromatics, the R-band moves to longer wavelengths. Vibrational fine structure may not be
observed in the spectra of substituted aromatics.
When an auxochromic group NH 2 , OH , etc., is attached to the aromatic molecules; B and E bands, i.e. E1,
and E2, undergo red shift. The intensity of the B-band of the parent molecule also increases while that of the E2
band does not show any definite trend. The effect of auxochromic substitution on B and E bands of benzene is
demonstrated in Table 5.9.
In case of benzene, the effect of ortho-, para- and meta-substituents on the red shift may be arranged in the
following two series:
⎯CN < ⎯ SO2R < ⎯ COOH < ⎯COCH3 < ⎯CHO < ⎯NO2
and Cl < OH < SR < NR2
Thus, it is evident that the magnitude of red shift increases as System lmax (K-band) (in nm)
the conjugating power of the substituent increases. When differ-
Benzene 200
ent kinds of substituents are in the para position, the K-bands of
benzene are strongly red shifted as is evident from the spectral Aniline 230
data indicated below: Nitrobenzene 269
The red shift of K-band of benzene due to conjugation effect p-dinitrobenzene 266
of various substituents affords a method for the determination p-nitroaniline 381
of the position of the substituent in the benzene ring, which is
m-nitroaniline 280
required for analysis of mixtures of meta- and para-substituted
compounds.
The absorption data of some substituted and unsubstituted aromatics and hetero aromatics is compared in
Table 5.13.
Table 5.13 Absorption data of some typical substituted and unsubstituted aromatics and hetero aromatics.
System E1-band (p -p*) K/E2-band (p-p*) B-band (p-p*) R-band (p-p*) Solvent
lmax (in nm) emax (103) lmax (in nm) emax (103) lmax (in nm) emax (103) lmax emax
Benzene 184 60 204 7.9 256 0.2

Naphthalene 221 133 286 9.3 312 0.289
Anthracene 256 180 375 9.0 submerged
o-nitroaniline 283 5.4 412 4.5
o-nitrophenol 279 6.6 351 3.2
m- nitroaniline 280 4.8 358 1.45
m-nitrophenol 274 6 333 1.96
p-nitroaniline 381 13.5 submerged
p -nitrophenol 318 10 submerged
Styrene 244 12 282 0 .45 Ethanol
Benzaldehyde 244 15 280 1.5 328 20 Ethanol
System E1-band (p -p*) K/E2-band (p-p*) B-band (p-p*) R-band (p-p*) Solvent
lmax (in nm) emax (103) lmax (in nm) emax (103) lmax (in nm) emax (103) lmax emax
Acetophenone 240 13 278 1.1 319 50 Ethanol
Nitrobenzene 252 10 280 1 330 125 Hexane
Quinoline 228 40 270 3.162 315 2.5 Cyclohexane
Acridine 250 20 358 10 Ethanol
Further, the lower members of the acene series have typical

benzenoid absorption bands with the following characteristics
(benzene is given for comparison). System lmax (in nm) emax
The larger the number of six-membered rings fused together
in a linear manner, (I) the stronger the bathochromic shift of light Benzene 225 230
absorption and more intense the absorption band. At the same time, Naphthalene 314 316
the stability of the compound decreases sharply, its aromaticity is
diminished, and it becomes increasingly unsaturated and closes to a Anthracene 380 7900
polyenic hydrocarbon with conjugated bonds. This is easily under-
standable since only one ring of an acene has a sextet of electrons. Tetracene 480 11,000
When the system of any acene, say anthracene, is symmetrical, the Pentacene 580 12,600
two electrons in the sextet are mobile and migrate from one end of
the system to the other as shown below:
The arrow indicates the migration of a pair of electrons of the
sextet.
The angular condensed hydrocarbons, (II) beginning with
phenanthrene, contain a larger number of completely aromatic
rings and the path covered by the pair of electrons from the sextet
is shorter.
These hydrocarbons absorb light in a larger wavelength region
(I)
of the spectrum, are more stable and aromatic.
Solutions of the red-coloured hydrocarbon rubrene (9,10,11,
12-tetraphenyltetracene) are decolourised in sunlight by atmospheric
oxygen due to the formation of peroxide (III) shown below.
(II)
This rubrene peroxide decomposes in vacuum at 140–150 °C into
C6H5 C6H5
molecular oxygen and rubrene. The evolved oxygen in the active sin-
glet state adds on to the conjugated C—C p-bonds, i.e.
O
O
+ O2
C6H5 C6H5
O O (III)
Based on the ultraviolet spectral data of all classes of organic compounds, i.e. fused-ring systems, many steroi-
des, terpenes, heterocyclic compounds and their derivatives, etc., rules have been worked out from which one can
predict the position of the absorption maximum in some organic systems. The rules will be discussed in detail in
the application part under the title ‘Structural Elucidation’.
5.16 INORGANIC ELECTRONIC ABSORPTION SPECTROSCOPY

Before discussing the electronic absorption spectra of inorganic molecules, the basic theories that laid the founda-
tion of electronic absorption spectroscopy of inorganic substances need be introduced here in brief.
All the five d-orbitals, i.e. dxy, dyz, dzx,, dz2, dx2y-2, in an isolated transition metal ion in the gaseous state and that
of an ion in a spherical electronic field are degenerate but the orbitals in the latter are at a higher energy level com-
pared to the former because of electron repulsion between metal-ion electron density and spherical field of nega-
tive charge. The splitting of energy levels or orbitals of a metal ion under the influence of an electric field of the
oppositely charged ions surrounding it as in a crystal is called crystal field splitting. The details and extent of split-
ting depend on the symmetry and the strength of the crystalline electric field. According to this theory, there is no
overlapping between the orbitals of the neighbouring ions and the central metal ion under consideration and only
repulsion is operative between the negative charge of the surrounding ions and electrons of the central metal ion,
i.e. the model assumes that the bonds between metal ions and oppositely charged ions surrounding it are of ionic
nature. This model also describes the properties of complexes where the metal ion is under the influence of the elec-
tric field created by the ligands (L). But this ionic or electrostatic model is not realistic since in complexes of transi-
tion metal ions in which the transition metal atoms, say M, are surrounded by a definite number of bound groups
called ligands (arranged in one of several definite symmetries: octahedral, tetrahedral, tetragonal or square planar)
whose charge clouds exert a marked influence on the electronic configuration of the transition element. These bound
groups may be monoatomic or polyatomic ions such as Cl−, F−, CN−, etc., or neutral polar molecules usually having
one or more lone pairs of electrons, e.g. H2O, NH3, CO, etc. The overlapping of electron clouds of ligands with that
of the central metal ion removes the degeneracy of some orbitals of the metal ion which have the same energy values
in the absence of the electric field of the ligand. This lifting of degeneracy of some orbitals of the metal ion by the
electric field of the ligand in the metal complexes is called ligand-field splitting. In other words, we can say that in
some complexes in which we have assumed electrons to be enitrely in the atomic d-orbitals of the metal, may they
actually spread out and spend some time in the orbitals belonging to the ligand and vice versa. Thus, there are three
kinds of interactions involving M and L: (i) L → M, s overlap of orbitals, (ii) L → M, p- overlap of orbitals, and
(iii) M → L, back bonding ( p*) interactions, i.e. dp–pp back bonding due to p-overlap of full d-orbitals of M with
empty p-orbitals of L. These interactions give rise to considerable covalency to M – L bonds which are now made up of
M–L molecular orbitals. Accordingly, the lone pair orbitals of ligand are combined with s, p and d orbitals of M to
give bonding, anti-bonding and sometimes nonbonding orbitals in the complexes. The electrons from the ligands
then occupy the bonding orbitals and impart covalency to the complexes. The remaining electrons partially fill the
anti-bonding orbitals.
Both the theories explain the splitting of d-orbital degeneracy of isolated metal ions in different ways, but they
give rise to the same general pattern of energy levels: The ligand field theory may thus be visualised as crystal field
theory with superimposed molecular orbital concepts i.e the ligand field theory is a modification of the crystal
field theory in which we drop the assumption that partially filled electron shell in one consisting of pure d orbitals.
Instead there is overlap between the d orbitals of metal and orbitals of the ligand atoms. The theory deals with three
broad categories: (i) crystal field strength is less than the spin-orbit (L.–S.) coupling (ii) crystal-field strength is of
intermediate strength in comparison to spin-orbit coupling and electron repulsion terms e2/rij, and (iii) crystal-field
strength is greater than the repulsion terms, i.e. e2/rij. Complexes of rare earths (4f) belong to the category (i) while
complexes of first transition group and covalent complexes belong to categories (ii) and (iii) respectively.
5.16.1 Energy Levels in Transition Metal Complexes

Let us consider an ion with single d electron, i.e. d1 configuration subjected to an external electric field (crystal
field) of octahedral symmetry. The five d-orbitals, viz., d xy , d yz , d xz , d z 2 , d x 2 − y 2 of a metal ion in the gaseous
state are degenerate, i.e. they have the same energy values. The single d-electron can occupy any one of the five
d-orbitals and the probability of occupancy is the same for all the d-orbitals. In the octahedral environment, out
of the five orbitals the two d z 2 d x 2 − y 2 correspond to high electron density along the axes while the remaining
three d xy , d yz , d xz correspond to high electron density along the 45° lines between the axes as shown in Fig. 5.26.
Thus, when a d electron is present in one of the d-orbitals along the axes, i.e. d z 2 d x 2 − y 2, there will be stronger
repulsion between the electron cloud of the d electron and that of the ligand. The repulsion will be less if the elec-
tron is present in one of the d-orbitals that are directed between the ligands with minimum charge density on the
axes, i.e. d xy , d yz and d xz . Consequently, d z 2 d x 2 − y 2 are high-energy orbitals as compared to the dxy, dyz and dzx
orbitals in an octahedral environment. Not all the five d-orbitals are degenerate in the octahedral environment but
split into two groups: one group of triply and a second group of doubly degenerate levels. The single electron in
the d x 2 y 2 orbitals is being repelled by four ligands, while the electron in the d z 2 orbital is only being repelled by
two ligands. Thus, the energy of d x 2 y 2 increases relative to that of d z 2 and if the ligand field is sufficiently strong,
the difference in energy between these two orbitals becomes larger than the energy needed to pair the electrons.
The group of dxy, dyz and dzx orbitals is labeled as 2 g or de orbitals while that of d z 2 d x 2 − y 2 as eg or dg. Thus,
the five d-levels are said to ‘split into eg and 2 g with t 2 g as the lower energy level; where the symbol ‘g’ stands
for gerade, ‘t’ for a set of triply degenerate orbitals and ‘e’for a set of doubly degenerate orbitals. The subscript
‘2’ indicates anti-symmetry with respect to the rotation axis other than the principal axis. The energy of the two
splitted levels is expressed in terms of the crystal-field parameters D and q, where ‘q’, is the magnitude of the
charge on the ligands and ‘D’ is a measure of the polarisability of the central metal ion. Since, in practice, it is
Z Z
X X
Y Y
dz 2 dx 2 −y 2
Higher energy orbitals (a)

Z Z Z
Degenerate
3d
Orbitals
X X X
Y Y
Y
dzx dyz dxy
Lower energy orbitals (b)

Fig. 5.26 High electron density 3d orbitals (a) along the axes, and (b) along 45° lines between the axes in a octahedral environment.
not possible to separate D and q, a single adjustable parameter Dq is considered. The energy of eg level, E(eg) =
+ 6Dq and that of t2g level E(t2g) = –4Dq (with respect to the energy of the degenerate levels, i.e. the bary centre
of the degenerate levels; each neighbour contributes to the blended value in proportion to its overlap at the bary
centre). Thus, the energy difference Δ0 between eg and t2g levels of an ion with d1 configuration in an octahedral
crystal field is 10Dq.
The energy level splitting of an ion with a single d electron in an octahedral crystal field is shown in Fig. 5.27.
The crystal field splitting of d-orbitals of metal ions in other crystal field symmetries, shown in Fig. 5.28, show
that the energy difference between the two separated groups of orbitals of metal ions in various crystal-field sym-
metries follows the trend:
eg /dγ
+6Dq
Δ0 or 10Dq
−4Dq
t2g /dε
Gaseous Spherical Octahedral

ion Field field
Fig. 5.27 Splitting of five degenerate 3d orbitals in an octahedral field.
Square planar >Tetragonal > Octahedral >Tetrahedral

1.30Δ0 Δ0 0.45Δ0
The magnitude of Δ can be obtained from the spectra of transition metals. The frequency v of absorption band
due to d–d transition in a transition metal complex with d1 configuration under resonance condition is given by
ΔE = hv. Thus, knowing v, we can determine Δ, e.g. the absorption maximum in [TiIII ( H2O)6]3+ is observed
at = 500 nm. Therefore, Δ0 = 500 nm = 20,000 cm−1 = 239.4 kJ mole−1.The magnitude of Δ depends on two quanti-
ties: (i) the nature of the ligands, and (ii) the d-electron configuration of the metal ion, which can vary for the same
element in different oxidation states. Note that for ions with more than one d electron, several electron transitions
are possible, giving rise to more than one absorption band. However, these transitions usually fall in or near the
dx 2 −y 2
Energy
dx 2 −y 2
dz 2, dx 2 −y 2
dxy, dyz, dxz
dz 2
dxy
dxy dz 2
dz 2, dx 2 −y 2 dxy, dyz, dxz
dyz, dxz
dyz, dxz
Tetrahedral Spherical Octahedral Tetragonal Square planar

Fig. 5.28 Splitting of 3d orbitals for various symmetries.
visible region. For the transition metal ions having the same electronic configuration dn, i.e. same charge, the value
of Δ0 is affected by the nature of the ligands which controls the strength of the ligand field. This explains the varia-
tion in colour within a given oxidation state for an ion of different ligands. On the basis of Δ values, ligands have
been arranged in the form of a series called the spectrochemical series CO ≈ CN– > NO 2− , >O-phenanthroline −
>
dipyridyl > ethylenediamine > pyridine ≈ NH3 > EDTA > oxalate ≈ H2O > OH > F > NO 3 >Cl > Br > Ι .
– – – – –
Strong ligand field...→ Weak ligand field

Blue end of spectrum... → Red end of spectrum
Low wavelength... → High wavelength.
–
The splitting produced by strong ligands is approximately twice that produced by weak ligand fields. The
spectrochemical series is useful to explain the colour of the transition metal complexes, e.g. NiSO4 is pale green,
[Ni(en)3]2+ is deep blue; [Cu(H2O)4]2+ is pale blue, [Cu(NH3)4]2+ is blue, and CuSO4 salt is colourless, [Ni(NH3)6]2+
is blue while aquo ion [Ni(H2O)6]2+ is green.
The value of Δ0 increases as the charge on the metal ion grows, i.e . Δ0(M3+) > Δ0(M2+). e.g d3 [CrIII(H2O)6]3+
(Δ0 = 17400 cm−1) > d4 [CrII(H2O)6]4+: (Δ0 = 13100 cm-1); d4 [MnIII(H2O)6]3+: (Δ0 = 21000 cm-1) > d5 [MnII(H2O)6]4+
(Δ0 = 7800 cm−1):
The magnitude of Δ0 also grows on descending a group of transition elements, e.g. Δ0[Co(NH3)6]3+ = 23000
cm−1 < Δ0[Rh(NH3)6]3+ = 34000 cm−1 < Δ0 [Ir(NH3)6]3+ = 41000 cm−1.
Further, note that any nonlinear molecule possessing orbital degeneracy being unstable, distorts itself in such
a way so as to remove degeneracy and become stable. This is called Jahn–Teller effect. Distortions are significant
when orbitals are directed towards ligands, i.e., eg orbitals in octahedral and t2 orbitals in tetrahedral complexes.
The distortions from t2g orbitals in octahedral and from e orbitals in tetrahedral complexes are too small to be
observed, since these orbitals do not point toward ligands. This effect in former cases lifts the degeneracy of
the upper level due to which asymmetric absorption band rather than a symmetric one appears in the electronic
absorption spectra of transition metal complexes.
5.16.2 Crystal Field Stabilisation Energy

Fig. 5.27 shows that the energy of eg orbitals increases by an amount +6Dq and that of the t2g orbitals decreases by
−4Dq compared to the energy of degenerate orbitals, under the influence of octahedral crystal field, i.e. an electron
from metal ion, placed in t2g orbitals stabilises the system by −4Dq (where negative sign refers to stabilisation)
while an electron in the eg orbitals destabilises the system by +6Dq (where the positive sign refers to destabilisa-
tion). Thus according to electrostatic model, an ion gets stabilised under the influence of an octahedral crystal
field and the actual energy of stabilisation is called stabilisation energy (CFSE). CFSE not only depends on the
symmetry of field but also on its strength.
The CFSEs of ions with different electronic configurations of 3d orbitals: d0, d1, d,2... d10, i.e. M 2+ ion of the
first transition series in an octahedral crystal field can be computed with the help of the following rules:
(i) An electron from a metal ion placed in t2g orbital stabilises the system by −4 Dq and in an eg orbital destabilises the system
by + 6 Dq.
(ii) Each electron from a metal ion placed in t2g orbital contributes to the stability of the system while in the eg orbital, con-
tributes to the destabilisation of the system.
(iii) In weak crystalline fields, i.e. high spin octahedral complexes, the 3d- electrons are distributed among the available orbitals
in such a way that maximum number of electrons remain unpaired while in strong crystalline fields, i.e. low-spin octahedral
complexes, the distribution is carried out to have maximum number of paired electrons, i.e. minimum number of unpaired
electrons among the available d-orbitals.
Mathematically, CFSE (octahedral) = −0.4nt 2 g + 0.6 ne g (5.60a)
where nt and neg are the number of electrons in t2g and eg orbitals respectively.
2g
Table 5.14 Crystal-field stabilisation energies of Ca++ to, Zn++ ions in weak and strong octahedral fields.
Ca++ Sc++ Ti++ V++ Cr++ Mn++ Fe++ Co++ Ni++ Cu++ Zn++
3d-electronic d0 d1 d2 d3 d4 d5 d6 d7 d8 d9 d10
ConFiguration
Distribution for high Spin
( i.e weak field) ConFigu- eg − − − − 1 2 2 2 2 3 4
ration:
t 2g 0 1 2 3 3 3 4 5 6 6 6
CFSE: –10Dq 0 4 8 12 6 0 4 8 12 6 0
Distribution forLow spin
( i.e high field) configura- t 2g 0 1 2 3 4 5 6 6 6 6 6
tion
eg − − − − − − − 1 2 3 4
CFSE: −10Dq 0 4 8 12 16 20 24 18 12 6 0
According to the above rules, the crystal field stabilisation energies of M2+ ions with d 0, d1, d,2..., d9, d10 configu-
rations in low,- and high,- spin octahedral complexes have been calculated and are recorded in Table 5.14.
Thus, according to the simple electrostatic model, all M2+ ions except Ca++, Mn++ and Zn++, i.e. d0, d5, and d10
configuration, will be stabilised in a weak crystalline octahedral field while in a strong crystalline octahedral field
only d 0 and d10 configurations will not be stabilised.
In tetrahedral complexes, there are four ligands (bulky) instead of six (small in size) and the direction of
approach of ligands does not coincide either with the e or t2 orbitals. The subscript ‘g’ is not considered in tetrahe-
dral complexes since they lack centre of symmetry. Each of these facts reduces the octahedral splitting by a factor
4 4
of 2/3. Thus, the tetrahedral splitting Δt, is roughly of the octahedral splitting Δ0, i.e Δt = Δ0 . Since Δt < Δ0,
9 9
the tetrahedral complexes are of high spin. Moreover, e is the lower energy level and t2 is the upper energy level
in these complexes, i.e. the energy-level diagram for tetrahedral complexes is the inverse of those for octahedral
complexes. The rules to calculate the CFSEs of metal ions with d0, d1, d2,...,d10 configurations in weak octahedral
fields equally apply to high spin tetrahedral complexes except that an electron in the e orbital stabilises the system
by 6Dq and in the t2 orbital destabilises the system by +4Dq.
Accordingly, Eq. (5.60a) for tetrahedral complexes is expressed as CFSE (tetrahedral) = −0.6ne+ 0.4nt2 (5.60b).
The values of CFSEs determined with the aid of Eq. (5.60b) are compared with those of CFSEs in an octahe-
dral field in Table 5.15. Table 5.15 reveals that all M2+ ions, except those with d0, d5 and d10 configurations, are
stabilised in high-spin tetrahedral as well as in high-spin octahedral complexes also.
Table 5.15 Crystal field stabilisation energies of Ca++ to Zn++ ions in high-spin tetrahedral and octahedral complexes.
Number of d- electrons Distribution of electrons in high Tetrahedral CFSE Octahedral CFSE

(M2+) spin tetrahedral field
e t2 D*t Do
d (Ca )
0 ++
− − 0.0 0.0
d (Sc )
1 ++
1 − −0.6 −0.4
d2(Ti++) 2 − −1.2 −0.8
d3(V++) 2 1 −1.2 + 0.4 = −0.8 −1.2
d (Cr )
4 ++
2 2 −1.2 + 0.8 = −0.4 −0.6
d (Mn )
5 ++
2 3 −1.2 + 1.2 = 0.0 0.0
d (Fe )
6 ++
3 3 −1.8 +1.2 = − 0.6 −0.4
d7(Co++) 4 3 −2.4 + 1.2 = 1.2 −0.8
d8(Ni++) 4 4 −2.4 + 1.6 = −0.8 −1.2
d (Cu )
9 ++
4 5 −2.4 + 2.0 = −0.4 −0.6
d (Zn )
10 ++
4 6 −2.4 + 2.4 = 0.0 0.0
Δt = E (e) + E (t2). These CFSE values can also be expressed in terms of Dq.
5.16.3 Charge Transfer Inter-Ligand Transitions in Transition Metal Complexes

Recall from Chapter 3 on infrared spectroscopy that charge-transfer complexation occurs with the transfer of an
electron from the donor molecule to the acceptor molecule. The colour of transition-metal complexes is attrib-
uted to d–d transitions. But there are compounds which have no d electrons but still have intense colours, e.g.
permanganate (VII) ion, (intense purple, d°), chromate (VI) ion (yellow, d°), mercury (II) iodide (brick red d10),
bismuth (III) iodide ( red, orange d10 s2). The colours of these types of complexes are attributed partly to electron
transfer from a ligand orbital to a metal orbital and vice versa with the absorption of light. Such electron- or
charge-transfer processes are designated as L→M and M→L electron transfer respectively. The energy of charge-
transfer transitions is generally higher than those of the d−d transitions and the charge-transfer band generally
appears in the ultraviolet or far ultraviolet region of the spectra. But if the metal is readily oxidisable and the
ligand easily reducible and vice versa, the charge-transfer transitions may occur in the visible region. The charge-
transfer spectra are very intense and mask the d−d transitions e.g., d−d transitions of metal iodide complexes are
hard to observe since the iodide ion is readily oxidisable. The CT transitions in MnO 4− and CrO 42− are attributed
to the transfer of an electron from nonbonding orbital of an oxygen atom to manganese or chromium (i.e. n−p*)
subsequently reducing the metals in the excited state. Alternatively the transition may occur with transfer of
an electron from the p-bonding molecular orbital of oxygen to the molecule orbital which actually is the metal
atomic orbital. In pyridine complex of iridium (III), the metal being easily oxidisable, the charge-transfer transi-
tion involves the transfer of an electron from the atomic orbital of iridium to an empty p* anti-bonding orbital
in pyridine. Charge transfer in gaseous sodium chloride takes place from the ion pair Na+ Cl− to an excited state
consisting of sodium and chloride atoms provided the Franck–Condon principle is obeyed, i.e. r(Na+ Cl−) =
r(NaCl). Intra-ligand transitions such as p−p* and s−s* in metal complexes are affected by coordination and
occur in UV region of the spectrum.
5.16.4 Ligand-Field Stabilisation Energy

In ML6 octahedral complex, as a result of strong L→M sigma overlap, the energy of eg level increases and hence
Δ0 value also increases while due to L→M p-overlap, the energy of t2g level increases but the Δ0 value decreases.
The M− L backbonding lowers the t2g level but increases the Δ0 value. Thus, the net effect of these M→L and L→M
interactions affects the Δ0 value and hence the stabilisation energies, called the ligand-field stabilisation energies
(LFSE) as opposed the CFSE, e.g. CN− and CO, being strong p-acceptor ligands favour strong M→L backbond-
ing interaction due to which these ligands cause large splitting in the energy range of 3.72 eV (= 30,000 cm−1).
On the other hand, the anionic ligands with lone pair of electrons such as Br−1, I−1 called donors to the p-orbitals
of metal, i.e. L→M p-interactions cause small splttings in the energy range of 1.24 eV (= 10,000 cm−1). Thus,
the neutral ligands with lone pair of electrons, which have no p-bonding capacity, may cause intermediate range
of splitting. According to the electrostatic model, this result is unexpected since the halogen anions I−1 and Br−1
with their negative charge should give smaller splitting than a neutral ligand like NH3 or H2O. The answer to this
discrepancy lies in the covalence effect. The lone pair of NH3 or H2O in the sp3 hybrid orbital interacts with the
empty metal hybrid orbital. In contrast, the filled p-orbitals on the halide ligand interact strongly with the empty
d-orbitals of M, i.e. electric charge is transferred from ligand to metal. These types of L→M interactions give
rise to charge transfer transitions. The charge-transfer from a ligand lowers the positive charge on the metal and
reduces the D term in ‘10Dq’. For the same metal, the greater the reducing power of the ligand, e.g. I− > Br− >
Cl−, the lower will be the energy of the charge-transfer trasition. Simialrly, for the same L, the greater the oxidis-
ing power of the metal ion, the lower will be the energy of charge-transfer transition. Thus, for metal ions having
same charge, the value of Δ0 is affected by the nature of the ligand which controls the strength of the ligand field.
On the basis of Δ0 values, ligands are arranged in the form of a series called the spectrochemical series reported in
Section 5.16.1. The order of s-bond interaction of some typical ligands with metal ions based on spectral studies
on octahedral ML6 complexes is
− − − −
NH3 > H2O > F > CI > Br > I
The order of p-repulsion is:
I− > Br− > Cl− > F− > NH3
5.16.5 Selection Rules for Transitions Between Energy States

of Transition Metal Complexes
Before discussing the spectra of inorganic systems, it is nice to summarise the selection rules wo rked out quan-
tummechanically in Section 5.2. The energy difference Δ0 between the two separated groups of orbitals, i.e. eg and
t2g, is identified with the d−d electronic transitions in the transition-metal complexes. The d−d electronic transi-
tions in the complexes are either said to be allowed or forbidden according to the following selection rules.
(a) Spin or Multiplicity Selection Rule Electronic transition between states of different multiplicity are
said to be multiplicity or spin forbidden, i.e. ΔS ≠ 0, is forbidden while between states of same multiplicity
are said to be spin allowed, i.e. ΔS = 0 is allowed. In other words, we can say that electronic transitions in
which the spin of an electron does not change are multiplicity allowed while those in which it changes are
forbidden.
(b) Laporte Parity Rule According to this rule, the only possible transition is between even and odd terms: a
term is said to be even or odd depending on whether the sum of the azimuthal quantum numbers of all electrons
in the atom is even or odd. Thus, the transition in which Δl/ΔL = 0 are Laporte forbidden while for which Δl/ΔL,
= ±1 are Laporte allowed, e.g. for Ca, s2 →s1 p1, Δl/ΔL = ±1, the transition is allowed and emax = 5000−10000 lit
mole−1 cm−1, while d−d transitions in transition metals are forbidden since Δl = 0. However electronic spectra
of very low intensity, i.e. emax = 5−10 lit mole−1 cm−1, are observed because of relaxation in the Laporte rule.
Accordingly, transitions within a given group of p or d orbitals are Laporte forbidden if the molecule has a
centre of symmetry. Thus, d–d transitions in perfect octahedral complexes are forbidden and many complexes
are colourless except the following.
(a) If the centre of symmetry is distorted due to mixing of d and p orbitals of the complex, then the transitions
are no longer pure d−d in character. The transitions actually occur between d levels with different amounts of p
character. The intensity of transition is roughly proportional to the amount of mixing and emax = 20 to 50 lit mole−1
cm−1, e.g. tetrahedral complexes, namely [MnBr4]2− and unsymmetrically substituted octahedral complexes such
as [Co(NH3)5 Cl]2−, are both coloured.
(b) According to the Laporte rule, the complexes with perfect octahedral symmetry (Oh), i.e. centre of sym-
metry should be colourless, e.g. [Co(NH3)6]3+ or [Cu(H2O)6]2+, but the observations are contrary to the rule, i.e.
these complexes are not colourless. In such cases due to the normal vibrations of the M–L bond of the octahedral
complex, some of the molecules lose their centre of symmetry and mixing of d and p orbitals occurs in the unsym-
metrical configuration. Consequently, a very low transition with emax ~ 5 to 25 lit mole−1 cm−1 is observed. Such
transitions are said to be vibronically allowed and the effect is called vibronic coupling. In general, we can say
that Laporte allowed transitions are very intense while Laporte forbidden transitions vary from weak intensity if
the complex is noncentrosymmetric to very weak if it is centrosymmetric. The selection rules are summarised in
Table 5.16.
Table 5.16 Summary of selection rules for transitions between energy states of transition metal complexes.
Laporte Spin Type of transition emax(lit mole–1 cm–1) Example
Allowed Allowed Charge transfer 10,000 [TiCl6 ]3−

Partly allowed, some Allowed d–d 500 [CoBrr4 ]2−
p–d mixing
[C Cl ]2−
Forbidden Allowed d–d 8 –10 [ ( H 2O) ]3+ ;

3+
⎡⎣ V ( H 2O )6 ⎤⎦
[MnBrr4 ]
2–
Partly allowed, some Forbidden d–d 4
p–d mixing
2+
Forbidden Forbidden d–d 0.02 ⎡⎣ Mn ( H 2O )6 ⎤⎦
It is to be kept in mind that spin-allowed transitions are broad while spin-forbidden transitions are usually
sharp. This is due to the fact that multiplicity allowed t2g→ eg transitions give rise to an excited state in which M–L
distance is larger than in the ground state. Thus, according to Franck–Condon principle, a broad band is observed.
This fact may provide an aid in the assignment of bands in the electronic spectra of complexes.
5.16.6 Bandwidths
The bandwidth of an absorption band due to d–d transitions in the electronic absorption spectrum of transition
metal complexes in solution phase not only broadens due to super imposition of vibrational and rotational levels
but also due to the Jahn–Teller effect and spin-orbit coupling. Bandwidths have been found to be of the order of
1000–3000 cm−1 but rarely narrower bands (width ~100 cm−1) are observed.
5.17 ELECTRONIC ABSORPTION SPECTRA OF TRANSITION METAL

COMPLEXES WITH d1, d2, d3,. . . d9 CONFIGURATIONS
IN OCTAHEDRAL AND TETRAHEDRAL FIELDS
Most metal salts are colourless except those of transition elements. Colour variation not only takes place between
elements, but also between oxidation states of the same elements and sometimes within oxidation states, e.g.
anhydrous CuSO4 is colourless while CuSO4· 5H2O is light blue and [Cu(NH3)4]SO4 is dark blue. The origin of
colour of transition metals and their compounds lies in their partially filled d-orbitals. The d–d transition under
the influence of light gives rise to an absorption band in the electronic absorption spectra of transition elements.
For ions with more than one d electron, several transitions are possible giving rise to more than one absorption
band. These bands fall in or near the visible region. The colour of a compound is seen provided the absorption
takes place over a wide range of wavelengths and the colour which we see is determined by the colour or colours
of white light which are transmitted, i.e. not absorbed by the compound, as shown in Fig. 5.29. Figure 5.29 shows
100 0
Violet Blue Green Yellow Red
Percent absorbance
Percent transmission
50 50
0 100
400 500 600
Wavelength (nm)
Fig. 5.29 Electronic absorption spectrum of Ti3+ ion in visible region.
that a single transition occurs with the absorption of the green component (lmax = 500 nm) of visible light and the
colour transmitted is purple, i.e. violet–red.
Before describing the electronic spectra of transition metal ions with different electronic configurations, i.e. d1,
d , d3... d10, it is nice to summarise ground terms for d1 to dl0 transition metal ions, the pattern of splitting of free
2
electronic states under the tetra- and octahedral-fields and the terms arising for s, p and d electronic configurations
in the form of Tables 5.17–5.19 respectively for ready reference. The deductions of the terms for free ions are not
discussed here and the readers interested in these aspects can consult any textbook on general inorganic chemistry
such as A New Concise Inorganic Chemistry by J D Lee:
Table 5.17 Spectroscopic ground terms for d1 to d10 transition-metal ions.
Ion Electronic Con- Ground Terms ml ∑ml = ML S

Figuration dn
2 1 0 -1 - 2
1
Ti3+ d1 2
D 2 2
V3+ d2 3
F 3 1
1
Cr3+ d3 4
F 3 1
2
Cr2+ d4 5
D 2 2
1
Mn2+ d5 6
S 0 2
2
Fe2+ d6(4) 5
D 2 2
1
Co2+ d7(3) 4
F 3 1
2
Ni2+ d8(2) 3
F 3 1
1
Cu2+ d9(1) 2
D 2 2
The quantities within parenthesis are the number of holes on the vacancies.
Table 5.18 Spectroscopic terms arising for s, p and d electronic conFigurations.
Electronic Configuration Ground state terms Excited state terms

s 1 2
S
s 2 1
S
p1, p5 2
P
p2, p4 3
P S, 1D
1
p 3 4
S P, 2D
2
p6 - S
1
d1, d9 2
D -
d2, d8 3
F P, 1G, 1D, 1S.
3
d3, d7 4
F P, 2H, 2G, 2F, 2D, 2P.
4
d4, d6 5
D H, 3G, 3F, 3D, 3P, 1I, 1G, 1F, 1D, 1S.
3
d5 6
S G, 4F, 4D, 4P, 2I, 2H,2G, 2F, 2D, 2P, 2S.
4
d 10 1
S -
When the subshell is more than half filled, the terms are worked out by considering the holes, i.e. the vacancies in the various orbitals, rather
than the large number of electrons actually present, e.g. the number of holes in the electronic configurations p4, p5, d8 and d9 are 2, 1, 2 and
1 respectively.
Table 5.19(a) Transformation of spectroscopic terms of free ions with sI , pI , d I ,f I and gIelectronic conFigurations into Mulliken
states under the octahedral and tetrahedral fields of ligands.
Electronic structure Spectroscopic terms Mulliken terms Tetrahedral field

of free ion octahedral field
s1 S A1g A1
p 1
P T1g T1
d 1
D Eg+ T2g E+ T2
f1
F A2g + T1g + T2g A2 + T1 + T2
g1 G A1g+ Eg + T1g + T2g A1+ E + T1 + T2
The singly degenerate s-orbital is referred to as a1g, where ‘a’ indicates the value of degeneracy, i.e. a = 1. The triply degenerate set of ungerade p-orbitals is
referred to as t1u. The former being spherical and the latter directional along X, Y and Z-axes, are not splitted under the influence of tetrahedral or octahedral
field. The splitting of d, f and g orbitals into a number of degenerate groups of orbitals under the influence of octahedral or tetrahedral field are denoted by
small letters, e.g. the five 3d orbitals are subdivided into two groups: the high-energy group contains two eg orbitals and the low-energy group contains three
t2g orbitals. The terms and symbols such as A1g, A1, T1g, T1, A2g, A2, T2, etc., are due to Mulliken and originate in group theory. The subscript g is not written in
terms of tetrahedral complexes since they lack centre of symmetry.
Table 5.19(b) Energy of the various Mulliken states corresponding to the d1, d2, d3...... d9, electronic conFiguration
of transition-metal ions in octahedral and tetrahedral fields of the ligands.
ConFiguration (Environ- Group state Spectroscopic Mulliken states 2T2g/5T2g Energy(E)

ment) terms
d1 (O)/d6(O) 2
Eg/ 5Eg +6Dq
or D/ D
2 5
d9 (T)/d4(T) 2
T2g/ 5T2g –4Dq
d (O)/d (O)
9 4 2
T2g/ T2g
5
+4Dq
or D/ D
2 5
d (T)/d (T)
1 6 2
Eg/ 5Eg –6Dq
d2 (O)/d7(O) 3
A2g(F)/ 4A2g(F) +12Dq
or F / 4F
3
d8 (T)/d3(T) 3
T1g(P)/ 4T1g(P)
3
T2g(F)/ 4T2g(F) +2Dq
3
T1g(F)/ 4T1g(F) –6Dq
d (O)/d (O)
8 3 3
T1g(P)/ T1g(P)
4
or F/ F
3 4
d (T)/d (T)
2 7 3
T1g(F)/ 4T1g(F) +6Dq
3
T2g(F)/ T2g(F)
4
–2Dq
3
A2g(F)/ 4A2g(F) –12Dq
For tetrahedral environment, the subscript g is ignored. The transition corresponding to d1(O)/d6(O) is 2T2g→2Eg/5T2g→5Eg.
The splitting of seven 4f1 degenerate orbitals of an ion in an octahedral crystal field is shown in Fig. 5.30. The
sevenfold degenerate level splits up into triply degenerate t1g level, triply degenerate t2g level and a singly degener-
ate a2g level. The t1g level is stabilised by −6Dq while the levels t2g and a2g are destabilised by +2Dq and +12Dq
respectively relative to the energy level of seven degenerate orbitals.
d1-ion The most simple and interesting example of an ion with a d1 electronic configuration is that of Ti3+
which is produced as follows. The electronic structure of titanium with an atomic number (Z) of 22 is Ar. 3d2.4s2
a2g
+12 Dq
t2g
2 Dq
−6 Dq
t1g
f1 - Degenerate
f1 - Orbitals in an Octahedral Field
Orbitals
Fig. 5.30 Splitting of 4f1 degenerate orbitals under the influence of an octahedral field.
where Ar represents the inner shell structure. The Ti3+ ion with an electronic configuration of Ar 3d1 results when
three electrons are lost by the titanium atom, i.e. Ar.3d2 4s2 ⎯−3e → Ar.3d1. This uncomplexed, Ti3+ ion has five
degenerate 3d orbitals, i.e. d 1 : d 1xy d yz d zx d x 2 y2
d z 2 , d xy d 1yz d zx d x 2 y2
d z 2 , d xy d yz d 1zx d x 2 y2
d z 2 , d xy d yz d zx d 1x 2 y 2 d z 2 ,
d xy d yz d x 2 y2
d 1z 2 as shown in Fig. 5.31 (a). The probability of a single electron residing in each of the five degener-
ate orbitals is the same. When six ligands, say 6H2O, approach a Ti3+ ion, its five 3d degenerate orbitals split up into
two degenerate groups of low and high energy under the influence of electrostatic octahedral field of the ligands.
The low-energy group (called the ground state) consists of three t2g (i.e. dxy, dyz, and dzx) and is stabilised by −4Dq
while the high–energy group (called the excited state) consists of two eg (i.e. dz2 and dx2−y2) degenerate orbitals and
is destabilised by +6Dq with respect to the bary centre of the five degenerate orbitals. The single electron from the
five 3d degenerate orbitals goes into the lower energy group, i.e. ground state as shown in Fig. 5.31(b). When such
a system is exposed to light of energy hv such that Δ0 = hv, the electron moves from the lower t2g, to the upper eg
state; the probability of residing the electron in both the d-orbitals is equal, [Fig. 5.31 (c)] through d–d transition,
i.e. t2g→eg. The energy corresponding to this transition gives to an absorption band in the middle of the visible
region having an absorption maximum around 500 nm [Fig. 5.29.]. The asymmetry in the band arises due to the
splitting of the excited state because of Jahn–Teller distortion. The purple colour of the [Ti(H2O)6]3+ ion results
eg
+6Dq
hν
Δo Δo = h ν
−4Dq
t2g
Ti3+ ion in spherical Ti3+ ion in octahedral [Ti(H2O)]3+ ion

field or uncomplexed crystal field of water in Excited State
Ti3+ ion Ligand or [Ti(H2O)6]3+
ion in ground state
(a) (b) (c)
Fig.5.31 Energy-level diagram illustrating the origin of electronic absorption
spectrum of d1-ion in an octahedral field of a ligand.
because absorption of light by the system occurs in the blue, green and yellow regions while transmission takes
place mainly in the red and violet regions of the spectrum.
The electronic absorption spectrum of an octahedral complex of dl-ion can also be expressed in terms of the
ground-state spectroscopic term for a free d1-ion, and the Mulliken terms corresponding to it in an octahedral
environment is as follows: The ground state spectroscopic term
for a virgin ion with d1-configuration is 2D. In an octahedral field
2E , E = +6Dq
of the ligand, the 2D state splits up into the 2T2g and 2Eg Mulliken g
states. The former originates from the triply degenerate set of

t2g orbitals, i.e. t 12 g d 1xy 1 1
x y , d y z d z x ; d xy , d yz , d xz d xy , d yz , d zx and is
+6Dq
stabilised by –4Dq, while the latter originates from the doubly
degenerate set of eg orbitals. The electronic configuration of 2Eg 2D
is e 1g : d 1z 2, d x 2 − y 2 , d z 2, d 1x 2 − y 2 and is destabilised by +6Dq by the
Energy
−4Dq
octahedral field of the ligand. Accordingly, the lower-energy 2Tg
state corresponds to the single electron residing in one of the t2g 2T , E = −4Dq
2g
orbitals and the upper-energy 2Eg state corresponds to the single
electron residing in one of the eg orbitals. The energy-level dia-
gram showing the 2D ground state term and its splitting into Ligand Field Stregth
Mulliken states under the influence of the octahedral field of the Fig. 5.32 Energy-level diagram for the spitting of 2D
ligand is shown in Fig. 5.32. The slope of the T2g line is −4Dq spectroscopic term of an ion with d1-conFiguration into
and that of Eg line is + 6Dq. The difference between the two Mulliken terms in an octahedral field of a ligand.
slopes ΔE, is 10Dq or Δ0. The absorption of photons of light hv
by the system promotes the electron from the 2T2g state to the 2Eg Ion Absorption maxima (in cm-1)
state provided Δ0 = hv. Consequently, a single absorption band at
20400 cm−1 results corresponding to the 2T2g → 2Eg transition in [TiCl6]3– 13000
the absorption spectrum of aquo ion [Ti(H2O)6] with d -config-
3+ 1
uration. Further, since E(2Eg) − E(2T2g) = 6Dq − (−4Dq) = 10Dq, [TiF6]3– 18900
therefore 10Dq = 20400 cm Moreover, Fig. 5.32 reveals that
−1.
the magnitude of splitting and hence the energy of d–d transition [Ti(H2O)6]3+ 20300
depends on the nature of the ligand and grows with the strength [Ti(CN)6] 3+
22300
of the ligand and the same has been observed indeed experi-
mentally. For example, the electronic absorption spectra of some
octahedral complexes of Ti with dl-configuration indicated below in the tabular form show that the magnitude of
splitting, i.e. Δ0, fits into the spectrochemical series, i.e. CN− > H2O > F− > Cl−.
The energy-level diagrams for the electronic absorption spectrum of dl ion in a tetrahedral field of a ligand have
been shown in Figs 5.33 and 5.34. In a tetrahedral field, the two splitted groups of five 3d degenerate orbitals,
i.e. d 1 : d 1xy d yz d zx d x 2 y2
d z 2 , d xy d 1yz d zx d x 2 y2
d z 2 , d xy d yz d 1zx d x 2 y2
d z 2 , d xy d yz d zx d 1x 2 y 2 d z 2 , d xy d yz d zx d x 2 y2
d 1z 2 (the
corresponding spectroscopic term is 2D) are inverse of that in an octahedral field of the ligand. The low-energy
t2
+4Dq
hν
Δt Δt = h ν
−6Dq
Uncomplexed d 1ion Ground State of a Excited State of a

Tetrahedral Complex Tetrahedral Complex
of d 1ion of d 1ion
Fig. 5.33 Energy-level diagram illustrating the origin of electronic absorption spectrum of d1 ion in a tetrahedral field of a ligand.
2T2 ; E = + 4Dq
+ 4Dq
Energy
2D
− 6Dq
2E ; E = − 6Dq
Ligand field strength

Fig. 5.34 Energy-level diagram for the splitting of 2D spectroscopic term of an ion with d1 conFiguration into Mulliken
terms in a tetrahedral field of a ligand.
group of two degenerate e orbitals (i.e. e 1 : d 1z 2 , d x 2 − y 2 , d z 2 , d 1x 2 − y 2 ; the Mulliken term corresponding to this set is
2
E) is stabilised by −6Dq while the high-energy group of three t2 orbitals (i.e. t 12 : d 1xy , d yz , d zx ;(;.....) the Mulliken
term corresponding to this set is 2T2) is destabilised by +4Dq with respect to the bary centre of the five 3d degen-
erate orbitals by the tetrahedral field of the ligand. Accordingly, the low-energy group of e orbitals constitutes
the ground state and high-energy group of t orbitals constitutes the excited state of the system. Thus, when such
a system is exposed to light of energy hv such that Δt = hv, the electron jumps from ground to the excited state
giving rise to an absorption band in the spectrum by d–d transition.
e→t2
or
2
E → 2T2
Further, Fig. 5.34 reveals that just like an octahedral complex of d1 ion, the magnitude of splitting Δt of a d1 ion
in a tetrahedral field of the ligand depends on the nature of the ligand and increases as the strength of the ligand
field increases.
It is to be noted that for tetrahedral complexes, the d-d absorption bands are considerably more intense than in
octahedral complexes. The reason being that tetrahedral complexes lack centre of symmetry, while the symmetry
in octahedral complexes is slightly distorted because of the vibronic effect. In other words, we can say that in
tetrahedral complexes the d-d transitions are Laporte-forbidden due to the absence of centre of symmetry while
Laporte-allowed in octahedral complexes.
(a) d9 ion The common example of an ion with, d9 electronic configuration is that of Cu2+. The electronic
structure of copper with atomic number (Z) of 29 is Ar 3d104s1. The Cu2+ ion with an electronic configuration of
Ar 3d9 results when two electrons are removed from the copper atom, i.e. Ar 3d104s1 ⎯−2e → Ar3d9. The free ion
having a d9 electronic configuration has five 3d degenerate orbitals (i.e. d9: d 1xy d yz2 d zx2 d z22 d x22 − y 2 ; d xy2 d 1yz d zx2 d z22 d x22 − y 2 ;
d xy2 d yz2 d 1zx d z22 d x22 − y 2 ; d xy2 d yz2 d zx2 d 1z d x2 y
; d xy2 d yz2 d zx2 d z22 d 1x 2 − y 2 ; the spectroscopic term corresponding to this set of
d-orbitals is 2D, Figs 5.35(a). and 5.36). On complexing with six ligands, 6H2O for example, the five 3d degen-
erate orbitals split up into two degenerate sets, the low-energy and high-energy sets under the influence of the
octahedral field of the ligand. The low-energy set consists of three t2g, while the high-energy set consists of two
eg orbitals. The ground state is stabilised by –6Dq and represents an electronic configuration t 26g e g3 : t 26g d x22 y2
d 1z 2 ,
1
t 26g d x 2 y 2 d z22 ; the Mulliken term corresponding to this configuration is Eg, while the excited state is destabilised
2
by +4Dq with an electronic configuration 2 g e g : d xy d yz d zx e g , t 2 g e g : d xy d yz d zx2 e g4 , d xy2 d yz2 d 1zx e g4 ; the Mulliken state
5 4 1 2 2 4 5 4 2 1
corresponding to this configuration is 2T2g. The single positive hole (o) resides in one of the two eg orbitals as
shown in Figs 5.35(b) or in 2Eg level; Fig. 5.36. On absorption of light of photon hv such that Δ0 = hv, the single
hole from the eg orbitals of ground state (or from the 2Eg level) crosses over to the t2g orbitals by a d–d transition
eg t2g
+ 6Dq
hν
Δ0 Δo = hν
− 4Dq
t2
g
Ground state of an Excited state of an

Uncomplexed d 9 ion Octahedral complex octahedral complex
of d 9 ion of d 9 ion
(a) (b) (c)
Fig. 5.35 Figure illustrating the origin of d–d transition in d9-ion in an octahedral field of the ligand.
(or to the 2T2g level); Figs 5.35(c) and 5.36. The energy corresponding to 2T
2g, E = +4Dq
this d–d transition, i.e. t2g → eg or t e [ E g ] t e [ T 2 g ] will give rise

6 3
2g g
2 5 4
2g g
2
+4Dq
to an absorption band in the electronic absorption spectrum of d ion in an
9 2 D
Energy
octahedral field and the same has been observed indeed as an asymmetric
−6Dq
band around 794 nm in the spectrum of [Cu(H2O)6]2+ ion. The asymmetry in
the band has been attributed to Jahn–Teller distortion. The pale blue colour 2E , E = −6Dq
g
of aquo ion [Cu(H2O)6] results because absorption of white light occurs
2+
Ligand field stregth
in the red region while a transmission occurs mainly in the blue region of
Fig. 5.36 Energy-level diagram of d9-ion
the spectrum. Again, it is evident from Fig. 5.36, that alike d1 octahedral in an octahedral field of a ligand in terms
and tetrahedral complexes the magnitude of splitting Δ0 between two sets of 2D spectroscopic term and the Mulliken
of orbitals in d9 octahedral complexes depends on the strength of the ligand terms corresponding to it.
and increases as the strength of the ligand increases.
In tetrahedral complexes of a d9 ion, the energy-level diagrams are the inverse of that of a d9 ion in an octahedral
field of the ligand [Figs. 5.37 and 5.36]. The spectroscopic term corresponding to the five 3d degenerate orbitals
for free d9-ion is 2D. The ground state of the complex can be written as 4 25 : e 4d 1xy d yz2 d zx2 , e 4d xy2 d 1yz d zx2 , e 4d xy2 d yz2 d 1zx ;
the Mulliken state corresponding to this configuration is 2T2 while the excited state can be written as
e 3t 26 : e 3d xy2 d yz2 d zx2 , e 3d xy2 d yz2 d zx2 , e 3d xy2 d yz2 d zx2 ; the Mulliken state corresponding to this configuration is 2E. In this case
the absorption of light of photon hv such that Δt = hv, causes the single electron to move from e [Fig. 5.37(b)] to
t2 level [Fig. 5.37(c)]. Accordingly, the e → t2 or e4 t 25 ⎡⎣ 2 T 2 ⎤⎦ → e 3t 26 ⎡⎣ ≡ 2E ⎤⎦ transition will result in a single band
in the electronic absorption spectrum of d9-ion in a tetrahedral field of the ligand and the same has been observed
indeed in the spectrum of cupric tetra-ammine, i.e. [Cu(NH3)4]2+ and that of [Cu(H2O)4]2+. Thus the intense blue
t2
+ 4Dq
hν
Δt Δt = hν
− 6Dq
e
t2
Ground state of a Encited state of a

Uncomplexed d 9 ion Tetrahedral complex tetrahedral complex
(a) (b) (c)
Fig. 5.37 Energy-level diagram illustrating the origin of electronic absorption band due to d-d transition in d9-tetrahedral complexes.
colour of [Cu(NH3)4]2+ is due to the single absorption band of Cu2+ falling in the red end of the spectrum, while the
pale blue colour of [Cu(H2O)4]2+ is due to the absorption band in the infrared with only a tail in the red end of the
spectrum. The blue colour of the solid salt of CuSO4.5H2O is due to the absorption band around 12000 cm−1. The
Cu2+ ion in the salt has octahedral environment of oxygen atoms. On the other hand, in anhydrous copper sulphate
2−
with a weak ligand field of 4 , the absorption band appears in the infrared and thus salt is colourless.
(b) d4 ion Cr2+ and Mn3+ ions have a d4-conFiguration. Cr2+ with an electronic structure of Ar3d4 is generated
from Cr atom having an atomic number (Z) 24 and electronic structure Ar3d54s1 by the removal of two electrons
from it, i.e. Ar3d54s1 ⎯ → Ar3d4, while three electrons are removed from the Mn atom with an atomic number
2e
(Z) 25 and electronic structure Ar 3d54s2 to form Mn3+ with an electronic structure Ar3d4, i.e. Ar3d5 4s2 ⎯−3e →
Ar3d4. The energy-level diagrams for the absorption spectrum of high-spin d4 octahedral complexes is qualita-
tively analogous to that d9 ion in the octahedral field as shown in Figs 5.35 and 5.36. In Fig. 5.36, the multiplicity
2 is to be replaced by 5. The five degenerate 3d orbitals of d4 ion are d4: d 1xy d 1yz d 1zx d 1x 2 y2
d z 2 ; d xy d 1yz d 1zx d 1x 2 y2
d 1z 2 ;
d 1xy d yz d 1zx d 1x 2 y2
d 1z 2 ; d 1xy d 1yz d zx d 1x 2 y2
d 1z 2 ; d 1xy d 1yz d 1zx d x 2 y2
d 1z 2 . The ground state spectroscopic term corresponding to
this configuration is 5D [Figs 5.38(a) and 5.36]. In an octahedral field of the ligand, the degenerate orbitals split
up into two degenerate sets: one of high energy and the other low energy. The low-energy set consists of three
degenerate T2g orbitals while the high-energy set consists of two eg orbitals; the former is stabilized by −4Dq in
contrast to latter which is destabilised by +6Dq by the octahedral field of the ligand. Accordingly, the low-energy
set of orbitals constitutes the ground state which is represented as t 23g e 1g : t 23g d 1x 2 y2
d z 2 , t 23g d x 2 y2
d 1z 2 ; the Mulliken
term corresponding to this configuration is 5Eg (Fig. 5.38b and 5.36 and the high-energy set constitutes the excited
state which is represented as t 22g e g2 : d 1xy d 1yz d zx e g2 ; d xy 1 1 2 1 1 2
x y d y z d zzxx e g ; d xy d yz d zx e g ; the Mulliken term corresponding to
this configuration is 5T2g [Figs. 5.38 (c) and 5.36]. In Fig. 5.36, the multiplicity 2 is to be replaced by 5. Thus the
absorption band will result due to the crossover of one hole from the eg level to the t2g level through d–d transition
provided Δ0 = hv, i.e. eg→t2g or t 23g e 1g [5E g ] t 22g e g2 [5T 2 g ].
eg
+ 6Dq
hν
Δ0 Δ0 = hν
− 4Dq
t2g
Ground state of a Excited state of a

Uncomplexed d 4 ion Octahedral complex octahedral complex
(a) (b) (c)
Fig. 5.38 Energy-level diagram illustrating the origin of d–d transition in octahedral complexes with d4-configuration.
As examples, aquo ion [Cr(H2O)6]2+ shows an absorption band at about 13900 cm−1 (= 719 nm). The absorp-
tion maximum falls mainly in the red region and transmission is maximum in the violet region of the spectrum.
This is sufficient to impart violet colour to the aquo ion of Cr2+. In case of [Mn(H2O)6]3+, absorption band with
maximum at 21000 cm−1 (≈476 nm) appears in the green region of the spectrum. Thus, the pink colour of aquo
ion, i.e. [Mn(H2O)6]3+, results due to the transmission of white light occurring mainly in the orange and red regions
of the spectrum.
The treatment for the electronic absorption spectrum of d4 ion in a tetrahedral field of a ligland is left an exer-
cise for the students. As a hint, the energy-level diagrams of d4 high spin tetrahedral complexes are the inverse of
that of d4 octahedral complexes but qualitatively similar to that of d9 octahedral complexes. The d–d transtition
is because of the transfer of a single hole from t2 to e level under the influence of light of photon hv provided
Δt = hv. The ground state (stabilised by −4Dq) of the system is represented as e2t2: e 2d 1xy d 1yz d zx , e 2d xy d 1yz d 1zx ,
eg ∗
+ 6Dq
hν
Δ0 Δo = h ν
− 4Dq
t2g
Uncomplexed d 6 lon Ground state of an Excited state of a

Octahedral Complex octahedral complex
of d 6 lon of d 6 lon
(a) (b) (c)
Fig. 5.39 Energy-level diagram depicting the origin of d–d transitioin in octahedral complexes with d6 electronic conFiguration.
e 2d 1xy d yz d 1zx ; the Mulliken term corresponding to it is 5T2 and the excited state (destabilised by +6Dq) is rep-
resented as e 1t 3 : e 1d 1xy , d 1yz , d 1zx ; e 1d 1xy d 1yz d 1zx ; e 1d 1xy d 1yz d 1zx , d 1xy d 1y z d 1zx , the Mulliken term corresponding to this
configuration is 5E.
(c) d6 ion The common examples of ions with a d6 electronic configuration are Fe2+ and Co3+. The electronic
structure of the Fe atom with an atomic number (Z) of 26 is Ar 3d64s2 and that of Co having an atomic number
27 is Ar 3d74s2. Fe2+ results when two electrons are removed from Fe atom, i.e Ar 3d64s2 ⎯−2e → Ar 3d6 while
Co3+ results when three electrons are expelled from the Co atom, i.e. Ar 3d74s2 ⎯ 3e → Ar 3d6. The energy-level
diagrams illustrating the origin of absorption spectrum of d6 ion in an octahedral field of a ligand are shown in
Figs 5.39 and 5.32. In Fig. 5.32, multiplicity 2 is to be replaced by 5.
In an octahedral field of the ligand, the five degenerate 3d orbitals of a d6 ion, i.e. 6 : d xy2 d 1yz d 1zx d 1x2 y 2 d z12 ,
d 1xy d yz2 d 1zx d x12 y2
d 1z 2 , d 1xy d 1yz d zx2 d 1x2 y2
d 1z 2 , d 1xy d 1yz d 1zx d x22 y2
d 1z 2 , d xy1 d yz1 d zx1 d x12 - y 2 d z22 , the ground-state spectroscopic term
corresponding to this configuration is 5D. [Fig. 5.39(a)], split into the low, i.e. t2g and high, i.e. eg energy degenerate
groups. The former group is stabilised by −4Dq by the octahedral field of the ligand and represents the ground state
of the system which is written as, t 2 g4 eg2 : d xy2 d 1yz d 1zx eg2 , d 1xy d yz2 d 1zx eg2 , d 1xy d 1yz d zx2 eg2 , the Mulliken term correspond-
ing to it is 5T2g while the latter group is destabilised by +6Dq and represents the excited state which is written as
3
e 3g : t 23g d 1x 2 y 2 d z22 ; t 23g d x22 y 2 d 1z 2 ; the Mulliken term corresponding to this configuration is 5Eg. [Figs 5.39(b) and
2g
5.32]. On absorption of light of photon hv such that Δ0 = hv, only the paired electron from the t2g[5T2g] state which
has opposite spin to all the other electrons moves to the eg[5Eg] state through d–d transition [Figs 5.39(c) and 5.32].
This d-d transition, i.e. t2g → eg or t42g e2g[ ≡5T2g] → t32g e3g[ ≡5Eg], gives to an absorption band in the electronic absorp-
tion spectrum of an octahedral complex with d6 electronic structure. Other transitions that involve inversion of spins
are forbidden and hence may give rise to weak bands in the spectrum. The arguments pertaining to Fig. 5.32 hold
good for octahedral complex of d6 ion also, i.e. the magnitude of splitting of five degenerate orbitals into two sets of
orbitals depend on the nature of the ligand and it increases as the strength of the ligand increases.
As an example, the absorption spectrum of [Fe(H2O)6]2+ in aqueous solution shows an asymmetric absorption
band with a maximum at 10,400 cm−1 (= 962 nm) that falls into the infrared with only its tail in the red region
of the spectrum. This tail in the red region is responsible for the pale green colour of the aquo ion [Fe(H2O)6]2+.
The asymmetry in the absorption band in this case is due to splitting of the energy level 5Eg into two non degen-
erate energy levels because of Jahn–Teller effect. The example of a low-spin octahedral complex of the Fe2+ is
[Fe(CN)6]4− i.e. the ferrocyanide ion. The examples of high-spin and low-spin octahedral complexes of Co3+ are
[Co(H2O)6]3+, [CoF6]3- and [Co(NH3)6]3+ with Δ0 = 23000 cm–1, [Co(CN)6]3− respectively.
The energy-level diagrams of d6 high–spin tetrahedral complexes are the inverse of that of d6-octahedral com-
plexes. The ground state is represented as e 3t 23: d x1 2 y2
d z22 t 23 ; d x2 2 y2
d 1z 2 t 23 ; the Mulliken term corresponding to it is
E and is stabilised by −6Dq while the excited state is represented as e 2 t 4 : e 2 d xy2 d yz1 d zx1 ; e 2 d xy1 d yz2 d zx1 ; e 2 d xy1 d yz1 d zx2 ; the
5
Mulliken term 5T2 originates from this configuration is destabilised by +4Dq by the tetrahedral field of the ligand.
The absorption band in the spectrum results from the transfer of only one electron without any change of spin
from e level to t2 level through d–d transition, i.e. e → t2 or e3 t32[ ≡5E] → e2 t42[ ≡5T2]. AS an example, [FeCl4]2− in
aqueous solution shows a crystal field broad absorption band around 4000 cm−1 in near IR region.
Finally, (i) the effect of strength of octahedral and tetrahedral ligand fields on the splitting of five degenerate
3d-orbitals in ions with d1, d4, d6, and d9, electronic configurations and (ii) the allowed d–d transitions respon-
sible for the single absorption band in the spectra of the high-spin octahedral and tetrahedral transition metal
complexes have been summarised by L E Orgel in a qualitative way in the form of a diagram known as the Orgel
diagram [Fig. 5.40]. The right half of the diagram is applicable to octahedral d1, d6 as well as to tetrahedral d4,
d9 high-spin complexes. The spectra of these complexes show only one band due to a d–d transition which, is
assigned as T2g→ Eg for d1, d6 octahedral and E → T2 for d4, d9 tetrahedral complexes. The left half of the diagram
describes the spectra of d1, d6 tetrahedral and of d4, d9 octahedral complexes. The spectra of these complexes also
exhibit only one band due to d–d transition which is assigned as E → T2 for d1, d6, tetrahedral and Eg → T2g for d4,
d9 octahedral complexes.
Energy 2E d 1, d 6 Oh
g
d 1, d 6 Td 2T2g
2
D
2T d 4, d 9 Td
2g
d 4, d 9 Oh 2E
g
O
Increasing Increasing
ligand ligand
field field
Fig. 5.40 Orgel combined energy-level diagram for ions with d1, d6, d4, d9 electronic configurations in an octahedral (Oh ) and a tetrahedral
(Td ) fields of a ligand. Subscript g is omitted in Td cases.
The Orgel energy-level diagram for d2-, d3-, d7-, and d8-ions in an octahedral and a tetrahedral environment of
a ligand is shown in Fig. 5.41. States between 3P and 3F are not indicated since they have different multiplicity.
Thus, transitions to these states are spin forbidden and usually not observed. The two T1g states, i.e. T1g (P) and
3
T1g(F) are slightly curved lines because they have the same symmetry and they interact with each other. This
inter-electronic repulsion lowers the energy of the lower state and increases the energy of the higher state. The
effect is much more pronounced on the left of the diagram since both the states are close in energy. In the absence
of isoelectronic repulsions, the energy states would be linear and cross each other and at the cross point, the two
electrons in one atom may have the same symmetry and energy. But according to noncrossing rule, two states of
the same symmetry can’t cross each other. Remember that in case of d7 −Oh and d8−Td complexes, the transition
lines should be drawn to the left-hand side of the cross point of 3A2g(F) and 3T1g (P) lines. This will give rise to
d 2d 7Td d 2d 7Oh
d 3d 8Oh Energy d 3d 8Td
3 3
T1(P ) Cross point A2g (F )
3P 3T (P )
1g
3T
1(F )
3T (F )
2g
3
F
3T
2(F )
3
3 T1g (F )
A2(F )
O
Increasing ligand field Increasing ligand field
Fig. 5.41 Orgel energy-level diagram for d , d , d ,and d octahedral and tetrahedral complexes. The suffix g is ommitted in tetrahedral
2 3 7 8
cases. For d3 and d8 systems, the spectroscopic terms are 4F and 4P all the Mulliken terms have a multipicity of four.
3
Tlg(F) → 3T2g(F), 3Tlg(F) → 3A2g(F), and 3Tlg(F) → 3T1g(P), transitions. For d2−Oh and d3−Td complexes, the transi-
tion lines are drawn to the right-hand side of the cross point. Accordingly, the following transitions will result:
3
Tlg(F) → 3T2g(F), 3Tlg(F) → 3T1g(P), 3Tlg(F) → 3A2g(F). If this sounds like gibberish, do not worry, you will see
some examples.
(d) d2 ion The simplest examples of a d2 ion is V3+. It is formed when three electrons are removed from the va-
nadium atom with an atomic number (Z) of 23, i.e. Ar 3d34s2 ⎯−3e → Ar 3d2. The energy-level diagram showing
the origin of absorption spectrum of d2-ion in an octahedral field of a ligand, the five degenerate 3d orbitals of a
bare ion [Fig. 5.42 (a)] split up into three low energy t2g and two high-energy eg orbitals. Two electrons occupy
two of the three t2g orbitals. [Fig. 5.42(b)] On absorption of light of photon hv such that hv = Δ0, the electrons
from t2g orbitals cross over to eg orbitals as shown in Fig. 5.42(c). The electrons in dxz or dyz orbitals require less
energy to move to the d z 2 orbital as compared to the dx2– y2 orbital. This discrepancy arises because the former
transition, i.e. dxz or dyz → d z 2 gives a d 1xy d 1z 2 arrangement in which the electron cloud of the two electrons is
eg
dx2−y 2
dz2
Δ0 hν Δ0 = hν
dyz
dxz
dxy
t2g (A) (B) (C)
Uncomplexed d 2 ion Possible distribution of Excited states of a d 2 ion

electrons in the ground in an octahedral field
state of a d 2 ion in an of a llgand
octrahedral field of a ligand
(a) (b) (c)

Fig. 5.42 Schematic diagram illustrating the origin of d–d transitions in octahedral complexes of an ion with d2-conFiguration.
uniformally spread in the X, Y, Z directions. Such a uniform spread of electron density reduces the interelectronic
repulsion compared to that of the d 1xy d 1x 2 y2
arrangement where the electrons are confined only in the (XY ) plane.
Thus the two permissible d–d transitions arising due to the transfer of an electron from t2g orbitals to eg orbit-
als are: dxz /dyz → d z 2 [Fig. 5.42(c), A] and d xy / d yz → d x 2 y2
[Fig. 5.42(c), B]. The third allowed d–d transition,
i.e. d xy d yz /d zx d yz /d yz d xy → d z 2 d x 2 − y 2 results because of the simultaneous promotion of both the electrons from
t2g orbitals to eg orbitals [Fig. 5.42(c), C]. The other possible transitions involving the reversal of spin of one of
the electrons such as dxy (↑)dxz (↑)→dz2(↓)dx2– y2(↑); dyz (↑)dxy (↑)→dz2(↑)dx2– y2(↓) are multiplicity forbidden. Such
transitions may give rise to very weak bands in the spectrum. Thus only three absorption bands, corresponding
to dxz /dyz→ dx2– y2, dyz /dxz→ dz2 and dyz dxz→ dx2– y2 dz2 transitions are expected in the electronic absorption spectrum
of a d2-ion in an octahedral field of the ligand.
The absorption spectrum of octahedral complexes of an ion with d2 configuration can also be described by
spectroscopic terms of a bare d2-ion and their splitting into Mulliken terms in an octahedral field of a ligand.
The spectroscopic states of a bare ion with d2-electronic configuration arise because of interelectronic repul-
sions between d electrons and are
Ground state: 3F.
Excited states: 3P, 1G, 1D, lS.
The transitions from the 3F ground state to the 1G, 1D or 1S states being multiplicity forbidden, will be very
weak. The only important allowed transition is 3F→3P. The energy of the 3F state for Ti2+ ,V3+ and Cr3+ free ions is
zero and that of the 3P state for the respective ions are 10600, 13250, and 15700 cm−1.
In an octahedral field of the ligand, the 3F state splits up into the triply degenerate 3T1g(F); (the origin of this
state is the orbital set t 22g d 1x y d 1y z d zx , d xxyy d 1y z d 1zzxx , d 1xxyy d y z d 1zzxx , i.e. the ground state of the system), triply degenerate
3
T2g(F) (the origin of the state is t 12 g e 1g : d 1xy d 1z 2 , d 1xz d 1x 2 y2
, d 1yz d 1x 2 y2
) and singly degenerate 3A2g(F); (the arrange-
ment e g2 : d 1z 2 d 1x 2 − y 2 produces a singly degenerate 3A2g state), Mulliken states while the 3P state is not affected by
the ligand field and transforms to the triplet 3Tlg(P) [the arrangement t 12 g e 1g ( P ) : d 1xy d 1x 2 y 2 ; d 1xz d 1z 2 ; d 1yz d 1z 2 produces
triply degenerate 3T1g(P) state] Mulliken state. The energies of these states as a function of Dq can be calculated by
evaluating the electrostatic perturbation of the ligand on the d-orbitals. Splitting of the spectroscopic states of the
bare d2 ion into Mulliken states under the, influence of an octahedral field of a ligand have been shown diagram-
matically in Fig. 5.43. Accordingly, the absorption spectra of octahedral complexes of d2-ions should consist of
three absorption bands corresponding to the d-d transitions indicated below:
Mulliken Energy Transition Transition energy

3
T1g(F) −6Dq 3
T1g(F) → 3T2g(F) 8Dq
3
T2g(F) +2Dq 3
T1g(F) → 3T1g(P) −
3
T1g(P) −
3
A2g(F) +12Dq 3
T1g(F) → 3A2g(F) 18Dq
However in practice, the absorption spectra of complexes of V3+ ion contain only two bands. As an example,
in aquo ion [V(H2O)6]3+, the absorption bands are observed at 17000 cm−1 (= 588 nm) and 24000 cm−1 (417 nm)
(Fig. 5.44). The discrepency between the theoretically expected and observed spectral data arises because of the
submersion of high energy bands corresponding to the 3Tlg(F)→ 3T1g(P) and 3T1g(F)→ 3A2g(F) transitions, i.e. the
overlapping of the high-energy bands occurs since the transitions take place close to the crossover point between
3
A2g(F) and 3Tlg(P) terms on the energy-level diagram [Fig. 5.43].
3A E = +12 Dq
2g (F )
3T
1g (P )
3 3E = +2 Dq
3P T2g (F )
Energy
3E
3T
1g (F ) E = −6 Dq
Ligand field strength

Fig. 5.43 Energy-level diagram for the spiitting of 3P and 3F spectrascopic terms of a bare ion, with d2-conFiguration into Mulliken terms in
an octahedral field of a ligand.
Further, in order to determine Δ0 we proceed as follows:

Since E(3T2g) − E(3T1g) = 2Dq − (−6Dq). = 8Dq = 17000 cm−1.
17000( ) × 10
Therefore Δ0 = 10Dq = = 2125 cm−1.
8
100 Violet Blue Green Yellow Red
Percent Absorption
50
400 500 600

Wavelength (nm)
Fig. 5.44 Ultraviolet visible spectrum of [V(H2O)6 ]3+ ion.
t2
dyz
dxz
dxy
hν
Δt Δt = hν
dx 2 − y 2
dz 2
e
(A) (B) (C)
Uncomplexed d 2 ion Distribution of electrons Excited states of a d 2 ion

in the ground state of a in a tetrahedral field of
d 2 ion in a tetrahedral a ligand
field of a ligand
(a) (b) (c)

Fig. 5.45 Schematic energy-level diagram, illustrating the origin of electronic absorption spectrum of d2-ion, in a tetrahedral field of a
ligand. Here in case (b), for convenience t2 (---) is represented as t2[≡] and e(--) as e[=]. This is so in the incoming diagrams also.
In tetrahedral complexes of a d2-ion the energy level 3P 3T (P)

1
diagrams are the inverse of that of d2-ion in an octahedral field
of the ligand as shown in Figs 5.45 and 5.46. 3T (F ); E = + 6Dq
1
On absorption of light, the electrons from e level
3
F
[Fig. 5.45(b) shifts to t2-orbitals according to the arrange- 3T (F ); E = − 2Dq
Energy
2
ments A, B and C [Fig. 5.45(c)]. The electrons in the dx2–y2
orbital require less energy to crossover to the dxy orbital 3A (F ); E = − 12Dq
than from the dz2 orbital since the former transition results 2
in lower electrons repulsion d1z2 d1xy arrangement compared Ligand field strength
to d 1x 2 y 2 d xy
1
arrangement due to the logic discussed ear- Fig. 5.46 Energy-level diagram for the splitting of 3P and 3F
lier in octahedral d2-complexes. Thus the two permissible spectroscopic terms of an ion with d2-conFiguration into Mul-
one electron d–d transitions are dz2 → dxy /dyz /dzx and dx2–y2 liken terms in a tetrahedral field of a ligand.
→ dxy/dzx /dyz. The third allowed d–d transition arising,
because of the simultaneous promotion of both the elec-
trons from e to t2 orbitals, is dz2 dx2–y2 → dxy dxz or dxz dyz or dyz dxy. The other possible transitions involving the
reversal of spin of one of the electrons such as dz2 (↑) dx2–y2 (↑) → dxy(↓)dxz(↑); dz2(↑)dx2–y2(↑) → dxy(↑) dxz(↓)
dx 2 − y 2 eg
dz 2
hν
Δ0 Δ0 = hν
dyz
dxz t2g
dxy
(A) (B) (C)

Uncomplexed d 8 ion Distribution of electrons Excited states of a d 8 ion
in the ground state of a in an octahedral field of
d 8 ion in an octahedral a ligand
field of a ligand
(a) (b) (c)
Fig. 5.47 Schematic energy-level diagram showing the origin of d–d electronic transitions of d8-ion in an octahedral field of a ligand. Open
circles represents holes.
are spin forbidden. Such transitions may give rise to very weak bands. Accordingly, three absorption bands
corresponding to the dx2–y2 → dxy /dyz /dzx, dz2 → dxy /dyz /dzx, and dx2–y2dz2 → dxydyz, transitions are expected in the
electronic spectrum of high spin tetrahedral complexes of metal ions with d2 configuration.
The description of Figure 5.46 is left as an exercise for the students. As a hint, the absorption spectra of high
spin d2-tetrahedral complexes should contain three bands corresponding to 3A2(F) →3T2(F), 3A2(F) →3T1(F) and
3
A2(F) →3T1(P), d—d transitions. In this case no overlapping of high-energy bands will take place since there is
no crossover of 3T1(P) and 3T1(F) lines. The red–purple colour of tetrahedral ferrate ion [Fe VI (O 2 )]4 + (stable in
strongly alkaline solution) results since transmission of white light occurs mainly in the red and violet regions of
the spectrum.
(a) d8 ion The d8-configuration occurs in Ni2+. This ion with electronic structure Ar 3d 8 is formed from Ni atom
with an atomic number (Z) of 28 by removing two electrons from it, i.e. A r 3d 8 4s 2 ⎯−2 e → A r 3d 8 . On complexing
with six-ligand molecules, the five 3d degenerate orbitals split into two degenerate sets: the low-energy (t2g) and
high-energy (eg) sets due to electrostatic octahedral field of a ligand. The energy-level diagram for the d8-ion in
an octahedral field is shown in Fig. 5.47. There are two holes in the eg orbitals. The d–d transitions are due to the
transfer of holes from eg to t2g orbitals under the influence
of light of photon hv. The transition which gives rise to a dz2
dxy arrangements, in which the electron cloud of electrons 3
P
3T (P )
1g
is uniformally spread in the X, Y and Z directions which
reduces the inter-electronic repulsion, is more probable as 3
T1g(F ); E = + 6Dq
compared with the dx2–y2 dxy arrangement where the electrons 3F
Energy
3T (F ); E = −2Dq
are confmed only to the XY plane. 2g
Thus, the two possible d–d transitions (dxy /dyz /dzx → dz2;
dxy /dyz /dzx → dx2–y2) arising due to the transfer of a hole from 3A (F ); E = −12Dq
2g
eg orbitals to t2g orbitals are: dz2 → dxy /dyz /dzx; dx2–y2 → dxy / Ligand field strength
dyz /dzx: The third possible d–d transition that results due to Fig. 5.48 Energy- level diagram for the splitting of 3P and 3F
the simultaneous promotion of both the electrons from the spectroscopic terms of an ion with d8 conFiguration into Mul-
tzg to eg states and simultaneously the migration of the two liken terms in an octahedral field of a ligand.
holes from eg orbitals to t2g orbitals is dz2 dx2–y2 → dxydyz or
dyzdxy or dxydxz. Accordingly, three absorption bands corre-
sponding to the dxy → dz2, or dxy → dx2–y2 and dzx,dyz→ dz2 dx2–y2 transitions are expected in the metal complexes of
d8ion. The energy- level diagram in terms of spectroscopic terms, i.e. and 3F of d8-ion and their splitting into Mul-
liken terms in an octahedral field of a ligand is presented in Fig. 5.48. The lines 3T1g(P) and 3T1g (F) do not cross
because they are of the same symmetry.
The Mulliken terms indicated in Fig. 5.48 originates from the following arrangements of t2g and eg orbitals
Mulliken term Arrangement of orbitals
3
A2g(F) t 26g e g2 : d xy2 d yz2 d zx2 d 1z 2 d 1x 2 y2
, [From Fig. 5.47(b)]
3
T2g(F) t 25g e g3 : d xy2 d yz2 d 1zx d z22 d 1x 2 y2
,
d 1xy d yz2 d zx2 d 1x 2 y2

d z22 ,
d xy2 d 1yz d zx2 d 1x 2 y2

d z22 , [From Fig. 5.47(c);(B)]
3
T1g(F) t 25g e g3 : d xy2 d yz2 d 1zx d 1z 2 d x22 y2
,
d 1xy d yz2 d zx2 d 1z 2 d x22 − y 2
1 2
d xy2 d 1yz d 2 zx d z 2 d x 2 − y 2 , [From Fig. 5.47(c);(A)]
3
T1g(P) t 24g e g4 : d xy2 d 1yz d 1zx d z22 d x22 y2
,
d 1xy d yz2 d 1zx d z22 d x22 − y 2
2 2 2
d 1xy d 1yz d zx d z 2 d x 2 − y 2 . [From Fig. 5.47(c);(C)]
According to Fig. 5.48, the three absorption bands are expected in the electronic spectra of d8–octahedral com-
plexes corresponding to the transitions indicated below.
Mulliken state Energy Transition Transition energy

3
A2g(F ) −12Dq 3
A2g(F )→ T2g(F)
3
10Dq
3
T2g(F ) − 2Dq 3
A2g(F )→ T1g(F)
3
18Dq
3
T1g(F ) +6Dq 3
A2g(F )→3T1g(P) −
3
T1g(P) −
and the same has been experimentally observed indeed.

For example, the absorption spectra of octahedral complexes of Ni2+ ion with various ligands such as H2O, NH3,
etc, i.e. [Ni(NH3)6]2+, [Ni(H2O)6]2+ shown in Fig. 5.49, conform to the theoretically expected spectra. The three
absorption bands fall in the regions 8000–13000 cm−1 (1250−769 nm), 15000–19000 cm−1 (666.6−526.3 nm) and
25000–29000 cm−1 (400–344.8 nm). The exact positions depend on the quantities Δ0 and b.
The emax of these bands is generally below 20. In the case of [Ni( H2O)6]2+ ion, the middle peak shows sign of
splitting into two peaks because of Jahn–Teller distortion. The amine of nickel [Ni(NH3)6]2+complex ion is blue
whereas the aquo ion [Ni(H2O)6]2+ is green. The absorption spectra of ions show that Δ0 is larger for ammonia
Violet Blue Green Yellow Red
[Ni (NH3)6]2+
Absorbance
[Ni (H2O)6]2+
400 600 800 1000

Wavelength (nrn)
Fig. 5.49 Electronic absorption spectra of high spin hexa-aquo and hexa-amine complexes of Ni2+ ion.
t2
dyz
dxz
dxy
Δt hν Δt = hν
dx 2 − y 2
dz 2
e (A) (B) (C)
Uncomplexed d a lon Possible distribution of Possible distribution of

electronc-in-the ground electronc in the excited
state of a d 8 lon in a state of a d 8 lon in a
tetradral field of a ligand tetrahedral field of a ligand
(a) (b) (c)

Fig. 5.50 Schematic diagram showing the origin of d–d transition in d -tetrahedral complexes. 8
than for water. For [Ni(H2O)6]2+ ion, the band maxima are at 8700, 14500 and 25300 cm−1. Thus in this case
E(3T2g ) −E(3A2g) = −2Dq −(−12Dq) = 8700 cm−l. Hence l0Dq = 8700 cm−1. For [Ni(NH3)6]2 ion, the absorption maxi-
mum corresponding to 3A2g(F) → 3T2g(F) transition lies at around 12907 cm−1Accordingly l0Dq = 12907 cm−1.
The treatment for the origin of d8 tetrahedral complexes is left an exercise for the students. As a hint, the
diagram for the origin of d-d transitions in these complexes is shown in Fig. 5.50 and the energy-level diagram
showing splitting of 3F and 3P spectroscopic states in Mulliken terms is similar to that of d2 octahedral complexes
except that E = −2Dq for 3T2(F) shown in Fig. 5.43 in a qualitative way. Two transitions due to the transfer of an
electron with antiparallel spin from e to t2 orbitals are dz2→dxy transition, gives higher interelectronic repulsion
arrangement dxydx2–y2(A) and dx2–y2→dxy transition, results in lower interelectron repulsion arrangement dz2 dxy(B).
The third possible transition is due to the simultaneous migration of two electrons with anti-parallel spins. The
Mulliken terms originates from the electronic configuration of e and t2 orbitals in Fig. 5.50.
3
T1(F) e 4t 24 : e 4d 1xy d 1yz d zx2 , e 4d xy2 d 1yz d 1zx , e 4d 1xy d yz2 d 1zx .
3
T2(F) e 3t 5 2 : e 3 (d 2 z 2 d 1x 2 − y 2 )d xy2y d 1yz 2 3 1 2 2 3 2 2 1
y d zx , e d xy d yz d zx , e d xy d yz d zx .
3
T1(P) e 3t 25 : e 3 (d 1z 2 d x22 − y 2 )d xy2y d 1yz 2 3 1 2 2 3 2 2 1
y d zx , e d xy d yz d zx , e d xy d yz d zx .
3
A2(F) e 2t 6 2 : e 2d xy2 d yz2 d zx2 .
The three-spin allowed transitions in Td complexes of Ni2+ are 3T1(F) → 3T2 (F); 3T1 (F) → 3A2 (F); and
3
T1 (F) → 3T1(P). For example, the absorption spectrum of [{(C6H5)3AsO}NiCl2] in benzene shows two bands at
7800 cm−1 and at 14000–16000 cm−1. The complex has distorted tetrahedral structure. The two absorption bands
are attributed to the following transitions:
3
T1 (F) → 3A2 (F), 7800 cm−1
3
T1 (F) → 3T1 (P), 16000 cm−1
Accordingly,
E (3A2) − E(3T1) = 12Dq −(−6Dq) = 18Dq = 7800 cm−1.
−1
7800( ) × 10
Hence, 10 Dq = = 4333 cm −1
18
The frequency of the band corresponding to the transition of lowest energy, i.e. 3T1 (F) → 3T2 (F)
7800( −11 ) × 8
= = 3467cm −1. But this is not observed in the electronic absorption spectrum since it falls in the
18
infrared region. Note that the band of lowest energy is often masked by absorption of either the organic part of the
molecule or the solvent. The other example of d8 tetrahedral complex is that of [Ni(Cl)4]2− in CH3NO2. It shows an
absorption band at 608 nm (16450 cm−1) as a shoulder (e =10) and two equally intense bands (e = 35) at 654 nm
(15290 cm−1) and 705 nm (14190 cm−1). Thus the green colour of this complex is attributed to absorption in the
yellow–red regions and hence transmission in green region of spectrum of white light. The first cationic tetrahe-
dral complex of Ni2+ was with phosphoramide, i.e. O–P–[N(CH3)2]3 abbreviated as HMPA. The similarity of the
spectrum of [Ni(Cl)4]2− was used to establish tetrahedral structure.
(f) d3 ion Cr3+ with an electronic configuration of Ar 3d3 results when three electrons are removed from Cr atom
−3e
with an electronic configuration Ar 3d5 4s1, i.e. Ar 3d5 4s1 ⎯ → Ar3d3. The origin of d-d transitions in d3octa-
hedral complexes has been depicted in Fig. 5.51. According to this diagram, three absorption bands are expected
corresponding to the to the two–one electron transition, i.e. d xy or d yz or d xz → dz
d 2 and d yz or d zx or d xy → d x 2 - y 2
and one two electron transition, i.e. d xy d yz or d yz d xz or d xy d zx → dz 2 d x 2 y.
The energy-level diagram for transition/s in metal complexes with d3 ion in terms of its spectroscopic terms and
their transformation into Mulliken terms under the influence of octahedral field of an ligand is similar to that of
d8 octahedral complexes except that the multiplicity 3 is to be replaced by 4 and the origin of Mulliken terms will
also be different [Fig. 5.48]. The origin of these terms,as worked out from Fig. 5.51, is expressed here.
eg
x −y 2
d 2
dz 2
Δ0 hν Δο = hν
dyz
dxz
dxy
t2g (A) (B) (C)
Uncomplexed d 3 ion Distribution of Electrons Excited States of a d 3 ion

in the Ground State of a in an Octahedral Field of
d 3 ion in an Octahedral a Ligand
Field of a Ligand
(a) (b) (c)

Fig. 5.51 Schematic diagram illustrating the origin of electronic transitions in octahedral complexes of d3 ion.
4
A2g (F) t32g : d1xy d1yz d1zx
4
T2g (F) t 22g e 1g : d 1xyy d 1yz 1 1 1 1 1 1 1 1
y d zx e g (d z 2 d x 2 − y 2 ), d xy d yz d zx e g , d x y d y z d zx
z eg .
4
T1g (F) t 22g e 1g : d 1xyy d 1yz 1 1 1 1 1 1 1 1
y d zx e g (d z 2 d x 2 − y 2 ), d xy d yz d zx e g , d x y d y z d zx
z eg
4
T1g (P) t 12 g e g2 : d 1xy d yz d zx e g2 , d x y d 1y z d zx e g2 , d xy d yz d 1zx e g2
Accordingly, the following transitions are possible in the d3 octahedral complexes:
Mulliken Term Energy Transition Transition energy

4
A2g(F) −12Dq 4
A2g(F) → 4T2g(F) +10Dq
4
T2g(F) −2Dq 4
A2g(F) → 4T1g(F) +18Dq
4
T1g(F) +6Dq 4
A2g(F) → 4T1g(P) −
4
T1g(P) −
Thus, the electronic absorption spectra of d3 octahedral complexes should contain three bands and the same has
been observed indeed. For example, the observed absorption spectrum of [Cr(F)6]3− ion exhibits three bands at
about 34400 cm−1 ( ≈ 290.7 nm), 22700 cm−1 ( ≈ 440.5 nm) and 14900 cm−1 (≈ 671.1nm) while that of [Cr(H2O)6]3+
ion shows three bands at 37879 cm−1 (≈ 264 nm), 24570 cm−1 (407 nm) and 17241 cm−1 (≈ 580 nm). The Δ0 values
for these complexes can be worked out as follows:
For [Cr(H2O)6] 3+ ion in an aqueous medium, we have E (4T2g) − E(4A2g) = −2Dq − (−12Dq) = 17241 cm−1, hence
Δ0 = 10Dq = 17241 cm−1 and for [Cr(F6)]3− we have E (4T2g) −E (4A2g) = − 2Dq − (−12 Dq) = 14,900 cm−1, hence
Δ0 = 10 Dq = 17,241 cm−1 and for [Cr(F)6]3− we have E(4T2g) − E(4A2g) = − 2Dq − (−12 Dq) = 14,900 cm−1, hence
Δ0 = 10 Dq = 14,900 cm−1. Accordingly, H2O is a strong ligand as compared to F− as expected on the basis of
spectrochemical series.
The bluish green colour of Cr3+ octahedral complexes results because blue/green components of white
light remain relatively unabsorbed. Another example of d3 ion is V2+. [V(H2O)6]2+ aquo ion shows absorption
bands at 358, 541 and 910 nm. In this case, the violet colour of light remains relatively unabsorbed.For this
complex, 10Dq = Δ0 = 10990 cm −1. Thus Cr3+ forms a stronger octahedral complex with water relative to V2+
as expected.
For d3 tetrahedral complexes, the diagram showing the origin of d–d transition/s, i.e. transfer of an electron/s
from e to t2 orbitals, is shown in Fig. 5.52 and the energy-level diagram showing the splitting of 4F and 4P spec-
troscopic terms into Mulliken terms in a tetrahedral field of a ligand is similar to that of d2 octahedral complexes
(Fig. 5.43) except the multiplicity and the orbital arrangements giving rise to Mulliken terms. The origin of
Mulliken terms as worked out from Fig. 5.52 is listed here.
t2
dyz
dxz
dxy
hν
Δt Δt = hν
dx 2− y 2
dz 2
e
(A) (B) (C)
Uncomplexed d 3 ion Possible Distribution of Excited States of a d 3 Ion

Electrons in the Ground in a Tetrahedral Field of a
State of a d 3 ion in an Ligand
Tetrahedral field of a Ligand
(a) (b) (c)

Fig. 5.52 Diagram illustrating the origin of d–d transitions in d3 ion in a tetrahedral field of a ligand.
4
T1(F) e 2t 12 : e 2d 1xy d yz d zx , e 2d xy d 1yz d zx , e 2d xy d yz d 1zx
4
T2(F) e 1t 22 : e 1 (d 1z 2 d x 2 − y 2 )d 1xyy d 1yz 1 1 1 1 1 1
y d zx , e d xy d yz d zx , e d xy d yz d zx
4
T1(P) e 1t 22 : e 1 (d 1x 2 y2
d z 2 )d 1xyy d 1yz 1 1 1 1 1 1
y d zx , e d xy d yz d zx , e d xy d yz d zx
4
A2 t 23 : d 1xy d 1yz d 1zx
Accordingly, the absorption spectra of d3-tetrahedral complexes should give rise to three bands corresponding
to the d–d transition/s indicated below.
Mulliken term Energy Transition Transition energy

4
T1 (F) − 6 Dq 4
T1 (F) → T2 (F)
4
+ 8Dq
4
T2 (F) + 2Dq 4
T1 (F) → T1 (P)
4
−
4
T1 (P) − − −
4
A2 12 Dq 4
T1 (F) → 4A2 18 Dq
(g) d7-ion Co2+ with an electronic configuration Ar 3d7 is an important example of d7-ion. It is produced by
ejecting two electrons from the cobalt atom (Z = 27) having an electronic configuration of Ar3d7 4s2, i.e. Ar 3d7
4s2 ⎯−2e → Ar3d7. The origin of d–d transitions in d7-octahedral complexes has been shown in Fig. 5.53. Two–one
electron transitions of the type dyz→dx2–y2 and dxz→dz2 occur due to the migration of an electron without change
of spin from t2g to eg orbitals. The third possible transition of the type dxydyz→dz2dx2–y2 is due to the simultaneous
crossover of two electrons without change of spin from t2g to eg orbitals. Accordingly, the absorption spectra of
d7-octahedral complexes should contain three absorption bands.
eg
dx2−y2
dz2
Δ0 = hν
Δ0 hν
dxy
dyz
dzx
Uncomplexed d 7 ion
Possible Distribution
of Electrons in the (A) (B) (C)
Ground State of d 7
ion Octahedral Field Excited States of d 7
of a Ligand Ion ion in a Octahedral
Field of a Ligand
(a) (b) (c)
Fig. 5.53 Schematic diagram depicting the origin of d–d transition in d7-octahedral complexes.
Further the energy-level diagram showing the splitting of 4F and 4P spectroscopic terms of d7-ion into Mulliken
terms under the influence of an octahedral field of a ligand is similar (except the multiplicity and arrangement of
orbitals due to d–d transition) to that of d2-octahedral complexes (Fig. 5.43). The arrangements of orbitals that
give rise to Mulliken terms as deduced from Fig. 5.53 are indicated below.
4
T1g(F) t 25g e g2 : d 1xy d yz2 d zx2 e g2 , d x2y d 1y z d z2x e g2 , d xy2 d yz2 d 1zx e g2 .
4
T2g(F) t 24g e g3 : d 1xyy d yz
2 1 3 1 2 2 1 1 3 1 1
y d zx e g (d z 2 d x 2 − y 2 ), d xy d yz d zx e g , d x y d yz
2 3
y d zx e g .
4
T1g(P) t 24g e g3 : d 1xyy d 1yz 2 3 2 1 2 1 1 3 1 2 1 3
y d zx e g (d z 2 d x 2 − y 2 ), d xy d yz d zx e g , d x y d yz
y d zx e g .
4
A2g(F) t 23g e g4 : d 1xy d 1yz d 1zx e g4 .
According to the energy-level diagram of d7 octahedral complexes, three absorption bands are expected to
correspond to the following transitions:
Mullikan term Energy Transition Transition Energy

4 4 4
T1g
1 (F ) −6Dq T1g
1 ( F ) → T 2g
2 (F ) 8Dq
4 4 4
T 2g
2 (F ) +2Dq T1g
1 ( F ) → T1g
1 ( P) −
4
T1g
1 ( P) −
4 4 4
A 2g
2 (F ) +12Dq T1g
1 ( F ) → A 2g
2 (F ) 18Dq
and the same has been observed indeed. For examples, [Co(F)6]4−, [Co(H2O)6]2+, etc.; the absorption spectrum of
the former shows three bands at 7150 cm−1 (1398.3 nm), 15200 cm−1 (658 nm) and 19200 cm−1 (521 nm) while that
of latter (Fig. 5.54) exhibits a well-resolved low-intensity band at about 8000 cm−1 (1250 nm) and a high- intensity
multiple absorption band consisting of three overlapping peaks at about 20,000 cm−1. The three peaks in the multiplet
Molar absorbance
25000 20000 15000 10000 5000
Frequency (cm−1)
Fig. 5.54 Electronic absorption spectrum of [Co(H2O)6]2+ in aqueous medium.
are at about 16000 cm−1 (625 nm), 19417 cm−1 (515 nm) and 21598 cm−1 (463 nm). The closeness of the peaks at
19417 and 21598 cm−1 indicates that the ligand field strength of water causes transitions close to the crossover point
of 4 T1g
1 ( P ) and
4
A 2 g terms. Thus, the three absorption bands at 8000, 19417 and 21598 cm−1 correspond to the
1 ( F ) → T 2g
three electronic transitions, viz., 4 T1g 1 ( F ) → A 2g 1 ( F ) → T1g
4 4 4 4 4
2 ( F ) , T1g 2 ( F ) and T1g 1 ( P ) and respectively.
−1
The extra band at about 16000 cm is attributed either to spin-orbit coupling effects or transitions to doublet states.
Further, in order to find Δ0, i.e. 10Dq; we write as
E ⎡⎣ 4 T 2 g ( F ) ⎤⎦ E ⎡⎣ 4 T g ( F ) ⎤⎦ = +2Dq − (−6Dq) = 8Dq = 8000 cm−1
−1
8000( ) × 10
Hence 10 Dq = 0000 cm −1 .
= 10000
8
The pinkish violet colour of octahedral complexes of Co2+ ions is attributed to the fact that transmission of
white light occurs in the red region of the spectrum, i.e. > 15,000 cm−1 < 10,000 cm−1.
dxy t2
dyz
dzx
hν
Δt Δt = hν
e
dz 2
dx 2−y 2
Uncomplexed d 7 ion
Possible Distribution
of Electrons in the
Ground State of d 7 (A) (B) (C)
ion in tetrahedral Field Excited States of d 7
of a Ligand ion ion in a tetrahedral
Field of a Ligand
(a) (b) (c)
Fig. 5.55 Schematic diagram illustrating the origin of d–d transitions of d7-ion in a tetrahedral field of a ligand.
7150( 1
) × 10
Further, for [Co(F)6]4−, 8Dq = 7150 cm−1, hence Δ0 = 10 Dq = = 8938 cm −1 .
8
Since, Δ0 (H2O) > Δ0 (F−), therefore H2O is a strong ligand than F− as expected according to spectrochemi-
cal series. The three non-overlapping bands of [Co(F)6]4− at 7150, 15200 and 19200 cm−1 correspond to the
1 ( F ) → T 2g 1 ( F ) → A 2g 1 ( F ) → T1g
4 4 4 4 4 4
T1g 2 ( F ) , T1g 2 ( F ) and T1g 1 ( P ) transitions respectively.
The splitting of five degenerate orbitals of d -ion into e and t2 sets of orbitals under the influence of tetrahe-
7
dral environment of a ligand as well as the origin of two–one electron transitions, i.e. d z 2 → dxz and d x 2 y 2 → dyz
and one–two electron transition, i.e. d z 2 d x 2 − y 2 → d xy d yz that take place due to the transfer of electron/s without
change of spin from e to t2 orbitals is shown in Fig. 5.55. The energy-level diagram showing the splitting of
spectroscopic terms 4F and 4P of d7-ion in tetrahedral environment into Mulliken terms (with multiplicity of 4)
is similar to that of d2-tetrahedral complexes (Fig. 5.46), except the multiplicity, and the arrangements of orbitals
giving rise to Mulliken terms. The origin of Mulliken terms is indicated below.
4
A 2 (F ) e 4t 23 : d z22 d x22 − y 2 d 1xy d 1yz d 1zx
4
(
T 2 ( F ) e 3t 24 : e 3 d 1z 2 d x22 y2 )d 2
xy d 1yz d 1zx , e 3 d 1xy d yz2 d 1zx , e 3 d 1xy 1 2
x d yz d zx .
4
(
T1 ( F ) e 3t 24 : e 3 d 1z 2 d x22 y2 )d 2
xy d 1yz d 1zx , e 3 d 1xy d yz2 d 1zx , e 3 d 1xy 1 2
x d yz d zx .
4
T1 ( P ) e 2t 25 : e 2d 1xy d yz2 d zx2 , e 2 d xy2 d 1yz d zx2 , e 2 d xy2 d yz2 d 1zx .
Thus, according to energy-level diagram of d7-ion in tetrahedral environment (Fig. 5.46), the absorptions spec-
tra of d7-tetrahedral complexes should consist of three absorption bands corresponding to the transitions indicated
below.
Mulliken term Energy Transition Transition energy
4 4
A 2 (F ) −12Dq A 2 (F ) → 4T2 (F ) +10Dq
4 4
T2 (F ) −2Dq A 2 ( F ) → 4 T1 ( F ) +18Dq
4 4
T1 ( F ) +6Dq A 2 ( F ) → 4 T1 ( P ) −
4
T1 ( P ) −
For example, the electronic absorption spectrum of [Co(Cl)4]2− [Fig. 5.56] shows two bands at 15000 cm−1
⎛ 666.5 nm;⎞ and 5800 cm−1 ⎛1724 nm;⎞ , and are assigned to 4 A ( F ) → 4 T ( P ) and 4 A ( F ) → 4 T ( F ) tran-
⎝ visible ⎠ ⎝ near IR ⎠ 2 1 2 1
sitions respectively. The band of lowest energy is expected in the IR region and will correspond to transition
4
A 2 ( F ) → 4 T 2 ( F ) . In order to determine 10Dq, we write, E ⎡⎣ 4 T1 ( F ) ⎤⎦ E ⎡⎣ 4 A 2 ( F ) ⎤⎦ = +6Dq − (−12Dq) = 18Dq
= 5800 cm−1.
Molar Absorbance
25000 20000 15000 10000 5000

Frequency (cm−1)
Fig. 5.56 Electronic absorption spectrum of [Co(Cl)4 ]2– in aqueous medium.
5800( 1
) × 10
Hence, 10Dq = = 3222 cm −1 .
18
Thus, the band of lowest energy should appear at about 3200 cm−1.
The band at 15000 cm−1 (666.5 nm) falls in the red and green regions of the spectrum. The characteristic blue
component of white light remains unabsorbed and hence imparts deep blue colour to [Co(Cl)4]2−, the high-spin
tetrahedral complex of Co2+ ion.
(h) d5-ion Mn2+, Fe3+ and Co4+ are some of the examples of ions with an electronic configuration of Ar 3d5.
They are produced by the removal of two, three and four electrons from their respective atoms. In an octahedral
field of a weak ligand, the five degenerate 3d orbitals split into two degenerate set of orbitals, the lower-energy set
and the higher-energy set: the former comprises three t2g orbitals while the latter contains two eg orbitals. Three
unpaired electrons with parallel spins reside in the lower set while the remaining two unpaired electrons reside
in the upper set. The t2g orbitals stabilises the system by −12 Dq while the eg orbitals destabilises the system by
12Dq. On absorption of light photon hv, such that Δ0 = hv, any d−d electron transition between two energy states
will occur with a reversal of spins. Hence like all other spin forbidden transitions, the absorption bands arising
due to the transitions involving reversal of only one spin will be extremely weak, while involving reversal of two
spins will be too weak to be observed and can be ignored.
The possible spectroscopic energy states for an ion with d5 configuration are
Ground state 6
S
Eleven excited states G , F, D, P, I, H, G, F, D, 2P, 2S.
4 4 4 4 2 2 2 2 2
The order of energy of spectroscopic states with a multiplicity of four is G < F < D < P.
3
The four excited states with a multiplicity of four (2 × + 1 = 4) involve only reversal of one spin while the
1 2
states with a multiplicity of two (2 × + 1 = 2) involve reversal of two spins. Thus, the transitions from the
2
ground state 6S, to Mulliken states with a multiplicity of two are spin forbidden and can be ignored. The only
important transitions are from 6S to 4G, 4F, 4D and 4P states respectively. In an octahedral field of a ligand, the
ground and excited states split into Mulliken states as reported in Table 5.20. Accordingly, ten extremely weak
bands are expected in the absorption spectrum of a d5-ion in an octahedral field of a ligand, for example in the
electronic absorption spectrum of [Mn(H2O)6]2+ shown in Fig. 5.57, extremely weak bands are observed, the
molar extinction coefficient e being about 0.02–0.03 lit mole−1 cm−1, which is approximately 250–333 times small
compared to spin-allowed transitions. Some of the bands are sharp while other are broad. Spin-allowed bands are
invariably broad. The positions of the bands are at about 18900 cm−1 (≈ 529 nm), 24970 cm−1 (≈ 402 nm), 25300
cm−1 (≈ 395 nm), 29700 cm−1 (≈ 337 nm), 23100 cm−1 (≈ 433 nm) and 28000 cm−1 (≈ 357 nm), i.e. the absorp-
tion region falls mainly in the violet, blue and green regions of the spectrum of white light, which is sufficient to
Table 5.20 The transformation of spectroscopic states of d5-ion
into Mulliken states in an octahedral field of a ligand.
Spectroscopic states Mulliken states
6
6
S A 1g
4
4
G T1g 4T2 g , 4 Eg 4
A1g
4
4
P T1g
1
4
4
D T2g 4 Eg
4
4
F A 2 g 4T11gg , 4T 2 g
4
A1g
4E (4G)
0.03 g
Molar absorbance
4
0.02 Eg(4D)
4
T1g(4P )
4T (4G)
1g 4T (4G)
4 2g
T2g(4G)
0.01
20000 25000 35000

Frequency (cm−1)
Fig. 5.57 Electronic absorption spectrum of [Mn(H2O)6]2+.

4T
2g
4T
1g
4F
4
40,000 A2g
4
T1g
4D 4
Eg
Energy (cm−1)
30,000 4T
4P 2g
4 4A
4G Eg 1g
4T
20,000 2g
4T
1g
10,000
6
S 6A
1g
500 1000
Ligand Field Strength Dq (cm−1)
Fig. 5.58 The Orgel energy-level diagram for Mn2+ ion in an octahedral field of water ligand.
impart pale pink colour to most of the Mn(II) salts. [MnF6]4− and [FeF6]3− are the other examples of octahedral
complexes of ions with d5-configuration. Iron(III) alum has pale violet colour.
The energy of the Mulliken states, corresponding to the respective spectroscopic states, increases from left-
hand side to right-hand side.
The Orgel energy- level diagram of Mn2+ ion in the octahedral field of water has been shown in Fig. 5.58. The
Orgel diagram reveals that 6S state is not split and transforms to 6 A 1g state which is drawn along the horizontal
axis. The 4 E g G ),4 A 11gg , 4 E g ( D ) and 4 A 2g
2 (F ) terms also appear as horizontal lines on the diagram indicating that
their energies are also independent of the crystal field of the ligand. Since the ligand in a complex oscillates about
mean position, so the crystal field strength 10Dq varies about a mean value. Thus, the energy for a particular tran-
sition varies about a mean value. Hence, the absorption peaks are broad. The degree of broadening of the peaks
is related to the slope of the lines on the Orgel diagram. Transitions between states of zero slopes, i.e. horizontal
lines, will give rise to intense bands and any deviation of excited state from zero slope will contribute to broad-
ening of the sharp band. Accordingly, the transitions from the 6 A 1g state to 4 E g G ), 4 A 11gg , 4 E g (D D 4
A 2g (F )
4 4
states respectively will yield sharp bands while to T1g (G ), T 22gg (G ) states will give broad bands.
Accordingly, the bands are assigned as follows.
The Orgel energy-level diagram for tetrahedral complexes of
d -ion is inverse of that of d5-ion is an octahedral field of the
5
ligand. According to Laporte selection rule, tetrahedral com- Transition Band positions (cm−1)
plexes such as [MnIIBr4]2− lacks a centre of symmetry, so mixing
of p and d orbitals can take place. This is true for all tetrahedral 6
A 1g → 4 T1g
1 18900
complexes. For example, [CoCl4] is deep blue is colour as com-
2−
pared with [Co(H2O)6]2+ ion. 6

A 1g → 4 E g (G )
The Orgel energy-level diagrams are no doubt useful for 24970 and 25300
6
interpreting the absorption spectra of high-spin complexes but A 1g → 4 A 1g
they are applicable only for spin-allowed transitions when the
6
number of observed bands is greater than or equal to the number A 1g → 4 E g ( D ) 29700
of empirical quantities: crystal field splitting Dq, modified Racah
6
parameter B′ and bending constant x. A 1g → 4 T 2g
2 (G ) 23100
The low-spin complexes can also be expressed on an Orgel
energy-level diagram but commonly Tanabe–Sugano diagrams 6
A 1g → 4 T 2g
2 (D ) 28000
are used for interpretation of low-spin as well as high- spin
complexes. Due to space problem, these diagrams are not reported here. They are well documented in standard
textbooks on inorganic spectroscopy and in the literature: Y Tanabe and S Sugano, J. Phys. Soc. Japan, 9(1964)
753, 766; RS Drago, Physical Methods in Inorganic Chemistry, Chapman and Hall Ltd. London, 1965.
5.18 RACAH PARAMETERS

The various Mulliken’s energy states of the same multiplicity corresponding to the high spin, low and high energy
spectroscopic terms of free ions having d2 , d3 , d7 and d8 configurations under the influence of octahedral or tetra-
hedral field of ligand are not linear as is evident from the Orgel energy-level diagrams. This nonlinear behaviour
of the various energy levels of the same multiplicity in the Orgel energy-level diagram arises because of inter-
electronic repulsion of d-electrons, also called configuration interaction, between them. The electron–electron
repulsion lowers the energy of the lower state and increases the energy of the higher state by the same amount
called the bending constant x. The nonlinear behaviour is much more pronounced in the regions where the states
are close in energy. In absence of isoelectronic repulsions, the energy states would be linear and cross each other,
and at the cross point, the two electrons in one atom may have the same symmetry and energy. But according to
noncrossing rule, two states of the same symmetry cannot cross each other, i.e. two states of different symmetry
cross each other, e.g. 3 A 2 g d 3T1g ( P ). The magnitude of inter-electron repulsion between various states in the
gaseous ion is measured in terms of parameters called Racah parameters and are designated as B and C. For
states of different multiplicity, both the B and C parameters are necessary to explain the positions of bands in the
spectrum. While in many cases, for terms of same multiplicity, only the quantity B suffices. The energy difference
between spectroscopic terms of highest spin multiplicity is expressed as ‘nB’ where n is an integer. Both n and B
vary for different ions. For most transition metal ions, B is a about 700–1000 cm−1 and C is approximately four
times B, i.e. C ≈ 4B. The values of B (in cm−1) for bare transition metal ions are given in Table 5.21.
The effective value of the Racah interelectronic repulsion term in the complex is denoted by B′ and is always
less than that for free ion value because of covalent character between metal ion and ligand in the complex rather
Table 5.21 The values of Racah parameter

B (in cm−1) for virgin transition metal ions.
Metal M2+ M3+
Ti 695 −
V 755 861
Cr 810 918
Mn 860 965
Fe 917 1015
Co 97 1 1065
Ni 1030 1115
than only electrostatic interaction where the ligands are considered as point charges and there is no covalency in
the metal–ligand bond. The covalence effect, also called nephelauxetic effect, causes the delocalisation of elec-
tron density on metal ion on to ligand and thus reduction in the electron–electron repulsion of d-electrons in the
complexed metal ion compared to the repulsions in the gaseous state. This causes a decrease in the energy differ-
ence between two spectroscopic terms of same multiplicity in the complex relative to that in the gaseous ion. A
percentage lowering of the upper spectroscopic state in the complex compared to the energy of the upper state in
the free gaseous ion is given by
( − ′)
β0 = × 100 (5.61)
B
where B equals the energy value of upper spectroscopic state in the free gaseous ion and B′ is the experimental
energy value of upper state in the complex.
Most often the quantity b is used instead of b° and is defined as
B′
β= (5.62)
B
Here, b is called the nephelauxetic ratio. b decreases as delocalisation increases and is always less than unity.
It usually varies from 0.7 to 0.9.
5.18.1 Determination of Dq and b for Transition-Metal Ions in Octahedral

and Tetrahedral Fields
The magnitude of Dq depends upon a number of factors and the contribution from each factor depends upon the
nature of the transition-metal complex. The various factors which contribute to the magnitude of Dq are perturbation
caused by electrostatic interactions, the M−L s bond, the L−M Pp bond, the M−L Pp bond and metal electron–ligand
electron–electron repulsions. Information regarding the metal ion–ligand interaction can be obtained from the eval-
uation of Dq and b. Theoretical expressions relating the energies of the various Mulliken states of transition-metal
ions to the Dq value of the ligand in octahedral or tetrahedral field can be employed to determine b and Dq provided
the spectral data on the complexes are available.
(a) M2+ Ions with d8 Configuration in an Octahedral Field. The energies E (in cm−1) of the various Mulliken
states for M ions with d8 configuration relative to the spherical field are related to Dq, the strength of the ligand
2+
field, on the assumption that the ligands behave only as point charges or point dipoles, by the expressions.
E = − 2Dq for 3
T2g (5.63)
E = − 12Dq for 3
A2g (5.64)
6Dq p − 16(Dq)2 + [−6Dq − p] E + E2 = 0 for 3T1g(F) and 3Tlg(P) (5.65)
The roots of the quadratic Eq. (5.65) will correspond to the energies of 3T1g(F) and 3T1g(P) Mulliken states.
Here, p is the energy of higher energy spectroscopic term, i.e. 3P state or 3T1g(P) Mulliken state.
For spin-free octahedral M2+ complex irrespective of the nature of the ligand; the energy difference between the
first two states, i.e. 3T2g and 3A2g is 10Dq [i.e. 3T2g − 3A2g = −2D q − (−12Dq) = 10Dq]
The energy difference between two spectroscopic terms of Ni2+ gaseous ion is 15840 cm−1, i.e. E(3P) − E(3F)
= 15B = 15840 cm−1. The above expressions hold good perfectly for an Ni2+ ion. The energy of the upper spectro-
scopic term 3P0, i.e. E(3P), i.e. E[3T1g(P)] is obtained from the energy of the transition 3A2g → 3T1g (P).
Substituting the values of E(3P) and Dq into the quadratic equation, we obtain the value of p, the energy differ-
ence between the two spectroscopic terms of Ni2+ ion in an octahedral field of the ligand, i.e. p = 15B′.
Once B and B′ are known, the values of b° and b are determined from Eqs (5.61) and (5.62) respectively.
In order to determine the position of the bands corresponding to the [3A2g →3T1g (F)] and [3A2g → 3T1g(P)] transi-
tions, substitute the value of p for the complexed ion and Dq into the quadratic expression (5.65). The values of
two roots of the equation will give the energies of the 3T1g (F) and 3T1g (P) states. Using these values, one can
obtain the position of the above said two bands in the spectrum. For other metal ions, these expressions become
complicated due to spin-orbit (so.) coupling. so. coupling splits the three t2g orbitals as shown in Fig. 5.58. It
also lowers the energy of ground state and lowering depends upon the magnitude of so. coupling. This lowering
because of so. coupling affects the energies of the spectral bands also. Thus, the contribution of Δso. to Δt value
should be taken into account in order to have accurate values of Dq and b. Hence, so coupling in the excited state
does not create any problem since transitions in both the split levels often occur and the average of the two ener-
gies can be considered. When the ground state splits, only the lower level is populated. Thus accurate values of
Dq and b can be obtained for ions even without taking into consideration Δso., the contribution due to so. coupling
to Δt, in which the ground state is A or E, e.g Ni2+. Jahn–Teller distortions also affect the energies of the levels and
hence the values of Dq and b. So in order to improve upon the values of Dq and b, contribution from this effect
to Δt should also be considered.
Note that complications due to so. coupling and Jahn–Teller distortion are insignificant, in case of Ni2+, Mn2+
(high spin), Co3+ (low spin) and Cr3+ octahedral complexes, thus accurate values of Dq and b can be obtained from
spectral data. Ti3+ has only minor complications from so. coupling and Jahn–Teller distortions. The magnitude of
splitting because of so. coupling becomes appreciable in tetrahedral complexes and tends to that of tetrahedral
4
crystal field splitting, i.e. Δso. → Δt = Δ 0 . Thus, in such cases contributions due to so. coupling to Δ0 must be
9
considered to obtain accurate values of Dq and b from positions of the absorption bands in the spectrum.
Problem 5.18: Calculate the values of Dq, b, b°, the theoretical energies corresponding to the 3A2g →3T1g (F)
and 3A2g →3T1g (P) transitions for Ni2+ octahedral complexes from the spectral data given below.
Variation number Ligand Absorption maxima (cm−1)

1 H2O 8500 15400 (13495) 26000
2 NH3 10750 17500 28200
3 HC(O) N(CH3)2 8503 13605 (14900) 25000
The energy difference between the two spectroscopic terms for gaseous Ni2+ ion is 15840 cm−1.
Solution The electronic structure of Ni2+ ion is 3d8. With the aid of energy diagram for 3d8 configuration, the
bands are assigned to the various transitions as under.
Ligand 3
A2g → 3T2g 3
A2g → 3T1g(F) 3
A2g → 3T1g(P)
H 2O 8500 15400(13495) 26000

NH3 10750 17500 282000
HC(O)N(CH3)2 N(CH3)2 8503 13605(14900) 25000
The absorption data is in cm−1.

H2O: The following steps are followed to obtain the results
(i) The spectroscopic terms for Ni2+ gaseous ion are 3P and 3F. The energy difference between two spectro-
scopic terms is 15840 cm−1, i.e.
E(3P) − E(3F) = 15 b = 15840 cm−1.
(ii) Dq is determined from the energy of the lowest energy transition 3A2g → 3T2g.
E (3T2g) − E (3A2g) = 8500 m−1 = 10 Dq
8500 cm −1
or Dq = = 850 cm −1 .
10
(iii) The energy of the upper spectroscopic term E(3P) is obtained from the 3A2g → 3T1g (P). transition as follows:
E[3T1g (P)] − E [3A2g] = 26000 cm−1
Since 3
A2g = −12 Dq
Therefore, E [3T1g (P)] = 26000 cm−1 − 12 Dq = 26000 cm−1 − 12 × 850 cm−1
=15800 cm−1.
(iv) On substituting the value of E(3P) = 15800 cm−1 and Dq = 850 cm−1 into Eq. (5.65), we get
15.75 × 10 7 cm 1
p= = 14719.6 cm −1
10700
Thus, the energy difference between two spectroscopic terms when the ion is under the octahedral field of H2O is
p = 15 b = 14719.6 cm−1.
(v) From Eq. (5.62), we obtain
14719.6 cm −1
β= = 0.929.
15840 cm −1
and from Eq. (5.61), we obtain
⎡15840 cm −1 − 14719.6 cm −1 ⎤
β0 = ⎢ ⎥ × 100 = 7.07 per
p cent
⎣ 15840 cm −1 ⎦
(vi) In order to obtain the energy of the 3T1g(P) and 3Tlg(F) states, put the values of p = 14719.6 cm−1 and Dq = 850
cm−1 into Eq. (5.65). On simplification, we arrive at the following equation.
E2 − 1.9819 × 104 E + 6.350 × 107 = 0
Solving for E, we obtain
E=
(1.9819 ± 1.1777)104 = 15798 and 4021 cm −1
2
Thus the energy of 3T1g (P) state = 15798 cm−1 and of 3T1g(F) state = 4021 cm−1.
(vii) The position of the absorption bands corresponding to the 3A2g → 3T1g (F) and 3A2g → 3T1g(P), transitions will be
E = [3T1g (P)] − E[3A2g] = 15798 cm−1 + 12 Dq
= 15798 cm−1 × 12 × 850 cm−1 = 25998 cm−1
E[3T1g (F)]−E[3A2g] = 4021 cm−1 + 12 × 850 cm−1 = 14221cm−1.
NH3 Using the required data from the problem and proceeding as in case of H2O, we obtain
(i) E (3F) − E (3P) = 15 B = 15840 cm−1
(ii) 10750 cm−1 = 10 Dq; or Dq = 1075 cm−1
(iii) E[3T1g (P)] = 28200 cm−1 − 12 × 1075 cm−1 = 15300 cm−1
11.692 × 10 7
(iv) p = 15B′ = = 13211 cm −1
8 85 × 10 3
13211 cm −1
(v) β = = 0.834
15840 cm −1
15840 cm −11 − 13211 cm 1
and β0 = × 100 = 16.6 per cent
15840 cm −1
(vi) p = 13211 cm−1; Dq = 1075 cm−1
E2 −1.966 × 104 E +6.672 × 107 = 0
(1.966 ± 1.094 ) × 10 4
E= cm −1 = 15295 cm −1 , 4355 cm −1
2
Thus E[3T1g (P)] = 15295 cm−1
and E[3T1g (F)] = 4355 cm−1.
(vii) E[3T1g (P)] − E[3A2g] = 15295 cm−1 + 12 × 1075 cm−1 = 28195 cm−1
and E[3T1g (F)] − E[3A2g] = 4355 cm−1 + 12 × 1075 cm−1 = 17255 cm−1.
HC(O)N (CH3)2 Using the required spectral data from the problem and proceeding as in case of H2O, we get,
(i) E(3F) − E(3P) = 15 B = 15840 cm−1
(ii) 8503 cm−1 = 10 Dq
or Dq = 850.3 cm−1
(iii) E [3T1g (P)] = 25000 cm−1 − 12 × 850.3 cm−1 = 14796 cm−1
13.1867 × 10 7
(iv) p 15B ′ = = 13601.5 cm −1
9.695 × 10 3
13601.5 cm −1
(v) β = = 0.858
15840 cm −1
15840 cm −11 − 13601.5 cm 1
and β0 = × 100 = 14.1 per cent
15840 cm −1
(vi) p = 13601.5 cm−1, Dq = 850.3 cm−1
E2 − 1.8703 × 104 E + 5.7824 × 107 = 0
E=
(1.8703 ± 1.0886)104 cm −1 = 14794.5 cm −1 , 3908.5 cm −1
2
Thus, E[3T1g (P)] = 14794.5 cm−1
and E[3T1g (F)] = 3908.5cm−1
(vii) E [3T1g (P)] − E[3A2g] = 14794.5 cm−1 + 12 × 850.3 cm−1 =24998 cm−1
and E [3T1g (F)] − E[3A2g] = 3908.5 cm−1 + 12 × 850.3 cm−1 = 14112 cm−1.
(b) M3+ and M2+ Ions with d3- and d7 Configurations Respectively in an Octahedral field of a Ligand. The
energies E( in cm−1) of the various d–d transitions of M3+ ion (with d3−configuration) and of M2+ ion (with
d7−configuration) in an octahedral field of a ligand are shown in Figs 5.59 (a) and (b). Thus, the Racah parameters
and bending constants of d3 and d7 ions in an octahedral field can be determined from the expressions deduced
from the Figs 5.59 (a) and (b) respectively. For d3 ions, the expressions are
E[4A2g → 4T2g(P)] = 12 Dq + 15 B′ + x (in cm−1) (5.66)
E[4A2g → 4T1g(F)] = 18 Dq − x (in cm−1) (5.67)
E[4A2g → 4T2g] = 10 Dq (in cm−1) (5.68)
For d ions, the expressions are
7
E[4T1g → 4T2g(F)] = 8 Dq + x (5.69)
E[4T1g →4A2g (F)] − E[4T1g → 4T2g (F)] = 10 Dq (5.70)
E[4T1g → 4T1g (P) ] = 15B′ + 2x + 6 Dq (5.71)
(a)
with Mixing
4T
4P x 1g (P)
No Mixing
15 B
x 4T
6Dq 1g (F)
4F
2Dq
4
T2g
12Dq
10Dq
4A
2g
( )
(b)
with Mixing
4
T1g (P)
2x
4
P No Mixing
15 B 4
A2g
18Dq
4T
2g
x
4F
6Dq 8Dq
4
T1g (F )
Fig. 5.59 The splitting of 4F and 4P spectroscopic terms of (a) d3, and (b) d7 ions in and octahedral field of ligand.
Problem 5.19: The electronic absorption spectrum of [Cr (H2O)6]3+ ion shows three bands at 37879, 24570
and 17241 cm−1. Assign these bands to the various transitions involved in the spectrum of the compound and cal-
culate the values of the Racah parameters and that of the bending constant. B = 918 cm−1 for virgin Cr3+ ion.
Solution According to Fig 5.59 (a), the bands at 37879, 24570 and 17241 cm−1 correspond to the 4A2g → 4T1g
(P), A2g → T1g (F) and 4A2g → 4T2g, transitions respectively. In order to determine B′, the values of Dq and x need
4 4
to be computed first.
From Eq. (5.68) we write
E[4A2g → 4T2g] = 10 Dq = 17241 cm−1
or Dq = 1724.1 cm−1
Substituting the values of Dq = 1724.1 cm−1 into Eq. (5.67) we obtain
E[4A2g → 4T1g (F)] = 18 Dq − x = 24570 cm−1
or x = (18 Dq) − 24570) cm−1
= (18 × 1724.1 − 24570) cm−1
= (31033.8 − 24570) cm−1 = 6463.8 cm−1
Substituting the values of Dq and x into Eq. (5.66), we obtain
E [4A2g → 4T1g (P)] = 12 Dq +15 B′ + x = 37879 cm−1
or 12 × 1724.1 cm−1 + 15B′ + 6463.8 cm−1 = 37879 cm−1
or 15 B′ = (37879 − 6463.8 − 20689.2) cm−1 = 10726 cm−1
or B′ = 715 cm−1
Thus, by Eq. (5.62);
B ′ 715 cm −1
β= = = 0.778.
B 918 cm −1
and from Eq. (5.61), β0 =

(918 − 715) cm −1 × 100 = 22.1 per cent
918 cm −1
Problem 5.20. The electronic absorption spectrum of [Co(H2O)6]2+ shows three bands at 8000, 16000 and
19417 cm−1. Calculate the values of Racah parameters as well as that of Dq and bending constant. B = 917 cm−1
for bare Co2+ ion.
Solution From Fig 5.59 (b) and Eqs. [5.69 − 5.71), we write
E(4T1g → 4T2g) = 8000 cm−1 = 8 Dq + x
E[4T1g → 4A2g (F)] − E[4T1g → 4T2g] = 10 Dq = (16000 − 8000) cm−1 = 8000 cm−l
and E [4T1g → 4T1g (P)] = 15 B′ + 2x + 6Dq = 19417 cm−1

Solving these Eqs, for Dq, x and B′, we get
Dq = 800 cm−1, x = 1600 cm−1 and B′ = 761.1 cm−1
Thus, β
761.1 cm −1
β0 =
(971 − 761) cm −1 × 100 21.6 per cent
971 cm 1 971 cm −1
(c) M2+ ions with d7 Configuration in tetrahedral field. The transition energies Es (in cm−1) of the various d–d
transitions of Co2+ ion in tetrahedral field are related to Δt by the following expressions.
E1 = Δt for 4A2 → 4T2 (5.72)
E2 = 1.5 Δt + 7.5 B′ − Q for A2 → 4T1 (F)

4
(5.73)
E3 = 1.5 Δt + 7.5 B′ + Q for 4

A2 → 4T1 (P) (5.74)
1
( B ′ ) + 0 64 Δ t2
2
Q Δ (5.75)
2
The spectroscopic ground terms of M2+ ion with d7 configuration, i.e. Co2+ , are 4F and 4P. The Mulliken terms
corresponding to 4F are 4A2, 4T2 and 4T1 (F) while 4T1 (P) for 4P. The energy difference between spectroscopic terms
of Co2+ gaseous ions is 14565 cm−1, i.e. 15 B = 14565 cm−1. The above expressions can be used to determine the
values of Δt and b for Co2+ ion in tetrahedral field from the spectral data as follows:
Addition of Eqs. (5.73) and (5.75) yields.
E 2 + E3 − 15 B ′
Δt = (5.76)
3
1
Q (E3 − E 2 ) (5.77)
2
Squaring both sides of Eq. (5.75) and rearranging the resultant expression, we obtain
8 B ′ Δt + 225 ( B ′ )
2
4Q 2 2
18 (5.78)
Substitute the values of Δt and Q from Eqs. (5.76) and (5.77) respectively into Eq. (5.78), and solve for B′.
Once B′ is known, the value of Δt can be obtained from Eq. (5.76) and b° and b from Eqs. (5.61) and (5.62)
respectively.
Problem 5.21: The electronic absorption spectrum of [Co(Cl)4]2− yields two bands at 15000 and 5800 cm−1.
The former has been assigned to 4T2→4T1 (P) while the latter to 4A2 → 4T1 (F) transitions. Calculate the values of
the Racah parameters for Co2− ion in the tetrahedral field due to chloride ions.
Solution Here, E3 = 15000 cm−1 and E2 = 5800 cm−1.

1 1 9200 −1
Q (E3 − E 2 ) cm −1 = (15000 − 5800) cm −1 = cm = 4600 cm −1
2 2 2
Substituting the values of E3 and E2 into Eq. (5.76), we get
Δt =
(5800 + 15000) cm −1 − 15 B ′ =
20800 cm −1 − 15B ′
1387cm −1 − B ′ ]
= 5[1387cm
3 3
Substituting Q = 4600 cm−1 and Δt = 5 [1387 cm−1 − B′] into Eq. (5.75),
we obtain a quadratic in B′, i.e.
4 × (4600)2 = 25 (1387 − B′)2 − 18 × B′ × 5 (1387 − B′) + 225 B′2
Simplification and rearrangement of the above expression yields
68 B ′ 2 − 38836 B ′ − 7309155 = 0
Solving the quadratic for B′, we get
38836 ± 59128.6
B′ = cm −1
136
97964.6
Thus, B′ = = 720 cm −1
136
The negative root is not considered.
For B′ = 720 cm−1,
Δt = 5 (1387 cm−1 − 720 cm−1) = 3335 cm−1.
Since, Δt = E1; therefore, E1 = 3335 cm−1
E1 (= 3335 cm−1) corresponds to the 4A2 → 4T2 (F) transition. A band at 3300 cm−1 has indeed been
reported in the infrared region of the spectrum.
The nephelauxetic ratio b is obtained from Eq. (5.62), i.e.
B ′ 720 cm −1
β= = = 0 74
B 971 cm −1
and the value of parameter b° from Eq. (5.61), i.e.
β0 =
(971 − 720) cm −1 × 100 = 25.8 per cent
971cm −1
5.19 BASIC PRINCIPLES OF A DOUBLE BEAM UV-VISIBLE

SPECTROPHOTOMETER
The self-explanatory block diagram indicating the basic elements of an UV-visible spectrophotometer is shown
in Fig. 5.60. The spectrophotometer consists of radiation source, two cells, one for sample solution and the
other called reference or blank cell for solvent only. The radiation from the source is passed through each of
these cells separately, and analysed and detected using a rotating prisim and a photomultiplier. The given range
of wavelengths is scanned for both the beams with a suitable optical arrangement of series of mirrors, prisms
and diffraction gratings not shown in the Fig. 5.60. The radiation is made to fall alternately on the detector
Oscilloscope
Chart
Recorder
Pen
Computer
C
Amplifier
Reference
Cell Blocking
Comb
Detector
Source
Rotating
Sample Prism
Cell
Fig. 5.60 Block diagram of a double beam UV-visible/infrared spectrophotometer.

with the help of a pair of rotatable chopping mirrors. When two beams have the same intensity, there is no
resultant signal from the detector and the blocking or compensating comb C is withdrawn from the reference
beam. When the radiation is absorbed by the sample at a particular wavelength, the comb C moves into the
reference beam so that the emergent beams retain the same intensity. The movement of the comb which is con-
nected to the chart recorder/computer/oscilloscope and the amplifying assembly is a measure of absorption
of radiation by the sample at a given wavelength. Thus, the absorption spectra are plot of absorption against
wavelength or a plot of log e against wavelength. The plot between log e and wavelength has an advantage
over the plot between absorbance and wavelength especially in related compounds. The absorption intensity
of related compounds is about the same but the parameter e brings out the differences better. This is probably
so when a large part of the molecule does not contribute to absorption which is due to chromophore only, but
certainly contributes to the molecular mass. When the molecular mass is not known then specific absorption
coefficient is used.
(a) Monochromators and Windows A monochromator is a device to isolate a band of wavelengths, usually
much narrower than is obtained by a filter. Glass is the material used for prisms, windows and lenses in the visible
region. Fused silica or quartz prisms, windows and lenses are used both in the UV and visible region out to 185
nm. Lithium fluoride is used out to 110 nm and calcium fluoride or sapphire out to 125 nm. Reflection grating
monochromators are used both in the UV and visible region.
(b) Sources of Radiation Sources of radiation in the UV-visible region of the spectrum are 400 W high pres-
sure Xe discharge lamp (300−1000 nm). Tungsten filament lamps in quartz envelopes (near ultraviolet), Tungsten
filament lamps (300−1000 nm), high pressure hydrogen or helium discharge lamps (300−200 nm or below); 1mm
of H2 or He, emits a continuum useful out to 100−60 nm. UV spectrum out to 185 nm can be studied by flushing
out air with nitrogen since atmospheric O2 starts absorbing at around is 195 nm. Lower wavelengths than this
require evacuated path.
(c) Detectors Three types of photosensitive devices, called detectors, used to detect-radiations are Barrier
layer or photovoltaic cells. Photoconductive cells permit operation down to 1 m (near infrared). Phototubes
or Photo-emissive (vacuum)Tubes These tubes are more sensitive than photovoltaic cells,and photomultiplier
tubes. (The sensitivity of a photo-emissive cell can be increased by employing the photomultiplier tube.)
Photographic detection is still a very useful means of obtaining ultraviolet spectra, particularly for high-
resolution studies.
(d) Sample Preparation and Cells The solvent used for the test solution should not absorb radiation in the
region of electromagnetic spectrum under investigation. Some of the most commonly used solvents along with
their lowest wavelength limit (in nm) for 1 cm cell are acetonitrile (190), water (191), cyclohexane (195), hexane
(201) or any other saturated hydrocarbon, methanol (203), ethanol (204), ether (215), methyl chloride (220)
chloroform (237), carbon tetrachloride (257), carbon disulphide (380), acetone (330), benzene (285), toluene
(290), dioxane (220). Cells used for the determination of spectra in solution vary in path length from 1 to 10 cm.
However cells are generally of 1 cm thickness and effective thickness, also called optical thickness, is changed by
changing the concentration of the sample. Microcells with an internal width of 2 mm and a path length of 0.1 mm
are used when only a small amount of solution is available. Glass cells are used in the visible region while for UV
region, cells made of quartz are used. When cell thickness is not mentioned, it is taken as 1 cm. Solid samples are
examined as thin films (cast or painted or evaporated), KBr pellets or ‘Nujolmulls’. Reflectance studies are made
on solids.
The samples for investigation are very dilute solutions, though gases can be used. They are held in fused
quartz cells, usually of 1.0 to 100 mm path length. Path lengths of 0.5−120 m can be achieved by using cells
containing mirrors. These cells are equipped with gas inlets and outlets. Celt jackets are available through which
liquids may be circulated for temperature control. The sensitivity of detection is 10−6−10−7 g. For UV spectral
studies, the instrument is evacuated for air since atmospheric oxygen UV absorption starts at wavelengths less
than 200 nm (i.e. in 195 nm, Schumann−Runge band systems; such studies may be covered under vacuum UV
spectroscopy).
Note that the block diagram of IR single and double-beam spectrophotometers applies equally to UV-visible
spectrophotometers but the sources and detectors and the materials used for prisms and the specifications of the
grating are different. Moreover, in IR the monochromator is placed before the detector and after the sample to
avoid stray light while in UV-visible, it is kept before the sample.
The more elaborated block diagram for double-beam UV-visible spectrophotometer applies equally to
IR-double-beam spectrophotometer, but again here the sources and the detectors and the materials used for prisms
and specifications of the grating used are different. The IR spectra are recorded in the transmission mode while
the UV-visible spectra are recorded in the absorption mode.
(e) Wavelength Calibration Wavelengths are checked from both sides by a mercury arc source which gives
lines at 690.7, 579.1, 577.0, 546.1, 496.0, 491.6, 435.8 and 404.7 nm, potassium flame which gives lines at 769.9
and 766.5 nm or even a hydrogen lamp for a quick check. Hydrogen gives three lines at 656, 486.1 and 379.9
nm. The wavelength is usually set by rotating the prism or grating or a mirror. The other sources for wavelength
checking are rare earth glasses (holomium, dydamium, etc.) or vapour absorption bands of benzene or any line
source like neon, Ar, Kr, cadmium, etc. Wavelength of the red cadmium line in air, 760 mm pressure and 15°C
is 6438.4696 Å and is used as a primary standard. Wavelengths of such sources along with their intensities both
in the UV and visible regions are given in standard data books such as ‘Handbook of Chemistry and Physics’ by
CRC Press.
(f) Calibration of Photometric Scale The photometric scale is checked by liquid standards such as KNO3
K2Cr2O7 or K2CrO4 solutions. The complete prescriptions are given in standard data books together with absor-
bance values. Solid standards with well-defined absorbances marked on them are also available, e.g. rhodiumised
glass or silica or other set of neutral absorbance filters. A dye suspended in gelatin and sandwitched between two
glass plates acts as an absorption filter. Coloured solutions of inorganic salts and organic dyes in rectangular cells
Table 5.22 Absorption characteristics of some common filters
Colour of filter Absorption band (in nm or mm)

Yellow 450
Red 575
Purple 450–650
Orange 500
Blue 480
Green 575–700
can also act as filters. Coloured glass filters are preferred as compared to solution filters because of their greater
thermal stability. Characteristics of some common filters are described in Table 5.22.
The filters transmit radiations of some desired wavelength which correspond to the absorption maximum but
absorb wholly or partially other unwanted wavelengths. Coloured glasses are used as filters in the visible region.
The other class of filters which function on the interference phenomenon, called interference filters, are also
in use. Such a type of filter consists of a layer of transparent material such as magnesium fluoride or calcium
fluoride which is coated on each side with a semitransparent thin film of silver. The wavelength bands are iso-
lated and called band pass, i.e. width of the transmission band at half the peak transmission are much narrower
(10−15 nm) and the peak transmittances much greater (40−60
per cent) than in case with glass filters (i.e. 35−50 nm; 5−20 Source
per cent decreasing with improved spectral isolations). Multi-
layer interference filters have transmissions of 50−70 per cent
and bandwidths of 1.0−5.0 nm. Filters of one type or another
are available which, alone or in combination, allow a selection Slit
of wavelength bands at almost any region of the spectrum, from
x-rays through the infrared. The band of wavelengths isolated by
filters is much broader than that obtained by a monochromator.
5.19.1 Spectrographs
Grating
A spectrograph is an instrument for producing a spectrogram, a
Photographic
photographic image of the spectrum. Both absorption and emis- plate
sion high-resolving power spectrographs are employed to record Fig. 5.61 Block diagram of a simple high-resolution
the spectra of gases. The self-explanatory flowsheet diagram of ultraviolet or visible spectrograph.
a simple high-resolution spectrograph is shown in Fig. 5.61. The dispersing element of the spectrograph is a
grating. Source of radiation for absorption studies is a discharge lamp and for emission spectrum, samples are
excited either thermally or electrically and the spectrum is recorded photographically. The photographic plate
or film is mounted so that successive wavelengths are focussed at the successive points. Thus, an entire spectral
region can be photographed simultaneously. The whole spectrograph is housed in an evacuated tank to allow air
to be removed, since atmospheric oxygen UV absorption starts around 195 nm (Schumann–Runge band systems)
[Vacuum UV spectroscopy]. The rotational fine structure of the electronic bands, even for many diatomics, is not
clearly resolved if the resolving power of the spectrograph is not very high. So, spectrographs with focal lengths
of several metres are used.
5.20 APPLICATIONS OF LOW RESOLUTION UV-VISIBLE SPECTROSCOPY

Chemical analysis or the art of qualitative identification of different substances and determining their amounts in
a quantitative way, is of great importance for the progress of science and technology. Just like other spectroscopic
techniques, namely IR, NMR, etc., UV Visible spectroscopy also meets the above qualifications. Wavelength
position of the absorption band is a criterion for the identification of chemical species and as such is used for
qualitative analysis. Variation of absorption with polarisation of radiation throws light on the internal structure of
the molecule. If the incident radiation used in an absorption experiment is polarised, only those transitions with
similarly oriented dipole moment vector will occur. In a powder, the molecules or complex ions are randomly
oriented.
All allowed transitions will be observed, for there will be a statistical distribution of crystals with dipole
moment vectors aligned with the polarised radiations. However, suppose, for example, a formaldehyde crystal,
with all molecules arranged so that their Z-axes are parallel, is examined. When the Z-axis of the molecules in the
crystal are aligned parallel to the plane of polarisation of light in the Z-direction, light will be absorbed for p–p*
transition. Light of this wavelength polarised in other planes will not be absorbed. If this crystal is rotated so that
Z-axis is perpendicular to the plane of polarisation of light, no light is absorbed. This behaviour supplements the
assignment of the band to p–p* transition. Intensity of light absorbed is a measure of the quantity of the absorbers
present and is useful for quantitative analysis. Based on these parameters, some of the qualitative, quantitative and
biological applications of this technique are described here.
5.20.1 Qualitative Applications

The qualitative applications are based on the position of the absorption maxima and factors affecting it, e.g. tem-
perature, pH, concentration, solvent, etc.
(a) Identification of Molecular Species The visible UV spectra are distinguishable for many different species.
Solutions of hydrated cobalt ion are pink while those of cobalt ion formed in excess of chloride ions are blue.
The absorption spectra of these two ions, i.e. Co[(H2O)6]++ and [CoCl4]2−, are completely different as shown in
Fig. 5.62. The absorption maximum in the former is at 515 nm and 667 nm in the latter.
Percent Transmission
[Co(H2O6)]2+
[CoCl4]2−
400 500 600 700

Wave length (nm)
Fig. 5.62 The absorption spectra of two complex ions of cobalt.
(b) Search for Extraterrestrial Life on Planets Among the more novel and fascinating uses of UV absorption
spectra is the plan to search for extraterrestrial life using automated instrumentation in space probes such as
Viking missions to Mars and other planets. Molecules, such as proteins that contain peptide bonds, absorb in a
band centred around 195 nm in the UV. Water molecule shows n−s* band at 167 nm (emax = 7000).
If a sample dug by the probe showed these absorptions, it would be
H R O R
suspected of containing a biological material and water. Thus, one can
infer from these type of data whether water and life in the present/past N C C N C C
was possible on Mars and on other planets or not. Such type of data
could also give scientists clues about how life on Earth began and the H H H O
origin of life on Earth. Peptide bond
(c) Study of Environmental Effects on Molecular Systems Molecules are generally affected by their en-
vironments. UV/visible spectral shifts provide valuable information about these effects since UV/visible spec-
tra are sensitive to intermolecular interactions such as solvent effects that modify the electronic states of the
molecule or ion, e.g. molecular iodine dissolves in nonpolar solvents such as carbon tetrachloride (CC14) or
cyclohexane (C6H12) with very little interactions between I2 and solvent molecules. These solutions of iodine
have a characteristic purple colour similar to the colour of solid and gaseous iodine (lmax = 520 nm). However,
in solvents such as pyridine (C5H5N), the colour changes to reddish brown with the change in the absorption
spectrum. The change is due to the interaction of iodine with the pyridine molecule and affects one or more
of the electronic states of iodine changing their energies and thereby changing the frequencies of the radiation
absorbed.
(d) Structural Elucidation Structural information such as bond-distances and angles can’t be deduced
from most UV/Visible absorption spectra but it provides a valuable tool in the identification of unsaturated
organic compounds and in elucidation of their structure. However, it has no match to infrared spectroscopy.
According to Prof. Woodward, no single tool has had more dramatic impact upon organic chemistry than
infrared measurements.
Position of the absorption band and its intensity is used for identification of the compound. The certainty of
identification can be improved by measuring the spectrum in a number of solvents or solution conditions such as
pH of both the known and unknown. According to Hartley’s rule, ‘compounds having similar structure would have
similar absorption spectra’.
Sometimes (but not always like IR, NMR, etc.) a direct comparison of absorption spectrum of an unknown
compound with those of model compounds of known structures may give information about the structure of an
unknown compound, e.g. chemical analysis of cannabidiol, (a substance isolated from wild hemp), suggests that
its structure could either be (I) or (II).
The UV spectra of this compound shows two bands around 276 and 282 nm. The shape and position of the
bands resemble very closely with that of 5-amyl resorcinol rather than
that of 4-amyl resorcinol which is quite different. Thus, the structure of C5H11 C5H11
cannabidiol corresponds to (I).
An aliphatic or alicyclic hydrocarbon, alcohol, ether, acid or a ketone R OH R
does not show any absorption band in the 200−800 nm region. A weak
band near 280 nm suggests an aldehyde or a ketone. A strictly substituted OH OH OH
phenyl or triene can also give a band around 280 nm but its intensity is (I) (II)
very strong.
The characteristic bands due to chromophores are an aid for deter-
mining the general structure of molecules. The conjugated double bonds in the molecule cause a red shift of
absorption band of the parent molecule. These features when examined systemati-
cally provide many useful facts about structure. If a compound contains two, three H
or four conjugated double bonds, a band is observed in the 220−300 nm region and
vice versa. This shows that a prior chemical information plays an important role CCl3 C O H
in identification. It is to be noted that these group determinations are possible only
in relatively simple but not for large chromophores like dyes. The procedure has
been particularly useful in fields of carotenoids, unsaturated fatty acids, alkaloids, OH
(I)
terpenes, steroids where the proposed structure contains double bonds, conju-
gated or unconjugated, cyclic (cis) or a cyclic (trans). We can only answer specific
Trans
Absorbance
Cis
200 250 300 350
Wavelength (nm)
Fig. 5.63 Absorption spectra of cis- and trans-stilbenes.
questions on the basis of UV/V spectral data. This is demonstrated here with the aid of some typical examples.
Chloralhydrate shows a band at 290 nm with emax = 33. However, this band is absent in aqueous medium.
Thus, the structure of chloralhydrate is (CC13 − CHO) H2O but not structure (I) since the low intensity band at
290 nm is attributed to carbonyl group. Geometrical isomers with conjugation usually have different UV spectra.
Cis-stilbene (1, 2-diphenyl ethylene) shows an absorption band at 280 nm with emax = 12300 in ethanol while
trans-isomer exhibits a corresponding band at 295 nm with emax = 25000 in ethanol (Fig. 5.63). Delocalisation of
the p-electrons, i.e. coplanarity, is destroyed by steric interference in cis-isomer. The molecule has to twist slightly
out of plane to avoid clashing of the hydrogen atoms at the o-positions of the two benzene rings. Consquently, a
low-intensity band is observed in the cis-form. A band in stilbene appears to be swamped by the intense band at
295 nm. The parent absorption bands of stilbene molecule shift to longer wavelengths and undergo increase in
intensity also as ‘n’ in C6H5 − (CH = CH)n − C6H5 increases. For n = 7, the molecule absorbs in the 400−465 nm
region with emax = 135000.
We can find whether the two double bonds in any unsaturated system are conjugated or not, e.g. mesityloxide
an a−b unsaturated ketone, absorbs at 230.6 nm with emax = 12000 in isooctane and 237.6 nm in chloroform. So it
must be −C = C− C = O. Further, we can also differentiate whether the two double bonds conjugated are alicyclic
diene or exocyclic or aliphatic diene. The former absorbs at 250 nm with emax = 5000 while the latter at 225 nm
with emax = 10,000.
Apart from these, specific empirical rules have been worked out, which enable us to deduce the position of
substituents or double bonds in a particular class of compounds, e.g. various heterocyclic compounds and their
derivatives, fused ring systems (naphthalene), many steroids, terpenes, flavone, anthraquinone, etc.
5.20.2 Woodward–Fieser Rules for Predicting the Wavelength

of Absorption Maximum in Conjugated Dienes
According to these rules, each type of diene has a certain fixed value. The wavelength of the absorption maximum
lmax depends upon (i) the number of alkyl substituents or ring residues on the double bond, (ii) the number of
double bonds which extends conjugation, and (iii) the presence of polar groups such as −Cl, −Br, −OR, −SR, −NR,
etc. These rules are summarised in Table 5.23.
These rules are obeyed by conjugated systems of four double bonds or less. For conjugated polyenes having
more than four double bonds, the Fieser–Kuhn rules hold good.
Table 5.23 Woodward–Fieser rules for diene and triene (p−p* transition)
absorption (contribution of various structural units towards lmax.).
(nm)
Parent value for:
Butadiene system or cyclic conjugated system 217

Heteroannular or open chaindiene 214
Homoannular diene 253
Increments for:
Alkyl substituent or ring residue 5

Exocyclic double bond 5
Double bond extending conjugation 30
Polar groups:
– O alkyl 6
– O acyl 0
– S alkyl 30
– Cl, –Br, 5
– N (alkyl)2 60
Solvents have negligible contribution to lmax of these p−p* transitions, i.e. solvent
correction is zero. In addition when both homo- and hetero-annular conjugated diene
systems are present in the same compound, the base value corresponding to homo-an-
nular diene, i.e. 253 nm, is used. When a diene system is present in bicyclic compound,
then a correction factor of 15 nm due to strain in the bicyclic compound is added to the
calculated value of absorption maximum.
Factors Affecting the Woodward−Fieser Rules The discrepancy between the observed and
calculated values of absorption maximum (i.e. lmax) may be due to the following parameters:
(i) Strain in the Molecule Strain in the molecule may cause a blue shift from the
calculated value, e.g. verbenene (I) shows an absorption maximum at 245.5 nm while Woodward−Fieser
rules yield a value of 229 nm, i.e. the blue shift deviation is 16.5 nm.
(ii) Distortion of Chromophore Distortion of chromophore in dienes may cause blue or red (I)
shifts depending upon the nature of distortion from the expected theoretical value. When distortion
of chromophore affects the coplanarity of the structure, loss of conjugation takes place, and consequently, a blue
shift results, e.g. the expected value for the diene (II) is 273 nm while the observed value is 220 nm (emax = 5500
nm). On the other hand, even when the coplanarity is not likely to be affected, still a red shift from the expected
value results, e.g. the expected value for the diene (III) is 238.5 nm while the observed value is 248 nm (emax =
15800), i.e. a red shift of 9.5 nm.
(iii) Ring Size The change of ring size in case of simple homo-annular dienes exhibits a blue shift from the expected
theoretical value, e.g. the expected value for diene (IV) is 263 nm while the observed value for cyclopentadiene is
238.5 nm (emax = 3400); cycloheptadiene is 248 nm (emax = 7500) and cyclohexadiene is 256 nm (emax = 8000)
(iv) Cross Conjugation Cross conjugation such as (V) in conjugated polyenes also affects the expected lmax
value.
(II) (III) (IV) (V)

In view of the foregoing discussion, one should be very careful while comparing the UV spectra of an unknown
compound with the model compound. Corrections regarding the shape of the molecule, strain in the molecule,
stearic hindrance, distortion of chromophore, etc., must be taken into consideration before jumping to any conclu-
sion about the structure of the unknown compound.
5.20.3 Fieser–Kuhn Rules

According to these rules, both the lmax and emax is related to the number of conjugated double bonds as well as
other structural units by the following equations:
lmax = 114 + 5 M + n(48.0 −1.7n) −16.5 Rendo −10 Rexo (5.79)
emax = (1.74 × 104) n (5.80)
where n is the number of conjugated double bonds, M is the numbers of alkyl or alkyl-like substituents on the
conjugated system, Rendo is the number of rings with endocyclic double bonds in the conjugated system and Rexo is
the number of rings with exocyclic double bonds in the conjugated system.
These rules hold good for polyenes with more than four conjugated double bonds, e.g. lycopene, a-and
b-carotenes, etc.
Before applying these rules, let us describe the general structure and electronic spectra of various types of
conjugated dienes. The four types of conjugated dienes are the following:
(a) Alicyclic Dienes or Dienes They are open-chain dienes and exist normally in the s-trans (transoid) con-
formation. The basic unit in them is the butadiene system (VI) i.e. emax in these dienes lies in the 8000–20000 range.
C C C C
(VI)
(b) Homo-annular Dienes They have conjugated double bonds in the same ring and exist in the s-cis
(cisoid) conformation. Since the transition moment in homodienes is small as compared to those of hetero-
annular and alicyclic dienes, the emax is of the order of ∼10,000. Typical examples of homo-annular dienes are
shown below.
1 3
2 3
4
1
2 1
2
Ring Residue = 2 =3 =4
(c) Hetero-annular Dienes They have conjugated double bonds in different rings. emax in these dienes is
comparable to that of alicyclic dienes, i.e. emax = 8000−20000. Typical examples of these systems are shown here.
3 4
2 3 2
2
1 1
1 3
2
3
Ring Residue = 3 =3 =5 1 =4
4
(d) Exocyclic and Endocyclic Dienes They contain exo- and endocyclic conjugated double bonds as shown
in the systems.
Exo to ring Y
X Y X Y
Exo Endo Exocyclic double Exo to ring X

bond to ring X
X Y
X Y
Exo to ring Y
Exo to ring Y
Problem 5.22: Calculate the absorption maximum for cholesta-3, 5,-diene in ethanol,
Solution It is a heteroannular diene and its spectroscopic conjugated skeleton is
Thus, from Table 5.23 we write

Structural unit Absorption (in nm)
Base value 214
Ring residues (= 3) 15
5
Exocyclic double bond (= 1) lmax = 234 nm
lmax (obs) = 235 nm (emax = 19000)
Obs lmax for cholesta-2, 4,-diene = 275 nm (emax = 10000)
Problem 5.23: Calculate the absorption maximum for
Solution It is heteroannular diene. Its spectroscopic conjugated skeleton is

Base value 214
lmax = 234 nm (obs: 234 nm)
Problem 5.24: Calculate the absorption maximum for 4-cyclohexenyl-2-pentene.
C CH CH CH3
CH3
Solution It belongs to a butadiene system and its spectroscopic conjugated skeleton is
C C C

Base value 217
Alkyl substituent (= 2) 10
Exocyclic double bond (= 1) 5
lmax = 242 nm
lmax (obs)= 242 nm
Problem 5.25: Calculate the absorption maximum for 2, 3-dimethylene bicyclo [2, 2,] haptane;
Solution From Table 5.23 we write

Base value 217
Exocyclic double bonds (= 2) 10
Bicyclic system (= 1) cause strain 15
lmax = 252 nm (obs: 254 nm)
Problem 5.26: Yellow plant pigments a-, b- and g-carotene (e.g. carrots contain 15 per cent a, 85 per cent b,
0.1 per cent g) and the red pigment of tomatoes, i.e. lycopene, are converted into vitamin A in the liver, the lack
of which in the diet causes hardening of the conjunctiva, corneal softening and to complete blindness, weight loss
and poor general physical condition; and a deficiency of vitamin A causes night blindness. All the four pigments
have the same molecular formula (C40 H56) but different structures given below:
(a) a-carotene
CH3 CH3 CH3
CH3 CH3
CH CH C CH CH CH C CH CH CH CH C CH
CH3
H3C CH3
CH3
CH CH C CH CH
H3C
H3C CH3
(b) b-carotene The ring on the right-hand side is
(c) g-carotene CH3

CH3 CH3
CH3 CH3
CH CH C CH CH CH C CH CH CH
CH3 CH3
CH3 CH C CH CH CH C CH
CH3
CH CH C CH2 CH2 CH C(CH3)2

(d) Lycopene
CH3 CH3 CH3
H3C
C CH CH2 CH2 C CH CH CH C CH CH CH C CH CH
H3C
CH3 CH3 CH3
CH3
CH CH C CH CH CH C CH CH CH C CH2 CH2 CH C
CH3
Calculate the absorption maximum for the given systems.
Solution (a) In a-carotene, there are ten conjugated double bonds, i.e. n = 10; and five methyl groups on the
conjugated system, i.e. M = 5. There are two rings with endocyclic double bonds in a-carotene. Thus from Eq.
(5.79), we obtain
lmax =114 + 5 × 10 + 10(48.0 − 1.7 × 10 ) − 16.5 × 2
= 114 + 50 + 480 − 170 − 33.0 = 441 nm
lmax (obs) = 445 nm.
and from Eq. (5.80), we get

emax= (1.74 × 104) 10 = 1.74 × 105 ; emax (obs)
= 1.45 × 105
(b) In b-carotene, n = 11; M = 10, there are two rings with endocyclic double bond in the molecule.
Thus Rendo = 2
lmax = 114 + 5 × 10 +11(48.0 − 1.7 ×11) − 2 × 16.5 = 114 + 50 + 528 − 205.7 − 33
= 453.3 nm; lmax (obs) = 452 nm (in hexane)
emax = (1.74 × 104) × 11 = 1.91 × 105; emax (obs) = 1.52 × 105 (in hexane)
(c) In g-carotene, n = 11, M = 9 and Rendo = 1. With the aid of Eq. (5.79), we get
lmax = 114 + 9 × 5 + 11(48.0 − 1.7 × 11) − 16.5 × 1 = 114 + 45 +528 −205.7−16.5
= 687 − 222.2 = 464.8 nm
and from Eq. (5.80), we obtain
emax = (l.74 × l04) × 11 = 1.91 × 105
(d) In lycopene, n = 11, M = 8 (six methyl groups and two chain residues). There is no ring system in lycopene,
i.e. Rendo = 0 and Rexo = 0.
Thus, from Eq. (5.79), we obtain
lmax = 114 + 5 × 8 +11(48.0 − 1.7 × 11) = 114 + 40 + 528 − 205.7 = 682 − 205.7 = 476 nm
lmax (obs) = 474 nm (in hexane)
and Eq. (5.80) yields
emax = (1.74 × l04) × 11 = 1.91 × l05.
emax (obs) = 1.86 × 10s (in hexane)
Problem 5.27: Calculate the absorption maximum for (a) Vitamin A1, and (b) Vitamin A2.
Solution (a) The molecular formula of vitamin A1 is C20H30O and its structural formula is
CH43 CH53
CH3 CH3
CH CH C CH CH CH C CH6 CH2R ( R OH )
CH3
Here n = 5, M = 6. There is one ring with endocyclic double bond in the conjugated system. Thus Rendo = 1.
Substituting these values into Eq. (5.79), we get
lmax = 114 + 6 × 5 + 5 (48 − 1.7 × 5) −16.5 = 114 + 30 + 240 − 42.5 − 16.5

= 384 − 59 = 325 nm.
lmax (obs) = 326 nm (in ethanol)
emax = (1.74 × 104)5 = 87000 or log emax = 4.94

log emax (obs) = 4.66 (in ethanol)
or emax = 45709
(b) The molecular formula of vitamin A2 is C20H28O.

It differs from vitamin A1, in that it has one extra ethylenic bond in it.
In vitamin A2, n = 6, M = 6 and Rendo = 1
Substitution of these values in Eq. (5.79) yields
lmax = 114 + 6 × 5 + 6 (48 − 1.7 × 6) −16.5 = 114 + 30 + 288 − 77.7

= 432 − 77.7 = 354.3 nm
lmax(obs) = 351 nm
CH34 CH35
H3C CH3
CH CH C CH CH CH C CH6 CH2R ( R OH )
All Trans
CH3
emax = (1.74 × l04) × 6 = 10200
or log emax = 5.01
5.20.4 Woodward–Fieser–Scott Rules for Predicting the Position

of Absorption Maximum (p-p * transitions) in Enones and Dienones
a, b-unsaturated aldehydes and ketones are called enones. The carbonyl group is conjugated with a ethylenic
group in these systems, e.g. cyclopentenone, mesityloxide, retinene, etc. The basic structural units in these com-
pounds are
b a d g b a
b C C C O d C C C C C O
Enone Dienone
Transitions in a, b-unsaturated carbonyl compounds are shown in Fig. 5.24.

Enones exhibit a strong absorption K-band (215–250 nm, emax = 10000-20000) and a poorly defined weak
absorption R-band (310–330 nm). The rules to estimate absorption maximum in these compounds are summarised
in Table 5.24.
Table 5.24 Woodward–Fieser–Scott rules to estimate absorption maximum (p-p* transitions) in a, b-unsaturated ketones and aldehydes.
Parent values (nm)
Acyclic a, b-unsaturated ketones 215

Six-membered cyclic a, b-unsaturated ketones 215
Five membered cyclic a, b-unsaturated ketones 202
* When carbonyl group is in five membered ring and doubled bond is
215
exocyclic to five member ring.
a, b-unsaturated aldehydes 210
a, b-unsaturated acids and esters Increments for: 195
Double-bond extending conjugation 30
Alkyl group, ring residue:
a 10
b 12
g and higher 18
*Ring strain (as in vervenone) 14
Auxochromes:
− OH a 35
b 30
δ 50
− OAC a, b, δ 6
− OMe a 35
b 30
g 17
δ 31
− Salk b 85
–Cl a 15
b 12
− Br a 25
b 30
−NR2 b 95
Exocyclic double bond 5
Homodiene component
(Two conjugated double bonds, both in the same ring) 39
λ max
EtOH
–
lmax(obs) − lmax(calc) = ± 3nm, emax for cisoid enones < 10,000 and for transoid enones >10,000. The solvent
correction must be applied to the predicted values (in ethanol). The solvent correction should be subtracted from
the calculated values. The solvent corrections (in nm) for a, b-unsaturated ketones are: methanol (0), dioxane (+5),
chloroform (+1), ether (+7), water (−8), hexane (+11), cyclohexane (+11). *Corrections for departure from the
rules.
Problem 5.28: Calculate the absorption maximum for mesityloxide (CH3)2C==CHCOCH3 in (a) chloroform,
and (b) water.
β α
Solution Mesityloxide is an enone: ( )2 C C HCOCH 3
Applying the Woodward–Fieser–Scott rules, we write

Base value =215nm
2 b-alkyl group = 24 nm
(2 × 12)
λ max
EtOH
= 239 nm
λ mab x (ε max = 12600)

Applying the solvent-correction, we get
λ max = 239 − 1 = 238 nm (obs : 237.6 nm

CHCl3
(a)
and
λ max = 239 + 8 = 247 nm (obs : 242.6 nm )
Water
(b)
Problem 5.29: Rhodospin is the protein opsin linked to retinene (or retinal) which is the vitamin A1 aldehyde.
The chain of retinal has an all trans-configuration. Rhodospin absorbs light and retinal bound by rhodospin isom-
erises to 11-cis retinene. The structural formula of retinal is
CH3 CH3
H3C CH3
CH CH C CH CH CH C CH CH O
d g b a
CH3
Calculate the absorption maximum for retinal.
Solution It is a a, b-unsaturated aldehyde.

Thus, from Table 5.24, we write
Retinene (structural units) Absorption (in nm)
Parent value = 210
a-substituent (=0) =0
b-substituent (=1) = 12
g and higher substituents (= 4) = 72 (= 4 × 18)
Double bonds extending conjugation (= 4) = 120 (= 4 × 30)
lmax(calc) = 414
Problem 5.30: Calculate the absorption maximum for the following compounds:
(a) O (b)
Ac O O
O
OH O
(c) (d)
(e) O
Solution
O d +1
(a) The principal spectroscopic conjugated skeleton in the basic molecular
structure is a d +2
b
Thus, g d
lmax = 215 nm (base value) + 60 nm (two double bonds extending conjugation) d
+ 10 (Two exocyclic double bonds) + 12 nm (one b-ring residue) +18 nm (one
g
δ + 1 ring residue) + 18 nm (one δ + 2 ring residues) + 18 nm (one δ + 2 alkyl
groups) + 0 nm (one heteroannular conjugated diene) = 351 nm (obs: 354 nm). b
(b) The principal spectroscopic conjugated skeleton in the molecular struc- a
ture is
Thus,
lmax = 215 nm (base value) + 30 nm (one double bonds extending conjuga- O
tion) + 10 (one a-ring residue) +12 nm (one b ring residue) + 18 nm (one δ ring O
residue) + 39 nm (one homo conjugated diene) = 324 nm (obs: 327 nm).
(c) The main spectroscopic conjugated skeleton in the molecular structure is a
Thus,
b
lmax = 202 nm (base value for cyclopentanone) + 30 nm (one double bond g
extending conjugation) + 12 nm (one b-alkyl group) +18 nm (one g-ring resi- d
due) +18 nm (one δ-ring residue) + 39 nm (one homo conjugated diene) + 5 nm
(one exo double bond) = 324 nm OH O
(d) The basic spectroscopic conjugated skeleton in the molecular structure is a
Thus,
lmax = 215 nm (base value) + 35 nm (one a-OH)+ 12nm (one b-alkyl group)
+12 nm (one b-ring residue) = 274 nm. b
(e) The principal spectroscopic conjugated skeleton in the molecular struc-
ture is
Thus, b
lmax = 215 nm (base value) + 10 nm (one a-alkyl group) + 24 nm (two b-ring
O
residues) + 5 nm (one exocyclic double bond) = 254 nm (obs: 254 nm) a
Table 5.25 Nielsen rules to predict the absorption maximum for a b-unsaturated acids, and esters
(emax > 10,000).
Base value (nm)

a or b-monosubstituted 208
a, b or b, b-disubstituted 217
a, b or b, b-trisubstituted 225
Increment for
exocyclic a, b-double bond 5
exocyclic a, b-double bond-in five or seven membered ring 5
calc obs
λ max λ max = ± 5 nm. Data of Table 5.24 may be used here. The values are for ethanol as solvent. The change in
lmax in going from acid to ester is not more than 2 nm. When the number of double bonds extending conjugation
is more than one, the contribution due to each bond extending conjugation should be taken as 36. The structure of
the basic units in these compounds are
R R R R
C COOH (a-monosubstituted), C HC COOH (b-monosubstituted), C C COOH

a b a b a
R
R R
(a, b-disubstituted), C CH COOH (b, b-disubstituted), C C COOH (a, b, b-trisubstituted),
R b a b a
R
Problem 5.31: Calculate lmax for the following acids:

(a) CH3-(CH=CH)2 COOH (obs lmax = 254 nm, emax = 25000),
(b) CH3-(CH=CH)4 COOH (obs: lmax = 332 nm, emax = 49000),
(c) C COOH (obs: λmax = 220, εmax = 14000),
CH3
(d) CH C COOH (obs:λmax = 210),
COOH
,(obs: l max = 217 nm)
(e)
COOH
(f) ,(obs: l max = 222 nm)
CN
(g) C COOH (obs: l max = 235, emax = 12500)
Solution
(a) CH3 CH CH CH CH COOH

d g b a
is a a, b-unsaturated acid. From Tables 5.25 and 5.24, we write.
Structural units Absorption (in nm)

Parent value 208
Double bonds extending conjugation (=1) 30
g or higher alky1 group (= 1) 18
lmax = 256 nm
(b) CH3-(CH==CH)4COOH is a a, b-unsaturated acid. From Tables 5.25 and 5.24, we write.

Base value 208
Double bonds extending conjugation (= 3) 108
g or higher alkyl group (= 1) 18
lmax = 334 nm
(c) C COOH is a a, b-unsaturated acid. From Tables 5.25 and 5.24, we write

Base value (b, b-disubstituted) 217
Exocyclic a, b-double bonds (= 1) 5
lmax = 222 nm
CH3
(d) CH C COOH is a a-monosubstituted unsaturated acid.
Thus lmax = 208 nm (Base value) + 5 nm (a-methyl group) = 213 nm.
COOH
a
(e) is a a, b-unsaturated acid.
b
From Tables 5.25 and 5.24, we get
Base value (a, b-disubstituted) 217
Exocyclic a, b-double bond in six membered ring (= 1) 0
lmax = 217 nm
COOH
a
(f ) b
is a a, b-unsaturated acid.
From Tables 5.24 and 5.25, we get
Base value (a, b-disubstituted) 217
Exocyclic a, b-double bonds in seven membered ring 5
lmax = 222 nm
CN
(g) C COOH is a a, b-unsaturated acid
From Tables 5.25 and 5.24, we write

Base value (a, b, b-trisubstituted) 225
a-substituent (= 1) 0
Exocyclic double bonds (= 1) 5
lmax = 230 nm
The absorption maximum for the derivatives of acylbenzens, aromatic ketones, aromatic aldehydes, aromatic
acids and aromatic esters can be calculated with the aid of rules devised by AI Scott. The rules are given in
Table 5.26. Note that in polysubstituted benzene rings, when the substitution leads to steric hindrance preventing
coplanarity of the carbonyl group and the ring, the contribution due to the substituent may be decreased markedly
by steric hindrance to coplanarity.
Table 5.26 Scott’s rules for calculating the principal band (p–p* transition) of substituted benzene derivative. R⎯C6H4 ⎯COX in ethanol.
System Orientation lcalc (nm)

Parent chromophore: C6H5
X = alkyl or ring residue (e.g. C6H5COR) 246
X = H(C6H5CHO) 250
X = OH or O alkyl 230
Increment for each substituent on parent chromophore, i.e. C6H5
⎡o , m −
R = alkyl or ring residue ⎢ p− 3
⎣
10
⎡o , m −
= OH, OMe, O alkyl ⎢ p− 7
⎣
25
⎡o −
⎢m −
R = O− ⎢ 11
⎢⎣ p −
20
78*
⎡o , m −
R = Cl ⎢ p− 0
⎣
10
⎡o , m −
R = Br ⎢ p− 2
⎣
15
⎡o , m −
R = NH2 ⎢ p− 13
⎣
58
⎡o , m −
R = NHAc ⎢ p− 20
⎣
45
R = NHMe p− 73
⎡o , m −
R = NMe2 ⎢ p− 20
⎣
85
*This value may be decreased markedly by steric hindrance to coplanarity.

Problem 5.32: Calculate the values of absorption maximum for

(a) 4-aminobenzoic acid
(b) 3-aminobenzoic acid
(c) 2-hydroxy-4-methoxy acetophenone 5 4
MeO
6
3
(d) 2-hydroxy-5-methoxy acetophenone in ethanol.
(e) 6-methoxytetralone 2
(f) 4-dimethylamino benzaldehyde.
O
Solution From Table 5.26, we get
(a) lmax = 230 nm (Parent chromophore) + 58 nm (p-NH2)
= 288 nm (obs: 289 nm; loge = 4.27) MeO
m
p Ring residue
(b) lmax = 230 nm (Base value) + 13 nm(m-NH2) o
= 243 nm (obs: 241 nm, log e = 3.85) m

o C
(c) lmax = 246 nm (Base chromophore) + 7 nm (o-OH) + 25 nm (p-OCH3)
= 278 nm (obs: 273.5 nm; log e = 4.2) O
(d) lmax = 246 nm (Parent chromophore) + 7 nm (o-OH) + 7 nm (m-OCH3) = 260 nm (obs: 255.5 nm, log
e = 3.95)
(e) The principal spectroscopic skeleton in the basic molecular structure is
Thus, lmax = 246 nm (parent value) + 3nm (o-ring residue) + 25 nm (p-OMe) =274 nm (obs: 276 nm, log e = 4.21)
(f) lmax = 250 nm (Base value) + 85 nm (p-N(CH3)2) = 335 nm (obs: 340nm, log e = 4.5)
5.20.5 Quantitative Applications

The quantitative applications of UV-visible spectroscopy are based on intensity of the absorption band. Based on
this quantity, a number of direct and indirect methods have been developed to expand the quantitative application
field of this technique in other fields such as applied and biological sciences.
Concentration Determination Volumetric analysis as an analytical tool considered as one of the classical
methods of chemical analysis. Volumetric titrations of all kinds are still being performed today in many indus-
trial, hospital and research laboratories because of their simplicity, reliability and cheapness. These days, the
sphetrophotometric analysis has in many cases replaced volumetric analysis as an analytical tool especially in
cases where it is cumbersome or impossible to analyse the samples volumetrically. Moreover, the methods of
spectrophotometric analysis are less time consuming as compared to volumetry, and requires a small amount of
the analyte (concentration of the solution as low as 10−6–10−7 g/mL can be determined). These methods of analysis
are used extensively not only in chemistry but also in many other fields of study including biology, biochemistry,
medicine, agriculture, etc. The quantitative analysis of samples by UV/V spectrophotometric technique is based
on Beer–Lambert’s law, i.e. A = ecl. Note that (i) when an impurity present in the sample absorbs at the absorption
maximum of the sought for substance, the wavelength other than the absorption maximum is chosen for analyti-
cal work, and (ii) when the substance is present in two forms and amount of each is concentration dependent,
e.g. acid–base indicator or cis-trans isomer, etc, then total concentration of both the forms is determined at the
isosbestic point, i.e. the point at which both the forms have equal value of e.
The sensitivity for analysis is enhanced on complexation of analyte with some suitable complexing agent.
Even substances with little absorption often will form complexes with other molecules that absorb very strongly,
thereby improving their sensitivity for analysis. Many important analytical procedures are based on this approach:
analysis of phosphates in water; detergents and steels; calcium and magnesium in hard water; iron, nickel and
copper; pH-measurements; food additives; acetone in blood and urine and detection of drugs, etc. The following
steps are involved in the analysis of multicomponent systems: (i) Determine the absorption curves for each com-
ponent in the same solvent. (ii) Determine the suitable wavelengths that show maximum difference in absorbance
and overlap to some extent. (iii) Draw the Beer–Lambert’s law plot for each component. (iv) Calculate the molar
extinction coefficient for each component at selected wavelengths. (v) Prepare known mixtures and calculate
absorbances and check whether the absorbances are additive. (vi) Record the absorbance of unknown sample
at selected wavelengths and solve the simultaneous equations for concentration of components of the system.
Theory of Beer–Lambert’s law and methods based on it has already been described in detail in Chapter 1 on intro-
duction and in Chapter 3 on infrared spectroscopy. Some of the techniques of analysis are being described here.
(a) Method of Standard Addition This technique is useful when the absorbance of the sample at zero con-
centration of the sought-for substance in the sample is known or assumed to be negligible, say it is A0. In such a
case, a known amount of the sought-for substance, say Cen is added to known portion of the sample and absor-
bance of the enriched sample is measured, say it is Aen. From the measured absorbance of the sample, say As, and
the enriched reference, one can determine the concentration of the sought for substance in the sample, say Cs,
under study from the expression:
As A0 C
= s (5.81)
A en A 0 C en
C en ( A s − A 0 )
or Cs =
(A en A0 )
This procedure balances the influence of some undesired constituents in the sample and is used for some
extracts of natural products.
(b) Method of Variable Reference Solution This technique is most suited to dyes which have similar ab-
sorptions, i.e. lmax and emax, e.g. fluorescien and its derivatives, m dibromo and tetrabromo derivatives. Since
they have similar spectra, they cannot be analysed very precisely by simultaneous analysis at three wavelengths.
In such a case, the analysis is carried out by simple comparison of the spectral properties of the standard sam-
ples with the unknown. Let Ax, Ay and Az be the absorbances of the three standard samples with compositions:
C x C y + C z , C x′ + C y′ + C z′ and C x′′ + C y′′ + C z′′ respectively, where Cs’ are the concentrations of the species x, y
and z in the sample. If Az is the absorbance of the unknown sample, then the composition of the unknown sample
will be C x′′ + C y′′ + C z′′.
(c) Curvature Inversion Method This technique is especially useful in the assay of pharmaceuticals which
may contain certain light scattering materials and other ingredients with unknown identity. In this method, differ-
ent concentrations of pure analyte are kept in the reference cell and the concentration of the analyte is found at
which the absorption inverts, i.e. goes below the baseline at the specific wavelength of interest.
(d) Method for Correction due to Colloidal Suspension or Turbidity in Photometric Methods This
technique is used for protein precipitates or dispersed fats which are common causes of turbidity in biological
samples. This method is based on the fact that the law of extinction and wavelength in colourless turbid media is
of the form.
e = k lx
or log e = log k + x log l (5.82)

where k and x are constants.
This gives a linear relationship between log e and log l where ls are the wavelengths of the absorption maxi-
mum and of two minima on both sides of it.
Let the absorption curve of pure substances show two minima at wavelengths l1 and l3 on two sides of a
maximum at wavelength l2. The Beer–Lambert curve is plotted between log e or log A and concentration of pure
substance at three wavelengths. The straight line through l1 and l3 crosses l2 at A 2* . Then the correct absorbance
of the substance will be A 2 A 2*
(e) Absorbance Ratio Method The basic principle of this method has already been discussed in Chapter 3 on
infrared spectroscopy. For a two-component system, we write:
At wavelength l1,
A ′ = a1′C x + a2′ C y (5.83)
and
At wavelength l2,
A ′′ = a1′′ C x + a2′′ C y (5.84)
If one of the points, say l2, is the isosbestic or isoabsorptive point then
A ′′ = a2′′ (5.85.)
A ′ a1′ C x + a ′ C y
R= = (5.86)
A ′′ a′′
where a″ is the absorbance at the isosbestic point, i.e. the point of intersection of absorption curves at wavelengths
l1 and l2.
Thus a plot between R, the absorbance ratio, and Cx or Cy will be a straight line. If the isosbestic point is not
taken then the plot between R and Cx or Cy will not be a straight but a curve joining 100 per cent of x to 100 per
cent of y. The advantage of the slope ratio method is that it can be expanded to three-component analysis roughly
resembling like a phase diagram.
(f) Method of Absorbance Difference This method is most suited when one of the constituents of the
unknown mixtures is to be estimated. In this method, the difference of absorbance is determined at a selected
wavelength, e.g. vitamin C has absorption maximum at 260 nm and disappears in its oxidised form. Thus, from
the absorbance measurements of the unknown compounds containing vitamin C in the unoxidised and oxidised
(ox) forms at absorption maximum of 260 nm, one can determine the amount of vitamin ‘C’ in the unknown
sample from the expression:
ΔA std
Δ A std − A std (ox ) C std l
ΔA mix
Δ A miix − A mix (ox ) C mix l
ΔA mix C mix
or =
ΔA std Cstd
ΔA
A mix
C mix = × C std (5.87)
ΔA
A std
where ΔAstd represents the standard absorbance difference between the two forms of standard vitamin C and
ΔAmix for the vitamin C in the unknown sample. Cstd and Cmix stand for the concentrations of vitamin ‘C’ in the
standard sample and unknown mixture respectively. The same can be followed by changing one isomer to another
as in arachidonic acid, i.e. 6, 10, 14, 18-Eico-satetraenoic acid [CH3(CH2)3(CH2CH:CH)4 (CH2)3–COOH], lmax
(log e) = 233 (4.24), 257 (4.09), 268 (4.21), 315 nm (3.82)] or by changing pH as in case of bromo thymol blue.
The molar extinction coefficients in this case are markedly different at pH 5 and 11. Hence, the difference of
absorbance values at these two pHs will be a measure of absorbance due to bromo thymol blue.
Problem 5.33: A sample of nonferrous alloy weighing 0.325 g is dissolved and treated with ammonia and
250 mL of a coloured solution is obtained. The absorbance of this coloured solution in a cell of 2 cm path length
is 0.254. Determine the percentage content of copper in the alloy given that the molar extinction coefficient of
copper ammoniate is 423.
Solution Here, l = 2 cm, A = 0.254, e = 423

Substituting these values in the Beer–Lambert’s equation: A = eCl, we get
0.254
C= = 3.002 ×10 −4 mole lit −1
423 × 2
002 ×10 4 × 63.5

3.002
= g/250 mL
4
= 4.765 × l0−3 g/250 mL
Hence, the percentage of copper in a given sample of nonferrous alloy is
4.765 ×10 3 × 100
= = 1.46
0.325
Problem 5.34: Calculate the molar extinction coefficient of copper if the absorbance of a solution containing
0.48 mg of copper in 500 mL is 0.14 in a cell that is 2 cm in length.
Solution Atomic mass of copper = 63.5

C = 0.48 mg/500 mL
0 96 10 −3
= 0 96 mg/lit = mole/lit
63.5
l = 2 cm,
A = 0.14
Substituting these values into the expression for Beer–Lambert law, i.e.
A = eCl, we get
0.14 63.5 × 103
ε= ( mole/lit )−1 cm −1
2 0.96
= 4630 (mole/lit)−1 cm−1
Problem 5.35: Both coffee and tea contain an alkaloid called caffeine with molecular formula C8H10O2N4H2O.
This alkaloid is responsible for the mildly stimulating effect of beverages made from coffee beans, tea leaves,
mate leaves, or cola nuts. Caffeine is also present in several common carbonated beverages (soft drinks), e.g.
Coca-Cola. Caffeine shows an absorbance of 0.560 for a concentration of 5.17 × 10−5 mole lit−1 at 272 nm. A 2.00
g commercial sample of an instant soluble coffee is mixed with water to a volume of 500 mL. 20 mL of this solu-
tion is treated with 25 mL of 0.1 N sulphuric acid in a 500 mL flask for clarification by the prescribed procedure
and the solution is then made up to 500 mL. This solution shows an absorbance of 0.402 at 272 nm. (a) Determine
the molar decadic coefficient. (b) What is the percentage of caffeine in the given sample of instant coffee? Assume
l = 1 cm.
Solution (a) lmax = 272 nm,
A = 0.560,
C = 5.17 × 10−5 mole lit−1
and l = 1 cm.
Substituting these values in the Beer–Lambert’s equation: A = eCl, we get
0.560
ε=
5 17 10 −5
( ) × 1(cm)
= 1.083 × 104 mole−1 lit cm−1
(b) For the unknown sample

A = 0.402
Therefore,
0.402
C=
1.083 × 10 −4
( ) × 1(cm)
= 3.71 × 105 mole lit−1
= 7.86 × 10−3 g lit−1 [since molecular mass = 212]
or
C = 3.93 × 10−3 g/500 mL
= 1.965 × 10−4 g/25 mL
Thus, the amount of caffeine in a 2 g commercial sample of instant coffee = 20 × 1.965 × 10−4 g
= 0.004 g or 0.2 per cent
Problem 5.36: 100 mL of water from an industry is treated with o-phenanthroline and evaporated to 25
mL. The absorbance of this coloured solution has been found to be 0.460. Determine the iron content of
industrial water (in mg lit−1) if the molar extinction coefficient of this coloured compound is 1100 for a path
length of 1 cm.
Solution We have l = 1 cm, e = 1100 mole−1 lit cm−1 and A = 0.460

Substituting these values into the expression for Beer–Lambert’s law, we get
0.460
C= = 4.181 × 10 −4 mole lit −1
1(cm ) × 1100 ( )
or
4.181 × 10 −4 × 55.84 × 25
C= g/25 mL = 5 84 10 −4 g/25 mL
1000
= 5.84 × 10−1 mg/25ml = 5.84 × 10−1 mg/100 ml of industrial water
= 5.84 mg/lit of industrial water [since 25 mL = 100 mL of industrial water]
Problem 5.37: To determine phosphorous in steel by dark-blue colour of a reduced phosphomolybdic com-
plexes on absorbance scale, the following data are observed for standard samples of steel:
Amount of phosphorous in steel (per cent) 0.15 0.024 0.031 0.040 0.055
Absorbance 0.318 0.538 0.722 0.921 1.302
(a) Determine the percentage content of phosphorous in the unknown sample whose reading on the absorbance
scale is 0.619.
(b) Determine also the percentage content of phosphorous from the straight-line expression derived from the
standard data.
Solution (a) Draw a straight-line graph between absorbance and amount of phosphorous in the steel. From the
graph, the amount of phosphorous in the unknown sample of steel corresponding to A = 0.619 has been found to
be 0.028 per cent.
(b) The mathematical expression for the straight drawn in part (a) may be expressed as
A = mC + Int
where A stands for absorbance, m stands for slope and Int represents the intercept. In order to determine m and
Int, select any two nonsuccessive values from the standard data and proceed as follows:
C per cent 0.024 0.040
A 0.538 0.921
0.921 − 0.538 0.383

Thus, m= = = 23.9
0.040 − 0.024 0.016
and A = 23.9 C + Int
In order to determine the value of intercept substitute any value of A and its corresponding value of C
from the standard data into the above expression, i.e.
0.538 = 23.9 × 0.024 + Int
Int = 0.538 − 23.9 × 0.024 = −3.56 × 10−2
Hence, the expression of straight line for the standard data is A = 23.9 C − 3.56 × 10−2
Thus, for A = 0.619
0.619 + 3.56 10 −2
C= = 0.027 per cent
23.9
Problem 5.38: A steel sample weighing 0.0532 g is dissolved in acid, the solution is treated with diethyl
dithiocarbaminate and its volume is made up to 100 ml. The solution is analysed spectrophotometrically at 328
and 368 nm in cell with a path length of 2 cm. The results obtained are A328 = 0.67 and A368 = 0.45. Determine the
percentage contents of nickel and cobalt in the steel if eNi = 35210 and eCo = 3910 at 328 nm and eNi = 21820 and
eCo = 14340 at 368 nm.
Solution The theory of the spectrophotometric method for the analysis of binary mixtures has been described
in Chapter 3 on infrared spectroscopy. Based on the theory we write following expressions:
At wavelength l1 : A1 = e1 C1 l + e2 C2 l and
At wavelength λ 2 2 ε1′ C1l ε ′2 C 2l
i.e. at l328: 0.67 = eNi l CNi + eCo l CCo
i.e. at l368: 0 45 = ε Ni
′ lC Ni + ε Co
′ lC Co
At l328: eNi= 35210 and eCo = 3910
At l368: ε Ni
′ = 21820 d ε Co
′ = 14340
Substitute these values together with the value of l into the above expressions and solve for CNi and CCo.
0.67 = 35210 × 2 CNi + 3910 × 2CCo
0.45 = 21820 × 2 CNi + 14340 × 2 CCo

or
[0.67 = 70420 CNi + 7820 CCo] × 28680
[0.45 = 43640 CNi + 28680 CCo] × 7820
0.67 × 28680 − 0.45 × 7820 = (70420 × 28680 − 43640 × 7820) CNi

1.569 × 10 4
C Ni = mole lit −1 = 9.350 × 10 −6 mole lit −1
1.678 × 10 9
= 9.350 × 10−6 × 58.7 g/lit = 5.488 × 10−4 g/lit
or CNi = 5.488 × 10−5 g/100 mL
Hence, percentage of nickel in the steel sample is
5.488 × 10 −5 × 100
= 0 10
0.0532
Substituting the value of CNi = 9.350 × 10−6 mole lit−1 into the equation in CNi and CCo, we get
0.67 = 70420 × 9.35 × 10−6 + 7820 CCo
0.67 − 0.658
or C Co = = 1.534 mole/lit
7820
= 1.534 × 10−6 × 58.93 g/lit = 9.04 × 10−5 g/lit
or
CCo = 9.04 × 10−6 g/100ml
9.04 × 10 −6 × 100
Thus, the percentage of Co in the given steel sample is = 0.017
0.0532
Problem 5.39: The solution in a dying vat after a hundredfold dilution yields the following absorbance data
in a cell with path length of 5 cm at two wavelengths l1 and l2: before dying, Al1 = 2.2 and Al2 = 1.6; after dying,
Al1 = 0.40 and Al2 = 0.25. Determine the degree of extraction of dyes from the vat if el1 = 3400 and el2 = 1500 for
the dye A and el1 = 1300 and el2 = 2500 for the dye B.
Solution Proceeding as in Problem (5.39), we write

Before dying,
2.2 =17000 CA + 7500 CB
1.6 = 6500 CA + 12500 CB
Solving for CA and CB, we get
CA = 9.465 × 10−5 mole lit−1; CB = 7.88 × 10−5 mole lit−1
Similarly, after dying, we write
0.40 = 17000 CA + 7500 CB
0.25 = 6500 CA + 12500 CB
Solving for CA and CB we obtain
CA = 1.908 × 10−5 mole lit−1 ; CB = 1.013 × 10−5 mole lit−1
Hence, the degree of extraction of the dye A
(9.465 × 1.908) × 10 −5 × 100
= 79.84 per cent
9.465 × 10 −5
( 7.88 − 1.013) × 10 −5 × 100
and of the dye B = = 87.14 per cent
7 88 × 10 −5
Problem 5.40: When incident light is passed through an empty cell placed in a single- beam photoelectric
colourimeter, the galvanometer scale reading is observed at 100 divisions while it is at 80 divisions when the cell
contains a solution having concentration C moles lit−1. What will be the reading in the absorbance scale?
Solution (a) Here Io = 100 divisions and It = 80 divisions. Thus

I 100
A = log o = log = log − log 80
It 80
= 2 − 1.903 = 0.097
(g) Spectrophotometric Titrations A plot of absorbance at a specific wavelength against a volume of titrant
is known as spectrophotometric titration curve and its nature depends upon the nature of titration. The change in
absorbance of a solution is used to follow the change in concentration of a light-absorbing species during titration.
The absorption varies linearly with concentration, i.e. Beer–Lambert’s law is obeyed. Thus, the titration curve
consists of two straight lines with different slopes intersecting at a point called the equivalence point or end point.
The curve may show a sharp angle or curvature near the end point. The curvature may be due to some equilibrium
reactions, and the sharp end point in such a case is obtained by back extrapolation method.
The titration is performed at a wavelength which lies well apart from an absorption maximum to avoid inter-
ference by other absorbing species and the absorbance is adjusted initially by adjusting the sensitivity and slit
width. The optimum concentration of the solution to be analysed depends upon the molar decadic coefficient of
the analyte, i.e. titre. It is of the order of 10−4−10−5 mole lit−1.
A measured volume of the reagent is added to the analyte placed in a cell, called titralyser, directly or by
circulation-type arrangement. The solution is stirred and absorbance is measured. The reading is taken at several
points before and after the end point and true end point is obtained graphically by the intersection of two straight
line segments. Some of the typical titration curves are shown in Fig. 5.64. For accurate results, the absorbance
V +v
‘Aapp’ must be corrected for dilution. It is to be multiplied by where V is the original volume of the solution
V
and v is the added volume. Thus the true absorbance will be given by
⎡V + v ⎤ (5.88)
A true A app ⎢ ⎥
⎣ V ⎦
A A A
(1) (2) (3)
VT VT VT
EP EP EP
A A A
(4) (5) (6)
VT VT VT
EP EP EP
A A A
(7) (8) (9)
VT VT VT
EP EP EP
Fig. 5.64 Schematic diagram showing the shapes of some spectrophotometric titration curves: A-absorbance, VT-volume oftitrant; EP-end
point. The volume correction improves the straight-line behaviour and hence the accuracy of the spectrophotometric titration.
This method is not used for only neutralisations, but also for redox and complex formation reactions. Weak
acids which absorb at different wavelengths from its anion or if either acid or anion does not absorb, can be
titrated by this method. Satisfactory results can be obtained provided the product of molar concentration and
ionisation constant of the acid is greater than about 10−12. Strong acids can’t be titrated by this method because
they are in the ionised state at all times. Some of the examples of spectrophometric titrations are titration of
m- and p-nitrophenol mixture with sodium hydroxide solution at l = 545 nm; mixture of lead and bismuth with
EDTA solution at l = 240 nm; mixture of sodium acetate and o-chloroaniline in glacial acetic acid with HClO4
solution at l = 312 nm, etc.
Problem 5.41: A 25 mL volume of solution containing a mixture of Pb (II) and Bi (III) salts is titrated with a
0.01 M solution of complexion (III) at l = 240 nm. The complexionate of Bi (III) is more stable than that of Pb(II)
and, therefore, two inflextion points appear on the titration curve (ePb.EDTA > eBi.EDTA). Calculate the concentrations
of Pb and Bi if the values of absorbances A1, and A2 before the point of equivalence, corresponding to the volumes
of added titrant, V1 = 1 mL and V2 = 1.8 mL, are known to be equal to 0.15 and 0.25, respectively. The absorbance
of the solutions A′ and A′′, after the first equivalence point corresponding to the volumes of the titrant V1 = 3.0
and V2 = 3.5 mL are equal to 0.75 and 1.00 respectively. The constant value of the absorbance reached at the end
of titration is 1.20. Ignore the effect of dilution upon titration.
Solution Using the data given above we construct a titration curve of the absorbance against volume of the
titrant added, (see Fig. 5.65).
From Fig. 5.65, we get two intersection points of the straight lines at EDTA volumes of 2.35 and 3.80 mL,
respectively.
1.20
0.80
Absorbance
0.40
1.00 2.00 3.00 4.00 5.00

Volume of Titrant (in ml)
Fig. 5.65 A plot between absorbance and volume of titrant EDTA in the EDTA mixture of Pb (II) and Bi (III) system.
Concentration of EDTA = 0.01 mole lit−1 = 0.01 × 328 × 10−3 g/mL = 3.28 × 10−3 g/mL
Volume of EDTA used for complexation with Pb = 2.40 mL
Volume of EDTA used for complexation with Bi = 3.80 − 2.40 = 1.40 mL.
Thus, the amount of Pb in the mixture = 2.40 (mL) × 3.28 × 10−3 (g/mL) = 7.87 × 10−3 g and of Bi = 1.4(mL)
× 3.28 × 10−3 (g/mL) = 4.59 g
(h) Kinetic Method of Analysis This method is used to determine the concentration of catalyst in catalytic
reactions. In this method, the variation of absorbance of a solution with time is measured in the presence of vari-
ous amounts of the catalyst. A graph is then plotted between absorbance and time at different concentrations of
the catalyst and the slopes of the resultant straight lines in the plot are then determined. A calibration graph is
then obtained between slope (tan a) and concentration of the catalyst. Next, we plot a graph representing absor-
bance versus time for the solution with an unknown amount of catalyst and determine the slope, say tan a ′, of the
straight line so obtained. The concentration of the catalyst corresponding to this slope, i.e. tan a ′, is then read off
from the standard graph between slope against concentration of the catalyst.
Problem 5.42: Ruthenium catalyses the reaction of reduction of Fe3+ by tin chloride. For a solution containing
different amounts of ruthenium, there have been obtained the following values of the absorbance of iron rhod-
anide as a function of time.
Time (in min)… 3 4 5 6 7

Absorbance without ruthenium... 0.55 0.50 0.46 0.43 0.40
2.4 × 10−6 g Ru... 0.48 0.42 0.36 0.31 0.26
4 ×10 g Ru...
−6
0.42 0.35 0.28 0.21 0.15
Find the amount of ruthenium in the solution if under the same conditions the values of the absorbance of the
solution vary with time as follows:
Time (in min)... 3 5 7

Absorbance... 0.39 0.32 0.23
Solution
(i) Draw a graph between absorbance and time at various concentrations of the solute and determine the slopes
of the straight lines in the resultant graph.
(ii) Draw a standard curve between slope and concentration of the catalyst.
(iii) Draw a curve between absorbance and time for solution with unknown concentration of ruthenium and deter-
mine the slope of the curve.
(iv) Determine the amount of ruthenium corresponding to the above slope from the standard graph. Our value has
been found to be 1.60 × 10−6 g.
Problem 5.43: The oxidation of phenylenediamine by potassium iodate resulting in the formation of coloured
products is catalysed by tungsten. In a determination by the fixed- time method, the following values of absor-
bance are obtained in 15 minutes at different concentrations of tungsten:
C(mg/mL)... 0 0.25 0.60 0.75 1.00

Absorbance... 0.12 0.38 0.60 0.74 0.93
A 1.5 gram sample of the slags of tungsten production is dissolved in acid, the resulting solution is treated and
diluted to 100 mL. In 15 minutes, the absorbance of 5 mL of this solution is 0.67. Determine the loss of tungsten
in one ton (= 1012 mg) of the slags. The volume of the coloured solution is 100 mL in all the cases.
Solution From the straight line plot between C and absorbance, the value of C corresponding to absorbance
= 0.67, has been found to be 0.635 mg/mL = 65.3 mg/100 mL. Thus, the loss of tungsten in 1 ton
(= 10 mg) of slag =
12
10 2 ( μ ) × 65 3 μg = 435 × 106 μg 435 g
1 5 × 106 μg
(i) Chemical Kinetics Studies Spectroscopy is one of the best methods of following the reaction and determin-
ing the rates when spectra of the products of reaction are different from any of the reactants. Spectra in visible and
ultraviolet regions have an advantage in the case over infrared because here a wide variety of solvents in which
reactions are normally carried out can be used in the visible and ultraviolet spectra. One can either follow the same
procedure as in usual chemical methods of removing the samples periodically and stopping the reaction before
estimation. This, however, does not work with very fast reactions. In such cases, intensity of a certain absorption
peak-either of product, which goes on increasing with time or of the reactant which goes on decreasing with time-
is monitored at different intervals of time. On a double-beam spectrophotometer, one can scan the spectrum re-
peatedly at a fast rate and get more information about the reaction mechanism. Kinetics of redox reactions where
one of the species of reactants or the products is preferably coloured can easily be followed by spectrophotometric
method. This is illustrated here by taking the example of oxidation of ascorbic acid by potassium ferricyanide, i.e.
K3Fe(CN)6 in an aqueous solution (100 mL of an ∼1.0 × 10−3 mole lit−1 solution of K3Fe (CN)6 + ~4 × 10−4 mole
lit−1 ascorbic acid in ~0.01 mole lit−1 nitric acid).
Vitamin C (C6H8O6) oxidises to dehydroascorbic acid (C6H6O6) in an aqueous solution
O O
H H
O OH O OH
CH Oxidizing CH
agent
HO OH CH2OH O O CH2OH
Ascorbic acid Dehydroascorbic acid
Vitamin C plays a vital role in the regulation of biological oxidation–reduction processes (aiding cellular res-
piration by acting as hydrogen transporter). Man cannot exist without consuming vitamin C in our diet (the daily
adult requirement is about 20 mg). It assists the body to withstand bacterial infection and toxins and is essential
for the prevention of scurvy due to antiscorbutic activity. Vitamin C was the first vitamin isolated in the pure state
in 1928 by A Szent-Gyorgyi from a number of natural resources. Later it was synthesised from glucose.
The stoichiometry of the oxidation reaction is
2[ ( )6 ]3− + C6H8O6 → 2[Fe(CN)6]4− + C6H6O6 + 2H+1

The overall reaction is of second order and is governed by the rate expression.
−d (a − x ) dx ⎛ x⎞
= = rate of exchange = k (a − x ) ⎜ b − ⎟ (5.89)
dt dt ⎝ 2⎠
where k is the specific rate constant, a is the concentration of [ ( )6 ]3−, b is the initial concentration of ascor-
bic acid and x is the concentration of [ ( )6 ]3− reacted at any time t. Thus, the concentration of ascorbic acid
x
reacted at any time t is .
2
Integrating the above expression and applying the condition t = 0 when x = 0 and a ≠ 2b, we get
2 b (a − x )
kt = ln
(a b) ⎛ x⎞
a b− ⎟
⎝ 2⎠
Rearranging the terms, we get
(a x ) k (a − 2b ) a
ln = t + ln (5.90
⎛ x ⎞ 2 b
⎜⎝ b − ⎟⎠
2
Now the problem arises how to determine the value of a – x, the concentration of unreacted [Fe (CN)6]3− at any
time by means of a spectrophotometer. Aqueous solutions of [Fe (CN)6]3− are yellow in colour and it has a lmax at
418 nm. The other reactant and product species do not absorb at this wavelength and therefore do not interfere in
the analysis. Assuming that the Beer–Lambert’s law is obeyed and if Ao is the absorbance (at l = 418 nm) of [Fe
(CN)6]3− solution having initial concentration ‘a’ mole lit−1, then the concentration of unreacted [Fe (CN)6]3− at
any time t, is
At
(a x) = × a mole lit −1 (5.91)
Ao
where At is the absorbance of the solution at 418 nm.
x ⎛ Ao − At ⎞ a
Thus, = (5.92)
2 ⎜⎝ A o ⎟⎠ 2
x
Substituting (a − x) and from Eqs (5.91) and (5.92) into Eq. (5.90), we obtain
2
⎡ ⎤
⎢ ⎥ k (a − 2b )
2A t ⎢ 1 ⎥= ⎛ a⎞
ln t + ln ⎜ ⎟ (5.93)
Ao ⎢ ⎛ Ao At ⎞ ⎥ 2 ⎝b⎠
⎢ 2b − ⎜⎝ A ⎟⎠ ⎥
⎣ o ⎦
⎡ ⎤
⎢ ⎥ k a − 2b )
The plot between ln t ⎢
A 1 ⎥ and t will be a straight line with slope ( and intercept
Ao ⎢ ⎛ Ao At ⎞ ⎥ 2
⎢ b − ⎜⎝ 2A ⎟⎠ ⎥
⎣ o ⎦
a
= ln . From the slope we can determine k, the second-order rate constant. Thus simply by monitoring intensity
b
of a certain absorption peak (in UV/V region) at different intervals of time, the rate of chemical reactions can be
followed.
Activation-energy parameters can be determined by carrying out the oxidation at three different temperatures.
1
The value of energy of activation (ΔE#) can be calculated from the Arrhenius plot lnk against . Entropy of
T
activation (ΔS≠), free-energy of activation ( ΔG≠), and heat of activation (ΔH≠) can be computed by employing the
Eyring equation from the absolute rate theory.
ΔH≠ = ΔE≠ − RT
ln k = ln (κT/h) + ΔS≠/R − ΔH≠/RT (5.94)
ΔG≠ = ΔH≠ − TΔS≠
It is to be noted that wherever possible, differential rate expressions for other ordered reactions can also be
derived in a similar fashion.
(j) Stoichiometry of Complexes in Solution Phase The major advantage of spectrophotometric method to
determine the composition of the complex formed between two components, say A and B, is that the equilibrium
is not affected during the course of study.
In order to study equilibrium ( A nB A B n ) involved in the formation of a complex, it is necessary to
determine the concentration of at least one of the species present in the mixture containing known formal concen-
tration of the reactants. The three well known methods which are routinely employed to measure the stoichiometry
of the complexes in the solution phase are the following:
• Mole-ratio Method (or Yoe and Jones Method) Here, a series of solutions are prepared by mixing A and
B as follows: The concentration of A is kept constant while that of B is varied. The absorbance of each solution is
then measured at a designated wavelength, lmax, on a spectrophotometer either in the ultraviolet or visible region.
A graph is then plotted between absorbance on the vertical axis and concentration of the variable component (B)
added on the horizontal axis. The point of intersection of two straight lines in the resultant curves will determine
the stoichiometry of the complex formed between components A and B in the solution phase, i.e. at the point of
interaction A:B::Moles of A:Moles of B.
• Continuous Variation Method (or Job Method) Here, equimolar stock solutions of components A and B
are prepared and a series of solutions containing A and B are then prepared as follows. The molar ratio of two
components is varied simultaneously and the total volume of the resultant solution is always kept constant. The
absorbance of each solution is then measured at a designated wavelength lmax. A graph is then plotted between ab-
sorbance on the vertical axis and VB /(VA + VB), where VB is the volume of the component B added to the volume VA
of the component A such that VA + VB = constant, on the horizontal axis. Generally, but not always, a parabolic-type
curve results. The two linear segments of the curve are back-extrapolated to obtain a definite point of intersection
which is a measure of the stoichiometry of the complex, i.e. A : B :: Moles of A : Moles of B.
It is to be noted that sharpness of the breaks in the curves of both these methods for the identification of a
complex depends on the magnitude of stability constant of the complex.
• Slope Ratio Method The theory of this method has already been discussed in Chapter 3 on infrared spec-
troscopy. Two series of solutions are prepared in this method. In one of the series, varying amounts of the com-
ponent A are added to a large excess of the component B and vice versa in the other series. In both the cases, a
straight-line graph is drawn at designated wavelengths say lmax (A) and lmax (B). The ratio of the slopes of two
resultant lines will correspond to the stoichiometric ratio of the components A and B in the solution, i.e. Slope A :
Slope:Moles of A: Moles of B.
Problem 5.44 Determine the stoichiometry of the coloured compound of Pb2+ and pyridylazoresorcinol (PAR)
by the mole-ratio method from the following data:
Concentration of PAR ( × 105 ) (mol lit−1) Absorbance

0.85 0.18
1.70 0.39
2.50 0.59
3.40 0.78
4.30 0.92
5.20 1.00
6.40 1.04
8.60 1.05
10.70 1.06
12.80 1.04
The constant concentration of lead ions in the solution is 4.4 × 10−5 g-ion lit−1.
Solution Using the data given above, a graph is plotted between absorbance (on Y-axis) and concentration of
PAR (on X-axis). From the break point of straight lines, we obtain Pb : PAR::1:1, i.e. the stoichiometry of the
complex is 1:1.
Problem 5.45 Determine the stoichiometry of a complex of cobalt with nitroso-R-salt from the following
spectrophotometric data obtained at pH = 4.12 and a wavelength of 570 nm.
Volume of cobalt solution (in mL); VA 24.0 15.0 12.0 10.0 7.5 6.0
Volume of nitroso-R-salt(in mL); VB 6.0 15.0 18.0 20.0 22.5 24.0
Absorbance 0.28 0.95 1.14 1.32 1.48 1.29
Equimolar stock solutions (10−3 mole lit−1) of cobalt and nitroso-R salt have been prepared in a buffer of
pH = 4.12 and used in spectrophotometric studies.
Solution
Concentration of cobalt solution = 10−3 mole lit−1
Concentration of salt solution = 10−3 mole lit−1
These solutions have been mixed in different proportions by volume and the total volume is made up to 30 mL.
Applying the molality/normality formula, i.e. M1V1 = M1 V2, we can calculate the molality of cobalt and nitrso-
R-salt in the final solutions.
Thus,
Concentration of cobalt × 104 (in mole lit−1), CA 8.00 5.00 4.00 3.33 2.50 2.0
Concentration of nitroso-R-salt × 104 (in mole lit–1), CB 2.0 5.0 6.00 6.66 7.50 8.00
VB CB
or 0.2 0.5 0.6 0.66 0.75 0.8
VA V B CA CB
Absorbance 0.28 0.95 1.14 1.32 1.48 1.29
Solution Using the data given above, plot a graph between absorbance (on vertical axis) and molar concentration
ratio of cobalt and salt or VB /(VA + VB) (on horizontal axis) and find the breaking point in the resultant curve. This
breaking point in the resultant curve will correspond to the stoichiometry of the complex formed between cobalt
and nitroso-R-salt in a buffer solution of pH = 4.2. Our composition of the complex is [Co]:[Salt] = 1:3
(k) Determination of Dissociation Constant of a Reagent Organic reagents can be considered to be weak
acids, HIn, which dissociate to give H3O+ and In− ions:
HIn + H 2 O H 3O+ + ln −
As weak acids, HIn must satisfy the condition

⎡ H 3O+ ⎤⎦ ⎡⎣In − ⎤⎦ ⎡IIn − ⎤
K= ⎣ or ⎣ ⎦ =
K
(5.95)
[HIn] [HIn] ⎡⎣ H3O+ ⎤⎦
and pK = pH + log
[HIn] (5.96)
⎡⎣In − ⎤⎦
⎡⎣In ⎤⎦
−
It follows that the reagent may exist in more than one form in the solution and the ratio is governed by
[HIn]
the pH of the solution. At very low pH, HIn form of the reagent will be predominant, while at very high pH,
In− form will be predominant. Thus at intermediate pH values, both the forms will exist in solution. Accordingly,
the visible or ultraviolet spectral absorption peaks of the reagent will be shifted when the reagent loses or gains a
proton. This affords an interesting method of determining the dissociation constants of such reagents. The usual
method is to measure absorbance (at the best wavelength, i.e. isosbestic wavelength, selected by measuring the
absorbances of solutions at different pH values) as a function of pH. A graph is then plotted between absorbance
and pH as shown in Fig. 5.66. The pH value midway between the maximum and minimum values of absorbances
is equal to the dissociation constant since according to Eq. (5.96) when [In−] = [HIn], pK = pH.
The numerical method to determine pK from Eq. (5.95) is to calculate the concentrations of In− and HIn. This
is done by measuring the absorbances at the maximum values of lIn– and lHIn respectively.
Thus at lIn−; A ε IIn C IIn l + ε HIn C HIn l
At lHIn: A ′ = ε IIn′ − C ′IIn− l + ε HIn

′ C HIn l
Amax
Absorbance
Aaverage
Amin
pKa
0 2 4 6 8 10 12 14
pH
Fig. 5.66 A plot between absorbance and pH.
where ε In , ε HIn and ε In′ − , ε HIn

′ can be obtained from Beer–Lambert’s plot at lIn− and lHIn respectively. The solu-
tion of these simultaneous equations will give the values of CIn– and CHIn. These values on substitution into Eq.
(5.95) will give the value of pK.
The other nongraphical method to determine pK is to measure the absorbances at very low pH (i.e. AHIn), at
very high pH (i.e. AIn–) and at intermediate pH (i.e. Amix) values.
These values when substituted in the formula (5.95) yield the value of K and hence pK since
pK = − log K
⎡In − ⎤ ⎡ H 3O+ ⎤⎦ ⎡ A − A HIn ⎤

K = ⎣ ⎦⎣ = ⎢ mix ⎥ ⎡⎣ H 3O ⎤⎦
+
(Eq. 5.96)
[ ]
HIn A
⎣ In − − A mix ⎦
Problem 5.46 Determine the dissociation constant of a reagent HR if at pH = 6.35 its absorbance is 0.360 and
at pH = 3.0 and 11.8 its absorbances are respectively, equal to 0.890 and 0.025.
Solution To solve the problem we make use of the equation
⎡ A − A HR ⎤ +
K HR = ⎢ mix ⎥ ⎡⎣ H ⎤⎦ (5.97)
⎣ A R − − A mix ⎦
Substituting the values given in the problem into Eq. (5.97) we get
pH = 6.35; [H+] = 10−6.35 = 4.46 × 10−7
0.360 − 0.890 0.53 × 4.46 × 10 −7

K HR = ⎡⎣ 4.460 × 10 −7 ⎤⎦ = = 7 06 × 10 −7
0.025 − 0.360 0.335
(l) Study of Keto-enol Tautomeric Equilibria Keto-enol tautomeric equilibria are shown by b-diketones.
The enolic form is stabilised by H-bonding. Acetyl acetone and ethyl acetoacetate are the classic examples of
keto-enol tautomerism. The tautomeric equilibria are temperature and solvent dependent.
O O O H O
CH3 C CH2 C CH3 CH3 C CH C CH3
l max (H2O) = 274 nm, (emax = 2050)

l max (iso-octane) = 272 nm, (emax = 12000)
O O O H O
CH3 C CH2 C C2H5 CH3 C CH C C2H5
l max = 275 nm, (emax = 16) l max = 244 nm, (emax = 16000)
The percentage content of the keto-enol forms in an equilibrium mixture can be deduced from the physical
properties of the mixture since the keto and enol forms (enol form is conjugated) show different light absorption
in the ultraviolet region and have different refractive indices. They will also exhibit different NMR signals. Since
absorbance is directly dependent on the concentration (i.e. A = e C l), the amount of tautomers in various solvents
and at different temperatures is determined from the intensity measurements at the absorption maximum of the
respective tautomer.
In acetyl acetone, enol tautomer is about 80 per cent in pure liquid, 15 per cent in a aqueous solution and 91–92
per cent in vapour phase and in solutions of nonpolar solvents. In ethylacetoacetate, enolic form is 8 per cent in
pure liquid, 51 per cent in hexane, 32 per cent in ether and 12 per cent in alcohol. The percentage of enol tautomer
in pure liquid ethylbenzoylacetate is 80 per cent.
(m) Study of Charge-Transfer Complexes The molecular complexes of charge transfer (CT) type are a
restricted case of the broad class of Electron Donor Acceptor (EDA) complexes or adducts, where partial or com-
plete transfer of negative charge occurs from an electron donor (D) to an electron acceptor (A). The complex can
be described by the resonance states
DA ↔ D+ A−
Covalent ionic
The formation of these complexes is explained by the quantum mechanical theory of Robert S Mulliken. His
contributions concerning chemical bonds and the electronic structure of molecules by molecular orbital method
were rewarded with a Nobel prize in chemistry in 1966, According to him, if ψ(DA) is the wave function of
no-bond case and ψ(D+ A−) is the wavefunction for ionic case involving transfer of charge, then a combination of
these two forms yields two wave functions ψ0 and ψ1, for the charge transfer complex, i.e.
ψ0 = ψ(DA) + l ψ (D+ A−)
ψ1 = ψ(DA) + l′ ψ (D+ A−)
where l2 is a measure of the amount of charge transfer, l′ > l. One of these functions, say ψ0, corresponds to an
energy lower than that of ψ(DA) or ψ(D+ A−). The transition between ψ0 and ψ1 states gives rise to a charge-transfer
band since charge is transferred from D to A in this process. Valence bond theory is favoured for weak complexes
while molecular orbital theory is favoured for strong complexes. The tendency of charge-transfer complexation
and energy change in the transition depend upon the ionisation potential of the donor and electron affinity of the
acceptor. The smaller the ionisation potential of the donor and larger the electron affinity of the acceptor, the
stronger will be the charge-transfer complex. Charge transfer occurs primarily in the excited state and is stabilised
relative to the ground state by electrostatic interaction between the separated charges on the donor and the accep-
tor. In H-bonded adducts, the electrostatic contribution is predominantly to the ground state of the complex.
The formation of molecular complexes can occur in the liquid/solution, solid or vapour phase. The solid state
CT complexes are also sometimes called organic metals, as these solids exhibit high electrical conductivity,
comparable with those of metals, though most organic solids are insulators, e.g. Quinolinium-TCNQ is the best
organic metal and MEM (TCNQ)2; where MEM = N-methyl-N-ethylmorpholinium behaves as a metallic conduc-
tor at 335 K. It is to noted that donor and acceptor molecules complexed in solid state consists of stacks of alter-
nate donor and acceptor molecules so that each donor has two acceptor neighbours and vice versa. The formation
of complexes may be proved and their composition established by study of spectral properties like absorption of
UV and visible radiations, absorption of infrared radiations, absorption of radio-frequency radiations under the
influence of applied magnetic field, absorption of microwave radiations under the influence of applied magnetic
field, absorption of g-radiations, diffraction and scattering of radiations, etc., and nonspectral properties like melt-
ing point (phase diagrams), dipole moment, conductance, etc.
The charge-transfer process is an instantaneous process and the formation of a CT complex is characterised by
the appearance of a new absorption band in the UV or visible region of the spectrum, e.g. benzene–Iodine system
shows a band at 37300 cm−1 ( ≈ 268 nm) with emax = 1650 in the vapour phase at 25°C while in CC14 solution at
33700 cm−1 ( ≈ 297 nm) with emax = 15400 at 25°C. Diethylether-iodine system exhibits a new band at 42700 cm−1
(≈ 234 nm) with emax = 2100 in the vapour phase at 25 °C while in CCl4 and n-heptane solutions at 40200 cm−1
(≈ 249 nm) with emax = 4700 at 20 °C and at 39700 cm−1 (≈ 252 nm) with emax = 5650 at 25 °C respectively. In
intramolecular CT complexes, the intensities of transitions are independent of concentration, the absorption band
in such complexes arises as a result of charge-transfer interactions across space, e.g. interaction, in nicotinamide
adenine-dinucleotide (NAD+) occurs between the pyridinium ring and the adenine moiety. The mere production
of a colour when two reagents are mixed has often been taken as the sole evidence for CT complex formation,
e.g. TCNE (Tetracyanoethylene) when mixed with benzene forms a bright orange CT complex, all the higher
condensed aromatic hydrocarbons anthracene, phenanthrene and acenes (condensed hydrocarbons with 4-, 5-,
6- and 7- benzene rings) readily form intensely coloured charge transfer complexes with trinitrobenzene, picric
acid, TCNE, tetranitromethane, etc. However more positive evidence, such as systematic variation of the energy of
the absorption band for complexes of a series of related donors with a given acceptor compared with other series
of complexes with other acceptors or with ionisation potential of the donors is desirable before attributing the
production of colour to the formation of a CT complex. Sometimes the instant appearance of colour may be due
to the decomposition of the CT complex instantaneously to give coloured products. In some cases the absorption
bands develop slowly and it is attributed to CT complexation. There is no evidence that CT complexes are formed
slowly. Most probably such absorptions arise from products of normal covalent chemical reactions.
(i) Multiple Charge Transfer Transitions The appearance of multiple charge-transfer band maxima is due to
the participation of more than one energy level of the donor. The energy difference between the two charge-trans-
fer transitions in the complex is expected to be equal to the energy difference between the two appropriate levels
of the donor and the same has been observed indeed. The energy difference between two observed transitions in
a crystalline 1, 3, 5-trinitrobenzene-anthracene complex is 8000 cm−1. This conforms to the calculated frequency
difference of the two highest filled orbitals of anthracene, i.e. 8210 cm−1. The difference between the charge trans-
fer bands of complexes of a given donor is independent of the particular acceptor, especially in cases where more
than two maxima have been observed as is evident from Table 5.27. The double maxima may also result due to
transition from the highest filled level of the donor to two vacant levels in the acceptor, e.g. the energy difference
between two observed transitions in hexamethyl benzene-1, 2, 4, 5-tetracyanobenzene in CHC12CHC12 has been
found to be 6400 cm−1 and it conforms to the calculated energy difference between the two lowest unfilled levels
in 1, 2, 4, 5-tetracyanobenzene. It is to be noted that the multiplicity cannot arise due to transitions from two donor
levels since the highest filled level of hexamethylbenzene is doubly degenerate. The energy difference between
the two maxima in such cases is independent of the nature of the donor for a particular acceptor as is evident from
Table 5.27.
Table 5.27 Typical examples of multiple charge-transfer band maxima.
Donor Acceptor Solvent ν1 (in cm−1) ν2 (in cm−1) ν3 (in cm−1) Δν1/2 (in cm−1)
Biphenyl TCNE CC14 20000 25700 5700
Biphenyl CA CCl4 23000 28700 5700
Anthracene TCNE CC14 14500 17100 21900 2600
Anthracene CA CC14 17200 19800 2600
Chrysene TCNE CC14 15900 18800 22900 2900
Chrysene CA CC14 18500 21500 3000
Mesitylene TCNB Pure solid .28300 33700 5400
Hexamethyl benzene TCNB CHCl2CHCl2 23500 29900 6400
Naphthalene TCNB CHCl2CHCl2 25000 731300 > 6300
Pure solid 24600 31500 6900
N, N′-dimethyl TCNB CHCl2CHCI2 18300 25000 6700
aniline
N, N, N′,N′-tetra TCNB CHCl2CHCl2 13900 20300 6400
methyl-p-phenylene-
diamine
TCNE-, Tetracyanoethylene; CA-Chloranil, TCNB-1, 2, 4, 5-Tetracyanobenzene.
• Estimation of Solvent Polarity or Effect of Solvent on the Energy of

the Charge-Transfer Band The energy of the charge-transfer band for COOH Ph
the complex formed between the iodide ion and a pyridinium ion is very
Ph
sensitive to solvents. The large dipole moment is perpendicular to the plane
of the pyridinium ring in the ground state of the complex which is effec-
tively an ion pair (I), which lies in the plane of the ring in the excited state
of complex. Solvating molecules orient themselves about the complex (the N+ I− Ph N+ Ph
so-called cybotactic region) to minimise the energy of the ground state.
According to Franck−Condon principle, there will be no solvent reorien- CH2CH3
tation at the instant of excitation. Thus the cybotactic region will tend to
increase the energy of the excited state. The greater the solvating power of (I)
Ph Ph
the solvent, the greater will be the decrease in energy of the ground state
and greater the increase in the excited state compared with the ideal sys-
O−
tem of a single isolated complexed pair in vacuo, i.e. gas phase, and hence
(II)
greater the hypsochromic shift. The solvent ionising power is expressed by,
Kosower in terms of a parameter called ‘Z-value’ which is defined as the
energy in kcal mole−1 of the peak maximum of the charge-transfer band of 4-carbomethoxy-1-ethylpyridinium
iodide in that solvent. This particular pyridinium iodide is chosen because of its greater solubility in the less sol-
vating solvents. Some Z-values are compared with the energies ET* (in kcal mole−1) of the peak maximum of the
intramolecular charge transfer absorption band of the pyridinium-N-phenolbetaine (II) which is a solvent sensi-
tive transition and is proposed as an alternative measure of polarity of the solvent in Table 5.28.
Table 5.28 ET* and Z values of some solvents at 298 K.
Solvent ET* (kcal mole−1) Z (kcal mole−1 )
Water 63.1 94.6
Formamide 56.6 83.3

Ethanol 51.9 79.6
Acetonitrile 46.0 71.3
Dimethylformamide 43.8 68.5
Acetone 43.2 65.5

Dichloromethane 41.1 64.2
Pyridine 40.2 64.0
Chloroform 39.1 63.2
Chlorobenzene 37.5 58.0
Dioxane 36.0
Hexane 30.9
2, 2, 4-Trimethylpentane 60.1
ET*values are energies, in kcal mole of the peak maximum of the intramolecular charge transfer
−1
band of the phenolbetaine.
• Determination of Equilibirium Constant Let us Consider the charge-transfer equilibrium

K
D+A DA ↔ D + A −
The equilibrium constant of 1 : 1 CT complex formed between D and A is then given by.
K=
[ ] =
C DA
(5.98)
[ ][ ] CD CA
where CDA, CD and CA are the concentration of complex, donor and acceptor respectively at equilibrium.
The general procedure is to measure the absorbance A of the complex (at λ max
DA
) either as a function of concen-
tration of donor or acceptor.
Thus,
Aabs = eDA lCDA (5.99)
If C°D and C°A are the initial concentrations of D and A respectively such that C°D >> C°A then Eq. (5.98) can
be written as
C DA
K= (5.100)
(C D C DA C )( A C DA )

C oA 1 1 C oA C
= o
+ o
+ 1 − DA (5.101)
C DA K C D C D C oD
Equating Eqs. (5.99) and (5.101), we get
C oA 1 1 Co C DA 1
= + o A − + (5.102)
A abs K ε DA l C D C D DA
o
D l ε DA C D l ε DDA l
o
Since C°D >> C°A and C°D and C°A are very small, the second and third terms of Eq. (5.102) can be neglected
so that
C oA 1 1 1
= + (5.103)
A abs ε DA l C D
0
DA l
D
Similarly, if C°A >> C°D, then Eq. (5.102) becomes

C oD 1 1 1
= + (5.104)
A abs ε DA l C A
o
DA l
D
Both Eqs (5.103) and (5.104) are called Benesi−Hildebrand equations.

C oA C oD 1 1 1
Thus, a plot between and o o should give a straight line with slope = and intercept
A abs A abs C D C A K ε DA l
1
= .
ε DA l
From the slope and intercept, both K and eDA can be obtained.
For example, the spectrum of iodine–mesitylene system in CC14 is shown in Fig. 5.67
The established equilibrium is given by
I 2 + (CH 3 )3 C6 H 3 (CH3 )3 C6 H3 ...I 2

Mesitylene does not absorb in the visible region. The CT complex formed absorbs at the violet end (332 nm) of
the visible spectrum while unreacted iodine absorbs at 500 nm, From the absorbance measurements at lmax = 332
nm and employing Eq. (5.103), the value of equilibrium constant and emax has been found to be 0.82 (lit mole−1)
and emax= 8850 lit mole−1 cm−1 respectively.
The equilibrium constants of some iodine-donor systems in CC14 at 25°C are given in Table 5.29 and of some
metalocene donor-acceptor systems, in Table 5.30.
Mes − I2 complex I2
Absorbance
200 300 400 500 600

Wavelength(nm)
Fig. 5.67 The absorption spectrum of iodine-mesitylene system in CCI4 at 25°C.
Table 5.29 Equilibrium constants of CT complexes of iodine with donors in CCl4 at 25°C.
Donor lmax (in nm) emax(lit mole−1 cm−1) K (lit mole−1)
Benzene 292 16400 0.15

Toluene 302 16700 0.16
Mesitylene 332 8850 0.82
Hexamethylbenzene 375 8200 1.35
Trimethyleneoxide* 452, 248 26.1, 25.2
Tetrahydrofuran* 455, 249 17.2, 20.0
Diethylether* 462, 252 5.9, 6.6

*- in n-heptane
Table 5.30 Association constants of some selected metallocene CT complexes.
Donor Acceptr Solvent Temp νCT K emax −ΔH −ΔS
(°C) (cm−1) (lit mole−1 cm−1) (kcal mole−1) (kcal mole−1 deg−1)
Ferrocene TCNE Cyclohexane 30 10000 30.0 474 3.8 5.5
Ferrocene TNB CH2ClCH2Cl 23 18700 2.8 630 1.8 4
Chloromer- TNB CH2ClCH2Cl 23 19400 .75
curiferrocene
TCNE-tetracyanoethylene, TNB-1, 3, 5-trinitrobenzene. Solid CT complexes are formed between ferrocene donor and acceptors such as p-benzoqui-
none, chloranil, tetracyanoethylene and DDQ. Cobaltocene forms solid complexes of ionic type with chloranil, tetracyanoethylene and DDQ: DDQ-2,
3-dichloro-5, 6-dicyano-p-benzoquinone. The latter being salts, are easily oxidisable.
Fe Fe A
(I) (II)
Fig. 5.68 (I) Filled p-orbital of a cyclopentadienly ring (II) nonbonding electrons of metal.
The acceptor in metallocene complex may interact either with the filled p-orbital of a cyclopentadienyl ring
(I) or with nonbonding electrons of the metal (II) as shown in Fig. 5.68. The x-ray diffraction of single crystal of
ferrocene-tetracyanoethylene complex shows that the geometry corresponds to (I) and not (II), Mössbauer studies
also conform to (I) and not (II).
Ferrocene/Nickelocene/Ruthenocene/Osmocene-TCNE systems in more polar solvents such as acetonitrile
show similar results, i.e. solvation in the first system causes ionisation into ferricinium ion and the radical ion of
tetracyanoethylene.
(ii) Chemical Reactions via Charge Transfer Complexes The molecular interactions between electron
donors and acceptors are generally classified into two types of reactions: The first type includes the formation of
a weak charge-transfer complex (also known as outer complex or p-complex) and the second type includes the
nucleophilic substitution reaction either via outer and inner complexes as intermediates or outer complexes and
ionic species as intermediates. Either of the reaction paths listed above depend upon mostly on the electron donor
characteristics, concentration of donor, temperature, permeativity of the solvent medium.
Accordingly, a CT band will appear due to 1: 1 CT complex formed between D and A in a nonpolar solvent. The
intensity of the band will not change with time suggesting thereby that the CT complexation proceeds according
to the scheme.
(DA )EDA ↔ (D )
Fast
D A D
A
However, when donor is mixed with an acceptor in a solvent of high dielectric constant, the intensity of the CT
band may decrease with time and simultaneously a new band with progressively increasing intensity may appear.
Concurrently, the electrical conductivity of the solution will also increase rapidly and the solution will become
paramagnetic and may give ESR signals corresponding to paramagnetic anionic and cationic species. This may be
due to the formation of inner ionic CT complex as an intermediate according to the scheme.
fast
D+A ( ) EDA
D ⎯slow
⎯⎯ → D+ A− (i)
Outer Complex Inner CT Comple
C x
The inner CT complex may further undergo reactions to yield isolatable or unisolatable products.
D+ + A − ⎯fast
⎯→ Products (ii)
Inner Complex
The outer CT complex may also yield products directly, i.e.
(DA) EDA
→ Products (iii)
Outer CT complex
Some typical examples concerning reactions via charge-transfer complexes according to the above schemes
are described here.
When a donor and accepter are mixed in a highly polar solvent, ionisation of the nonionic CT complex will take
place. The driving force is the energy of solvation of the ions formed. The chemical reaction
+ −
DA + Solvent Dsolv A solv
is analogous to the ionisation of HCl in a polar solvent, e.g. the CT complex of N, N, N′ N′- tetramethyl -p-phe-
nylenediamine (TMPD) and tetrachloro-p-bezoquinone (chloranil) when dissolved in cyclohexane shows a broad
band due to CT complex at 870 nm. This band is absent in methanol solution. However, absorptions at 568 nm,
618 nm characteristic of the cation radical TMPD+ are observed, together with absorptions at 426 nm, 452 nm
due to semiquninone ion derived from chloranil. The assignments have been confirmed
from the spectra of methanolic solution of TMPD+ perchlorate salt which shows bands
O−
at 568 and
618 nm and methanolic solution of the product obtained by the action of sodium iodide Cl Cl
on a solution of chloranil in acetone which shows bands at 426, 452 nm. It is pertinent to
note that initially TMPD−CA complex ionises in solution to form D+ and A − , i.e. TMPD+
and the dinegative ion of chloranil (I).
Cl Cl
No ESR absorption is obtained for solution of TMPD−CA in solvents of low ionising
power such as chloroform. However, a strong absorption is observed for acetonitrile solu-
O−
tion immediately after mixing. This consists of singlet due to chloranil semiquinone ion,
(I)
superimposed on a set of thirteen triplets attributed to TMPD+ as shown in Fig. 5.69 (a).
−
This absorption is charaterised by ESR spectrum of TMPD+ ClO 4 . The relative instabil-
ity of chloranil semiquinone ion compared to TMPD+ is the collapse of singlet absorption in the aged solution as
shown in Fig. 5.69 (b).
(a) (b)
Fig. 5.69 ESR spectra of TMPD-CA system in acetonitrile (a) immediately after mixing, and (b) aged solution.
Ethanol and methanol form 1 : 1 CT complexes with iodine with lCT ~ 225 and 220 nm, respectively. The
intensity of CT band decreases with time, and simultaneously progressive intensification of triiodide ion absorp-
tion at 297 and 363 nm is observed. Concurrently, the conductance of solution of iodine in ethanol/methanol also
increases with time and ultimately becomes constant. The transformation of initially formed 1 : 1 outer CT com-
plex of alcohol with iodine to inner CT complex followed by very fast reaction of inner CT complex with iodine
to form triiodide ion takes place. according to the scheme:
fast
a
R + I2 R I2 ⎯sslow
⎯⎯ → R .I δ + ...I δ −
Outer CT complex Inner CT complex
fast
a −
RI δ ... I δ I −I R .I + + I 3
where, R = CH3 OH or C2H5OH.

The rate-determining step is monomolecular transformation of the outer CT complex to inner CT complex.
The specific rate constant for the monomolecular transformation can be obtained by making use of the first-order
rate constant equation.
2.303 A Ao
k= log ∞ (5.105)
t A∞ At
Here, A0, At and A∞ are the absorbances measured immediately after mixing iodine in an alcohol, after time t
and at the end of the reaction respectively.
Form the straight-line plot between log(A∞ − At) and time, one can determine the value of k from the slope.
Note that the absorption measurements as a function of time are either carried out at the absorption band of CT
complex or at the product band. In case of ionic reactions, the specific rate constant can also be obtained from the
expression
σ ∞ σ t1
ln = kt + φ (5.106)
σ ∞ σt2
where, s∞, σ t1 σ t 2 are the conductances at time t∞, t1, and t2 respectively and f is the intercept due to time lag.
TCNE with N, N-dimethylaniline in chloroform instantaneously forms a blue charge-transfer complex (I) with
absorption maxima at 675 nm. The intensity of the band gradually decreases as an s-intermediate (II) develops
and the final product has been isolated as a solid and is identified as N, N-dimethy1-4-tricyanovinyl aniline (III).
The kinetic data for the reaction is consistent with the bimolecular reaction of the CT complex with a molecule of
the free aniline. The reaction proceeds according to the scheme.(S-I)
The reaction of N-methylaniline with TCNE in chloroform follows a sequence analogous to the reaction involv-
ing N, N-dimethyl aniline except that the decomposition of CT complex is fast and formation of final product is
slower than in the reaction of N, N-dimethylaniline. A similar sequence is observed for the reaction of indole with
TCNE in dichloromethane. The final product has been identified as 3-tricyanovinyl indole.
When chloranil is directly dissolved in excess of N, N-dimethylaniline, a broad band appears at 650 nm and
is attributed to the formation of a neutral CT complex (I). The intensity declines with time and simultaneously a
new band develops at 410 nm. Concurrently, the conductivity of the solution grows rapidly and an ESR absorp-
tion band corresponding to the chloranil semiquinone ion (II) is observed. This band assignment conforms to
ESR band of solid sodium-chloranil compound as well as of electrolysed solutions of chloranil. The absence of
NMe2 NMe2
NC CN NC CN
C C
+
C C
NC CN NC CN
(I)
CN NC CN
−C NC
C CN NC CN CN
−
NC C C NC C
H C
NC
+
NMe2 NMe2 NMe2
(III) S-I (II)
ESR signal due to N, N-dimethyl-aniline cation (III) at any stage of the reaction might be due to broadening of
the signal through rapid electron exchange reaction between radical cation of N, -N dimethylaniline and the larger
excess of neutral N, N-dimethylaniline molecules. After a number of days, the absorption at 410 nm decreases
and somewhat later the intensity of ESR signal also decreases. The conductivity of the solution however does
not change. The aged solution contains crystal violet as its cation (IV). Thus this reaction via CT complexation
according to Kosower proceeds according the scheme (S-II).
Problem 5.47: When solutions of N-heterocyclics, i.e. pyridine (Py). a-piocoline (a-P), b- picoline (b-P),
g-picoline (g-P) and fluoranil (FA) in chloroform (concentration of N-heterocyclics being in far excess than that
of fluoranil) are mixed, an absorption band appears around ~ 340 nm. The intensity of the band increases with
time and concurrently the conductivity of the solution grows rapidly. The absorbance of N-hetrocyclics-fluoranils
systems at different time intervals in CHCl3 solution at 298 K is given in Table 5.31, Assuming that 1 : 1 complex
is formed in these systems and no products are isolated in the course of the reaction as well as from the final
solutions,
(a) suggest the tentative mechanism in these systems, and
(b) calculate the rate constants for the reaction in these systems.
Solution
(a) We know that the formation of DA complex of CT type is an instantaneous process and band positions
remain unaltered on changing the electron donor in these systems. The absence of absorption band characteristic
of CT complex is due to very low concentration of DA complex left at any moment due to its immediate conver-
sion to the final product according to the following scheme.
fast
N heterocyclics + FA DA Complex slow
→ Products
Since the position of the band does not change on changing the donor, the reaction species, i.e. product or a DA
complex formed are structurally similar. The appearance of a band at around ~340 nm and concurrent increase in
conductivity of the solution suggests the formation of ionic products.
(b) The specific rate constants k at the products bands have been calculated using Eq (5:105)
2.303 A Ao
k= log ∞ (Eq. 5.105) System k × 103
t A∞ At
Py − FA 7.53
where the terms and symbols have their usual meanings. The straight-line a − P − FA 10.60
plot between log (A∞ − At) and t for various systems is shown in Fig. 5.70.
b − P − FA 16.80
The values of specific rate constants at 298K as determined from the slope of
g − P − FA 26.80
straight lines in Fig. 5.70 are
•+
NMe2 O NMe2 O NMe2 O−
Cl Cl Cl Cl Cl Cl
+ + +
Cl Cl Cl Cl Cl Cl
O O O•
(I) (II) (III)
Me CH2 NMe2 NMe2 Me CH2

N N+ OH
Cl Cl
+
O Cl Cl
Cl Cl
O
Cl Cl
O
Me +• CH2 NMe2 Me + CH NMe2

N O− N OH
Cl Cl Cl Cl
+ +
NMe2
Cl Cl Cl Cl
O• O−
NMe2
NMe2 NMe2
Me C NMe2 Me CH
N N
Chloranil
NMe2
H
NMe2
O
Cl Cl NMe2
Cl Cl NHMe O−
OH Cl Cl
Me2N C+ + +
Cl Cl
OH
NMe2
(IV)
S-II
Table 5.31 The absorbance of N-heterocyclic − FA systems at different time intervals in CHCl3 solution at 298 K.
Time (min) Absorbance
Py − FA α − P − FA β − P − FA γ − P − FA
348 nm 341 nm 338 nm 339 nm
0.0 0.7661 0.6490 0.6440 0.6010

1.0 0.7676 0.6518 0.6485 0.6046
2.0 0.7706 0.6545 0.6539 0.6123
3.0 0.7718 0.6590 0.6578 0.6157
4.0 0.7739 0.6610 0.6626 0.6160
5.0 0.7765 0.6650 0.6654 0.6201
6.0 0.7776 0.6675 0.6707 0.6232

7.0 0.7787 0.6705 0.6758 0.6268
8.0 0.7822 0.6740 0.6804 0.6303
9.0 0.7837 0.6765 0.6833 0.6332
10.0 0.7844 0.6800 0.6877 0.6363
11.0 0.7871 0.6825 0.6918 0.6389
12.0 0.7891 0.6860 0.6959 0.6415
13.0 0.7904 0.6890 0.6992 0.6449
14.0 0.7916 0.6920 0.7029 0.6468
15.0 0.7916 0.6950 0.7052 0.6485
16.0 0.7940 0.6975 0.7094 0.6510
17.0 0.7968 0.7000 0.7123 0.6539
18.0 0.7983 0.7025 0.7159 0.6566
19.0 0.7999 0.7065 0.7186 0.6567
20.0 0.8027 0.7080 0.7226 0.6591
∞ 1.0370 0.9490 0.9194 0.7480
5.20.6 Biological Applications

Biological applications of UV/V spectroscopy are centred around (i) quantitative analysis of biological samples
from plant and animal systems in vivo and vitro carbohydrates, lipids, aminoacids and proteins, enzymes, nucleic
acids, vitamins, pigments, phenolics, etc., and (ii) the study of inter- and intra-molecular interactions in biological
systems. It has been suggested that intramolecular charge-transfer interactions across space may be involved in the
structures of some biological macromolecules and various cellular systems and the intermolecular interactions are
involved in biochemical reactions. Some of the typical biological applications of this tool are reported here.
0.51 0.60
0.53 0.70
0.55 0.80
−log(A∞-At)
−log(A∞-At)
0.57 0.90
0.59 1.00
0.61 1.10
0.63 1.20
0.0 5.0 10.0 15.0 20.0

Py, a–P β–P, γ –P
Time (min)
Fig. 5.70 Pseudo first-order rate plots for the N-heterocyclic-FA systems in CHCI3 solution at 298 K • − Py. 340 nm; Δ−a− P, 341 nm,
▲−b−P, 338 nm; o–g−P, 339 nm.
(a) Determination of Chlorophyll and Carotenoids in Plants All green vegetation contains the alcohol-
soluble pigment chlorophyll. It is always associated with the carotenoids, carotene and lutein. The carotenoids
act as photosensitisers in conjugation with chlorophyll. With the help of chlorophyll, plants assimilate carbon
dioxide from the atmosphere converting it into water, oxygen and carbohydrates. Chlorophyll actually consists of
two individual substances−chlorophyll ‘a’ and chlorophyll ‘b’ differing somewhat in colour. The only difference
between them is that the chlorophyll ‘a’ contains methyl group while an aldehyde group is present in the chloro-
phyll ‘b’. Both chlorophylls are esters. A complete synthesis of chlorophyll was established in 1960 by Robert
B Woodward and he was rewarded the Nobel prize in chemistry for his outstanding achievements in the art of
organic synthesis. The chlorophyll content is estimated by the formula of Anderson and Boardman while total
carotenoids are computed by the formula reported by Jenson and Jenson.
V
Chlorophyll ‘a’ (in mg/g tissue) = (12.7 A663 − 2.69 A645) × (5.107)
1000w
V
Chlorophyll ‘b’ (in mg/g tissue) = (22.9 A645 − 4.68 A663) × (5.108)
1000w
H3C CH CH2 H 3C CH CH2
HC CH HC CH
N N
H3C CH3 H 3C CHO
N Mg N N Mg N
H2C C2H5 H 2C C2H5
N N
C20H39OOC H2C C CH H 2C C CH
HC C20H39OOC HC
CH3 CH3
O O
O O
C C
OCH3 OCH3
Chlorophyll a Chlorophyll b
CH3 CH3
where C20H39 = CH2 CH C CH2 (CH2 CH2 CH CH2)3 H
Phytol residue
(I)
Total chlorophyll (in mg/g tissue) = (20.2 A645 + 8.02 A663) × V (5.109)
where, V = Total volume of solution made (in mL) 1000w
w = Fresh mass of sample (in g)
54A 420 × chlorophyll ’aa’ + 10.1 × chlorophyll ’b’
Caroteniods (inn mg/g tissue ) = (5.110)
2185
Problem 5.48: 50 mg of green leaves crushed in 2 mL of 80 per cent acetone was centrifuged. The supernatant
was collected in a 5 mL flask. The residue was again treated with 2 mL of 80 per cent acetone and then centri-
fuged. The supernatant was again transferred to a 5 mL flask and the total volume was brought up to 5 mL with
80 per cent acetone. The absorbance of the above extract is measured at 663 nm, 645 nm, and 420 nm using 80
per cent acetone as blank and the following data are obtained.
A663 = 0.821, A645 = 0.345 and A420 = 0.765
Calculate (a) the percentage of chlorophyll ‘a’, chlorophyll ‘b’, and (b) total carotenoids in the sample under
study. Note that the amount of these species in infected leaves differs from that of healthy leaves. The amount of
chlorophyll and carotenoids in leaves also differs from that of stem of plants.
Solution (a) Substituting the given values in the respective formulae, we obtain chlorophyll ‘a’ (in mg/g tissue)
=
(12.7 × 0.821 − 2.69 × 0.345) × 5 = (10.426 − 0.928) = 0.995
50 × 10 −3 × 1000 10
(22.9 0.345 − 4.68 0.821) × 5 (7.900 − 3.842) = 0 40

Chlorophyll ‘b’ (in mg/g tissue) = −3
=
50 × 10 × 1000 10
(20.2 0.345 + 8.02 0.821) × 5 (6.969 + 6.584) = 1 35
Total chlorophyll (in mg/g tissue) = =
Thus, 10 10
0 95
Chlorophyll ‘a’ = × 100 = 70.4 per cent
(0.95 0.40)
and
0 40
Chlorophyll ‘b’ = × 100 = 29.6 per cent
(0.95 0.40)
(b) Substituting the given value of A420 together with the values of chlorophyll ‘a’ and chlorophyll ‘b’ from Part
(a) into Eq. (5.110), we obtain
54 × 0.765 × 0.95 10.1 × 0.40
Total carotenoids =
2185
39.244 + 4.040
= =1.980 × 10−2 mg/g tissue.
2185
(b) Estimation of Tyrosine and Tryptophan in Intact Protein We know that a-amino acids are of great
biological importance and are the building blocks of proteins. They can be derived from natural materials by the
hydrolysis of proteins present in meat, skin, gelatin, wool, hair, feather, and also the proteins of protoplasm and
nuclei of plants and animal cells, casein from curds, a number of hormones such as insulin, enzymes (e.g. pepsin)
etc. UV/V spectroscopy is used to determine the amount of tyrosine (tyr) and tryptophan (try) in intact proteins.
Pure tyrosine shows band at lmax = 294 nm while tryptophan exhibits maximum absorption at lmax = 280 nm in
0.1 N NaOH or KOH. Both the bands overlap to some extent, i.e. both acids in solution will contribute to the
absorption at lmax = 294 nm as well as at lmax = 280 nm. From theory, we write
At lmax = 294 nm,
Atyr = etyr Ctyr l + etry Ctry l
At lmax = 280 nm,
A tryr ′ C tryr l
ttry
ry ′ C tyr l
tyr
εty r εt′yr and εtrry εt′ry are determined from the absorbances measurements on virgin acids in 0.1N KOH at the two
wavelengths, i.e. at lmax = 294 and 280 nm.
A tyr
λ max = 294 nm: ε ttyr =
C tyr l
A try
ε tryr = and at
C tryr l
lmax = 280 nm.
A tyr
ε ′tyr =
C tyr l
A tryr
ε ′ tryr =
C tryr l
Thus, simply by measuring the absorbance of a protein solution in 0.1 N NaOH at lmax = 280 and 294 nm and
making use of the standard equations, we can determine the amounts of the two amino acids in a given protein.
UV absorption spectra of human and bovine erythrocuprein in the wavelength region of 240 to 310 nm have
been compared. Absorption spectrum of human erythrocuprein shows bands at 254, 260, 266, 326 nm and shoul-
ders at 270, 282 and 289 nm. The low intensity of the band in 280 nm region (e265 = 17,000) is attributed at the low
content aromatic amino-acid residues, especially tyrosine and tryptophan. The absorption spectra in the visible
region show a broad absorption band at 675 nm (e675 = 275)) and shoulder at 440 nm. Thus erythrocuprein is clas-
sified as nonblue copper protein. The ratio of absorbances A 259 = 31.4 (5.111) in case of bovine erythrocuprein
A 680 A
since e259 = 9840 and e680 = 313. This ratio in case of human erythrocuprein has been found to be 265 = 61.8.
A 675
This suggests the absence of tryptophan in bovine erythrocuprein provided it is clear whether or not the human
protein really contains two tryptophan residues.
Problem 5.49: Total sugars, i.e. pentoses and hexoses, when treated with concentrated H2SO4 and phenol (5%)
yield a green–blue coloured condensate of furfural or hydroxymethylfurfural from pentose/hexose and phenol.
The blue-coloured condensation product has a maximum absorption at 490 nm. To plot a calibration curve for
the determination of total sugars colourimetrically, 10 mg of glucose was dissolved in 100 mL of H2O. Five
standard solutions of varying concentrations (10−90 mg) were prepared. The appropriate amounts of reagents
were added to these standard samples, the final volume was made up to 7 mL with water indicated below. After
these, the absorbances of all the solutions were measured with respect to the blank, i.e. 1 mL H2O + 1 mL phenol
(5 per cent) + 5 mL H2SO4.
Volume of standard glucose solution(10 mg/100 mL) 0.1 0.3 0.5 0.7 0.9
Concentration of glucose (in mg) 10.0 30.0 50.0 70.0 90.0
Water (in mL) 0.9 0.7 0.5 0.3 0.1
Volume of phenol (5 per cent) added 1.0 1.0 1.0 1.0 1.0
Volume of concentrated H2SO4 added (in mL) 5.0 5.0 5.0 5.0 5.0
Absorbance at lmax = 490 nm 0.08 0.24 0.40 0.56 0.72
(a) Free sugars were extracted from 0.5 g of finely ground grains (or leaves) with 80 per cent ethanol. The
extracted sugars were dissolved in 50 mL of H2O. The absorbance of 0.2 mL of the extracted sugars in water
measured under the same conditions has been found to be 0.48. What will be the percentage of total sugars in the
given sample of grains?
(b) 0.5 g of sample starch obtained from 100 g of gram seeds was hydrolysed with 52 per cent perchloric acid to
obtain 50 mL extract containing glucose. The absorbance of 0.2 mL of extract measured under the same condi-
tions was 0.54. Calculate the amount of starch in 100 g of seeds.
Solution (a) Form the standard plot between absorbance and concentration of glucose, the amount of sugars
in the unknown sample corresponding to the given absorbance, i.e. 0.48 is 60 mg, i.e. 60 mg of sugars is present
in 0.2 mL of the extract.
The amount of total sugars in 50 mL of extract 50 × 60 = 15000 mg = 0.015 g
02
0.015 × 100
Thus, the amount of total sugars in 100 g of grains = = 3 per cent
05
(b) From the standard graph, the amount of glucose corresponding to the given value of absorbance, i.e. 0.54,
comes out to be 67.5 mg.
So the amount of glucose in 50 mL of extract or 0.5 g of starch sample
50 × 67.5
= = 168.75 × 102 mg = 0.0168 g
02
100 × 0.0168
Therefore, the amount glucose in 100 g of starch sample = = 3.36 g
05
Thus, the amount of starch equivalent to 3.36 g of glucose = 3.36 × 0.9 = 3.02, i.e. 100 g of seeds, contain
3.02 g of starch.
Problem 5.50: Deoxyribonucleic Acid (DNA) is present in all living cells and is the substance of heredity.
DNA when treated with diphenyl amine reagent, the 2-deoxy-D-ribose of DNA reacts to yield a blue-coloured
product (CH3COCHOHCH2CH N(C6H5)2 OH i.e. C17 H19 O3N) which shows a sharp absorption maximum at
595 nm. To plot a calibration curve for the determination of DNA colourimetrically, 15 mg of DNA was dissolved
in 100 mL of ethanol. Four standard solutions of varying concentrations (45−90 mg) were prepared. The appro-
priate amounts of reagents were added to these standard samples, the final volume was made up to 5 mL with
ethanol, indicated below. All the solutions were kept in boiling water for more than 10 minutes. After these, the
absorbances of all the solutions were measured with respect to the blank, i.e. 1 mL ethanol + 4 mL reagent.
Volume of standard solution of DNA (15 mg/100 mL) 0.3 0.4 0.5 0.6
Concentration of standard solution of DNA (in mg) 45 60 75 90
Volume of alcohol added (in mL) 0.7 0.6 0.5 0.4
Volume of diphenylamine reagent added (in mL) 4.0 4.0 4.0 4.0
Absorbance at lmax = 595 nm 0.03 0.04 0.05 0.06
From 3 g of plant material, 50 mL of DNA extract was obtained. The absorbance of 0.5 mL of DNA extract mea-
sured under the same conditions was 0.02. What is the percentage of DNA in the plant material?
Solution Form the calibration graph between absorbance and concentration of standard DNA (in mg), the
amount of DNA corresponding to absorbance = 0.02, has been found to be 30 mg, i.e. 30 mg of DNA is present in
0.5 mL of DNA extract. The amount of DNA present in 50 mL of DNA extract = 30 × 50 = 3000 mg = 3 × 10−3 g.
05
−3
Thus the amount of DNA present in 100 g sample of plant = 100 × 3 10 = 0.1 per cent.
3
Note that in DNA, only the deoxyribose of the purine nucleotides reacts, so the value obtained represents half
of the total deoxyribose present. Moreover, the reaction is given by 2-deoxypentoses in general and is not specific
for DNA.
Problem 5.51: Free fatty acids present in oils are responsible for keeping the quality of oils. Lipids or oils
react with cupric acetate pyridine reagent to yield a greenish blue-coloured complex which shows an absorp-
tion maximum at 715 nm. To plot a calibration curve for the determination of free fatty acids colourimetrically,
20 mg of oleic acid was dissolved in 100 mL of benzene. Four standard solutions of varying concentrations were
prepared. The appropriate amounts of reagents were added to these standard samples, and the final volume was
made up to 7.2 mL with benzene, indicated below. All the solutions were warmed for two minutes and allowed
to stand. After these, the absorbance of supernatants of all the solutions were measured with respect to the blank,
i.e. 5.2 mL benzene + 2 mL cupric acetate pyridine reagent.
Volume of standard oleic acid solution (20 mg/100 mL) 0.2 0.4 0.6 0.8
Concentration of standard oleic acid (in mg) 40.0 80.0 120.0 160.0
Volume of benzene added (in mL) 5.0 4.8 4.6 4.4
Volume of cupric acetate pyridine reagent (in mL) 2.0 2.0 2.0 2.0
Absorbance at lmax = 715 nm 0.21 0.42 0.63 0.84
400 mg of ground seeds were treated with petroleum ether to extract oil. The amount of oil extracted was 50 mL.
The absorbance of 0.2 mL of extracted oil measured under the same conditions was 0.04. What will be the
percentage of free fatty acids in the given sample of seeds?
Solution Form the plot of absorbance against concentration of standard oleic acid (in mg), the value of concen-
tration of free fatty acids in 0.2 mL of extract corresponding to the given value of absorbance, i.e. 0.04, comes out
to be 7 mg. So the amount of free fatty acids in 50 mL of extracted oil = 50 × 7 = 1750 mg
02
= 1750 × 10−6 × 103 mg = 1.75 mg
Thus, the percentage of free fatty acids in the given sample of seeds
100 × 1 75
= 0 43
400
Problem 5.52: Creatinine (1-Methylglycocyamidine, C4H7N3O) plays an important role in storage and trans-
mission of phosphate bond energy. In the presence of picric acid in alkaline medium, it gives a red or light-red
coloured tautomer of creatinine-picrate which exhibits an absorption maximum at 530 nm. To plot a calibration
curve for the determination of creatinine colourimetrically, 50 mg of creatinine was dissolved in 100 mL of water.
Five standard solutions of different concentrations (100−400 mg/mL) were prepared. There samples were mixed
with reagents in appropriate amounts, and the final volume was made up to 10 mL with water, indicated below. The
solutions were allowed to stand for 10 minutes. After these, the following readings were obtained on an absorption
spectrophotometer with respect to blank, i.e. 7 mL H2O + 1 mL 1 N.NaOH + 2 mL saturated picric acid.
Volume of standard creatinine solution (50 mg/100 mL) 0.2 0.4 0.5 0.6 0.8
Concentration of standard creatinine (in mg) 100 200 250 300 400
Volume of water added (in mL) 6.8 6.6 6.5 6.4 6.2
1 N NaOH added (in mL) 1 1 1 1 1
Saturated solution of picric acid added (in mL) 2 2 2 2 2
The absorbance of 0.6 mL of urine sample measured under the same conditions was 0.20. Determine (a) the
amount of creatinine in mg per 100 mL of urine, and (b) the creatinine coefficient, assuming body weight to be
50 kg and urine collected in a single day is 1500 mL.
Solution
(a) Plot a graph between absorbance and concentration of creatinine (in mg). From the graph, our value of
urine sample corresponding to the absorbance, i.e. 0.2, is 70 mg.
Thus, amount of creatinine in 100 mL of urine sample = 100 × 70 = 1.166 × 104 mg
06
= 1.166 × 104 × 10−6 × 103 mg = 11.66 mg
(b) Creatinine coefficient is defined as the amount of creatinine present in urine sample collected per day
per kg body weight. The amount of creatinine present in
1500 × 11.66
1500 ml urine = = 174.9 mg
100
Thus, the creatinine coefficient = 174.9 mg/day/kg body weight

50
= 3.5 mg/day/kg body weight
(c) Folding and Unfolding of Protein Molecules The folding of polypeptide chains into compact globular
structures is often assumed to be responsible for their biological functions in the cell. The unfolding of globular
proteins such as ribonuclease, myoglobin, chymotrypsinogen,
lysozyme, trypsin, etc., can easily be accomplished from
0.02
their optical melting curves. The optical equilibrium curves
are obtained by plotting a graph of absorbance change Tm
per degree as a function of temperature (Fig. 5.71). The
Absorbance change per Degree
pronounced asymmetry of the resultant differentiated melting

curve is due to the change of specific heat upon folding of the
protein molecule. Such a presentation of melting curves is very
convenient and helpful since details of the equilibrium data, 0.01
like the pronounced asymmetry, are more easily recognisable
than they are from the integrated melting curves.
The unfolding of the globular proteins is based on the fol-
lowing two hypothesis: (i) The transition takes place between
two macroscopic states, the native state A and the unfolded
state B, which means that in a melting region, a mixture of
fully native molecules and unfolded molecules exist with
30 40 50 60
negligible amounts of intermediate states. The Van’t Hoff
Temperature (°C)
analysis, i.e. free energy change in a function of the tem-
Fig. 5.71 Optical equilibrium melting point curve of chy-
perature and of the pressure of unfolding of protein based motrypsinogen at acid pH (=2) and tryptophan absorbance
on intramolecular conformational change between the two at 293 nm.
Table 5.32 Comparision of specific heat (in kJ mol−1 K−1) upon unfolding of globular
proteins as determined by calorimetric and optical melting methods.
Protein ΔCP (calorimetric) ΔCP (Van’t Hoff)
Chymotrypsinogen 16 14(16*)
a-chymotrypsin 12.5 14
Ribonuclease 5.5 8.5 (7.1*)
Myoglobin 11.3 10.5
Lysozyme 6.7 8.8
*Values within brackets are based on the pressure dependence of unfolding.
states, lends support to this hypothesis. (ii) There is a gradual unfolding of every protein molecule in the transition
region with many different intermediate states, in such a way, for example, that at the melting temperature Tm, half
of every molecule is unfolded, the rest being folded. Such behaviour is expected for the helical-coil transition of
very long polypeptides.
The change of specific heat upon unfolding of globular proteins as deduced from calorimetric measurements
and Van’t Hoff analysis of optical melting curves are compared in Table 5.32.
(d) Molecular Interactions and Mechanism of Biological Systems Pyridine nucleotide and flavin nucleotide
coenzymes are the oxidation−reduction coenzymes and play an important role in biological oxidation−reduction
process.
Nicotinamideadenine dinucleotide (NAD+) [I, R = H) nicotinamideadenine dinucleotide phosphate (NADP)
[I, R = OP (OH)2] are typical pyridine nucleotide coenzymes. They play a vital role in biochemical processes
involving NaD+ in oxidation–reduction in which the pyridnium ring of NAD+ [II, R = ribosyl adenine phosphate
residue] is reduced by hydride addition in the 4-position to give NADH [III, R = ribosyl adenine phosphate resi-
due]. NADH has absorption maximum at 340 nm while NAD+ has at 260 nm. The coenzyme NADP+ may be
reduced in a similar way to NADPH.
The two flavincoenzymes, flavinmononucleotide (FMN) [IV] and flavinadenine dinucleotide (FAD) [V], are
derivatives of riboflavin [VI]. Just like pyridine coenzymes they are also involved in redox processes in which
hydrogen is transferred at the isoalloxazine ring. The enzymitically reduced forms of FMN and FAD, i.e. FMNH2
and FADH2, are represented by the structures [VII] in which R stands for the side chain of FMN and FAD
respectively.
Mixtures of reduced flavin mononucleotide [VII] ([FMNH2] with NAD+ (which has a absorption maximum at
260 nm) show a new absorption band at lmax = 21,000 cm−1 (480 nm) and is attributed to an intermolecular charge
transfer transition. The CT type interaction in such systems is substantiated by quantitative spectrophotometric
measurements on solutions containing NAD+ analogues, i.e. substituted methylpyridinium chlorides with FMNH2
(Table 5.33). The values of association constant and optical characteristics of the complexes are consistent with
charge-transfer complexation. The relatively high energy of the absorption band of the FMNH2−NAD+ complex
may be because of steric effects of the large group attached at the ring nitrogen atom in NAD+.
The reduced form of flavoprotein enzyme, lipoyl dehydrogenase, forms a green CT complex with either (i) NAD+
with lmax at 14000 cm−1 (700 nm); or (ii) the hypoxanthine derivative of NAD+ with lmax at 14000 cm−1 (700 nm),
or (iii) oxidised thionicotinamide adenine dinucleotide with lmax at 13500 cm−1 (740 nm). The intensities of the
Table 5.33 Association constants of the complexes FMNH2−NAD+ analogues.
R in N-methyl pyridinium chloride νCT (cm−1)

AD
max K(l mole−1)
3-CONH2 19000 830 4.3

4-CONH2 15300 600 5.0
3-COOCH3 18500 1000 7.7
4-COOCH3 14100 1050 8.8
H 18500 600 1.1
ADPR* 22000 560 11.4
*ADPR−represents the adenine−phosphate−ribosyl moiety of NAD+.

O NH2
C NH2
O O N
N
N + O O
CH2 O P O P O CH2 N N
HO OH O− O−
HO OR
CONH2 H H
+H− CONH2
N+
−H−
N
R
R
II III
CH2 P O O NH2
O OH
CH2 N N
CHOH OH O P O P O CH2
CHOH O N N
CHOH OH OH
CHOH H H
CHOH H H
CHOH
OH OH
CH2
9 1 CH2
8 1
H3C 7 N N 2 O 8 9
H3C 7 N N 2 O
NH
H3C 6 4 3 NH
N 4 3
5 H3C 6 N
10 5 10
O
O
(FMN) (FAD)
IV V
CH2OH CH2 O P OH
CHOH CHOH OH
CHOH
CHOH
CHOH
CHOH
CH2 H
CH2
H3C N N O
H3C N N O
NH
NH
H3C H3C N
N
O
O H
VI (FMNH2)
VII
bands grow as the temperature is reduced and the effect is reversible. CONH2
H
+ +
NAD and NADP when mixed with indole and its derivatives, including trypto- N
phan, show broad featureless absorptions extra to the absorptions of the component
species. The extra band in these systems is attributed to CT type interaction between N+
indole ring and pyridinium ring since absorption spectra of l-(b-indolylethyl)-3-carb- CH2 CH2
amidopyridiniurn chloride [VIII] in methanol solution and of an equimolar mixture of VIII
tryptamine hydrochloride and nicotinamide methochloride have been characterised by
a broad band tailing into 22000 cm−1 (450 nm) range of the visible spectrum, with an
estimated maximum at 30,800 cm−1 (325 nm). This low-energy absorption band in [VIII] is assigned to a charge-
transfer transition across space between indole ring and pyridinium ring. The reluctance to undergo intermo-
lecular association between NAD+ and NADH may be as a result of intramolecular charge-transfer interaction
in NAD+ between the pyridinium ring and the adenine moiety. The broad absorption band shown by solutions of
glyceraldehyde-3-phosphate dehydrogenase containing NAD+ may be the result of a charge transfer interaction
again between the pyridinium ring acting as the acceptor site and an indole moiety as the donor site.
Mixtures of FMNH2 and FMN show a new absorption band at lmax 11000 cm−1 (900 nm) and is attributed to
charge-transfer complex formation.
The use of electronic spectroscopy to study the interaction of drugs with biological systems met with a limited
success. A transient magenta colour develops when procaine hydrochloride is mixed with RNA. The band appears
at 18200 cm−1 (550 nm). The band is not due to CT type interaction since the colouration does not appear imme-
diately.
Proteins and amino acids in vivo are probably surrounded by very strong polarisation forces. Electronic absorp-
tion spectra of proteins or amino-acids-I2 systems in water show a new band at 226 nm. When amino acid or
protein is added to aqueous iodine, the intensity of the band at 455 nm due to solvated iodine decreases (iodine
absorption in an inert solvent appears at 520 nm) while those of at 353 and 287 nm due to triiodide ion, I 3−
increases. A new band at 226 nm due to negative iodine ion, I− appears. With passage of time the intensity of band
at 226 nm increases while those of other, i.e. I2 and I 3− species decreases, the decrease may be due to slow iodi-
nation. The iodine presumably attaches itself to the amino group as the I− ion. On the basis of CT complexation
between beta-carotene (C40H56) and I2 and amines and I2 is according to the scheme.
⎡⎣ C40amine ⎤⎦ +2I 2 ⎡ C40 H56 I + ⎤ +I 3− ,

+
H56
⎣ or amineI
i I ⎦
⎡aminoacid ⎤
it is suggested that CT complexes of the type ⎢ or I +⎥⎥ are initially present in the solutions of amino acids or
⎢
proteins and iodine in water. ⎢⎣ protiens ⎥⎦
The most interesting and fascinating biological process where electronic spectroscopy provides an aid to
understand its mechanism is photosynthesis. We have already discussed that chlorophyll in plants absorbs light,
assimilates carbon dioxide from the atmosphere, convert it into water, oxygen and carbohydrates. The question
arises how this absorbed light can be trapped and used in such processes without being dissipated as heat. The
answer to this problem comes from the ‘hopping model of electrical transport’ which is applicable to lamilar
phthalocyanine−electron acceptor systems. The conductivity of a pure donor or acceptor changes when a thin
layer of a second component is deposited on the surface, e.g. the conductivity of film of phthalocyanine (I) is potenti-
ated by 107 when treated in this way with o-chloranil and vice versa. Similar large increases in conductivity have been
observed when a matrix of violanthrene is coated with o-chloranil. According to the ‘hopping model’, an electron
is transferred from a donor molecule to the neighbouring acceptor molecule to yield a dative structure D+ A−. The
charges may separate subsequently via orbital overlap of D+ with another D molecule. Because of such succes-
sive interactions, the positive hole will become free of the negative charge and will cause hole conduction. The
stronger the D−D+ interaction, the lower will be the activation energy for the charge separation and hence greater
will be conductivity.
It is suggested that a similar process is taking place at the biological site of a photo-
chemical energy transfer, called chloroplast, which also has lamilar structure. Chlorophyll N
(II) with its conjugated dihydroporphyrin ring structure acts a donor and forms a charge- N N
transfer complex with plastoquinone, a quinone specific to chloroplasts. Absorption of light N H
H N
by chlorophyll causes the electron to move to the plastoquinone. This is followed by hole N N
migration via neighbouring chlorophyll molecules so that back reaction is prevented. The N
migrating hole eventually reaches a cytochrome system at which electron-transfer from the
I
iron of the cytochrome or some other electron donor (cytochrome oxidase CH2CH3 CH3
enzymes contain iron and copper), restores the chlorophyll to its initial
state. CH3 CH2CH3
N N
The charge-transfer process in photosynthesis is substantiated by the
H
CT complexation between chlorophyll and other porphyrin-like molecules. H
Some of these porphyrins are involved in the related electron-transport N N
chain in respiration. When an electron acceptor is added to a solution of CH3CH2 CH3
a porphyrin, a broadening of the band structure with decrease in inten-

sity of the porphyrin takes place. The appearance of isosbestic points is CH3 CH2CH3
an evidence for the formation of 1 : 1 complex, e.g. aetioporphyrin I (III) III
-1, 3, 5-trinitrobenzene, and tetraphenylporphyrin-1, 3, 5-trinitrobenzene
systems in nitrobenzene solvent show a charge-transfer band in the region 20,000 cm−1. The association constants
for complexes of 2, 4, 7-trinitrofluorenone electron accepters with some porphyrin type molecules are given in
Table 5.34.
Table 5.34 Association constants of some complexes of porphyrin-type molecules with 2, 4, 7-trinitrofluorenone
electron acceptors at 23°C.
Acceptor Donor Solvent λ≠ (in nm) K (lit mole−1)
2, 4, 7-trinitrofluorenone Meso p CHCl3 497 2006

2, 4, 7-trinitrofluorenone Aetio p CHCl3 499 2356
2, 4, 7-trinitrofluorenone Co (II) meso p CHCl3 522 3038
2, 4, 7-trinitrofluorenone Ag (II) meso p CHCl3 561 2261
Meso p ≡ mesoporphyrindimethylester (IV), aetio p ≡ aetioporphyrin I; ≠ wavelength at which optical measurements have been made.
The comparable values of association constants of metal-free porphyrins and CH2CH3 CH3
metal porphyrins for the same acceptor show that electrons from the porphyrin
CH2CH3
ring but not from the metal are involved in the complexation. CH3
N N
It has been known for more than two decades that plants protect themselves H
from excess light, which can lead to oxidative damage to chlorophyll and other H
N N
key photosynthetic pigments. But the biophysical mechanism of the protective CH CH3
3
process (known as feedback de-excitation) remained mysterious. Very recently,
using femtosecond, i.e. ultrafast spectroscopic techniques and plant genetics, it CH2 CH2
was found that when spinach leaves are exposed to intense light triggers, zeaxan-
thin cation radicals are produced. These cation radicals result when zeaxanthin CH2 CH2
or zeaxanthol [C40 H56 O2; λ max (log e): 275(4.34), 453(5.12), 480(5.07)], a caro-
alcohol
COOCH3 COOCH3
tenoid known to be produced by plants in response to bright sunlight, binds to
IV
photo-excited chlorophyll molecules. The zeaxanthin gives up an electron to the
excited chlorophyll, yielding a chlorophyll anion radical and a zeaxanthin cation
radical. These products subsequently undergo charge recombination, allowing the excited state energy of chloro-
phyll to be safely dissipated as heat. Thus, plants rely on a carotenoid to protect themselves from over-exposure to
light. This feedback de-excitation mechanism might be manipulated for practical uses such as protecting artificial
photosynthetic systems used for solar energy generations from light-induced damage.
(e) Study of Dye/Drug-Protein Interactions Coloured organic dyes are consumed by humans through food
and medicine as they are present in the form of additives, preservatives, etc. When these coloured dyes, like other
nutrients, enter the circulation in blood, they are bound specifically by a plasma protein in blood called serum
albumin (having 580 amino-acid residues)—the transporting protein for various drugs, vitamins, hormones as
well as for dyes in the blood. However, the dyes have no physiological role and have to be eliminated rapidly
from the human system to avoid intoxication. When anionic substrates (organic small molecules) interact with
serum albumin, the absorption spectra often show marked changes. Thus, the distribution and elimination of
water-soluble anionic dyes such as eosine, metanil yellow, erythrosine, etc., can be judged from the strength of
binding of dyes with plasma protein in the visible region of spectrum where albumin does not show absorbance.
This enables to follow the absorbance of dye without interference from albumin. Bovine Serum Albumin (BSA)
having molecular mass of 69000 DA being cheaper is an excellent model protein for studies. To a fixed [dye]
varying amounts of BSA (in phosphate buffer of pH = 7.4 and at 298 K) are added and absorbances are measured
at the absorption maxima of the dye. The change in absorption indicates the dye-protein interaction. The binding
of the dye to BSA can be quantified by the absorbance change method, i.e.
C B A d A pb
= (5.111)
CT A d A cbd
where CB is the bound and CT is the total [dye], Ad is the absorbance of the dye alone; Apb is the absorbance of the
dye in the partially bound condition, Acbd is the absorbance in the completely bound condition (at and above this
[BSA], there is no change in absorbance of dye).
Further, the reversible binding of dyes to BSA can be represented by the equilibrium indicated below:
BSA + dye Complex (5.112)
For 1:1 complex, i.e. one amino acid residue binding with one dye molecule,
CB = [Complex] (5.113)
Thus, the binding constant for dye−BSA complex formation is
CB
K = (5.114)
[ ]free C f
where Cf is the free [dye] and can be found as Cf = CT − CB. If [BSA] >>[Dye], [BSA]T ~ [BSA]free. Under these
conditions, Eq. (5.114) modifies to
CB
K= (5.115)
[ ]T C f
From a set of K values, determined with the aid of Eq. (5.115), mean value of K can be determined.
According to theoretical model of Martin (Fig. 5.72) when dye bound has K of the order of 104, then the frac-
tion unbound is to the extent of 70 to 80 per cent and only least affected by the total dosage of the dye. If K is of
the order of 107, at low dosage the dye is mostly bound to albumin and unbound fraction is low. However, at high
dosage, the bound fraction decreases, leading to increase in unbound concentration. Thus the binding constants
‘K’ of the order of 103 to 104 can easily be eliminated from the system and are less toxic to the human system.
In this respect, the binding constants of dye-albumin complex aids in discerning the careful selection of dye,
and thus dyes having strong affinity towards albumin should be avoided in using them as additives, preserva-
tives, flavouring or colouring agents. Many drug molecules have strong ultraviolet or visible absorption spectra.
When such drugs interact with proteins, nucleic acids and other biological molecules, their absorption spectra
show marked changes but many drugs, and most biological meromolecules absorb strongly at wavelengths below
100
Percent of Unbound Dye in Body
4
K = 10
80
60 5
10
=
K
40
07
=1
K
20
2 4 6 8
Amount of Dye in Body (mM)
Fig. 5.72 Theoretical curves due to Martin showing the relationship between percentage of unbound dye
in body and amount of dye in body at differentvalues of K.
300 nm. Spectral changes in this wavelength region are often Phenylbutazone + HSA
small and rather complex and are best observed by means (calculated) Phenylbutazone + HSA
of difference spectroscopy. When an interaction between a (found)
drug and a macromolecule alters the spectrum of one or both 0.6
components then a difference spectrum will be generated.
The observed spectrum of the phenylbutazone−human serum Phenylbutazone
albumin (HSA) complex, for example, is not that which
0.4
would be expected from an addition of the absorption spectra
Absorbance
of the isolated drug and macromolecule (Fig. 5.73). These HSA
spectral changes give rise to a biphasic difference spectrum
which characterises this particular interaction. The difference
0.2
spectrum generated by dissolving phenylbutazone in the cat-
ionic detergent cetrimide (Fig. 5.74) is very similar to that
observed when the drug binds to HSA. This suggests that the
binding site for phenylbutazone is in a hydrophobic region of
0
the protein. +0.08 Difference spectrum
5.21 PHOTOACOUSTIC UV/ V 0
SPECTROSCOPY (PAUVS/PAVS) −0.08

260 280 300
Wavelength, (nm)
We have already introduced in Chapter 3 on infrared spec- Fig. 5.73 The observed ultraviolet absorption spectrum
troscopy, the historical background and the basic principles of phenylbutazone, human serum albumin (HSA) and
of photoacoustic spectroscopy. It was Bell who in 1881 phenylbutazone−HSA mixture. The phenylbutazone + HSA
showed that PA effect is applicable to all matter and even (calculated) curve is obtained by adding the spectra of the
individual components. The difference spectrum is obtained
visible light could produce PA effect. According to him, ‘the by combining the drug and protein in the sample beam and
nature of the rays that produce PA effect on different sub- separating them in the reference beam. All the measurements
stances depends upon the nature of the substances that are have been made in sodium phosphate buffer of pH of 7.4.
exposed to the beam and that the sound are in every case due
to the rays of the spectrum that are absorbed of the body.
5.21.1 Modern Photoacoustic Spectrophotometer

The self-explanatory block diagrams of single-and double-beam photoacoustic spectrophotometers have been
shown in Fig. 5.75(a) and (b). The modem spectrophotometers resemble the spectrophone employed by Bell.
Phenylbutozone
in 0.05 percent
Cetrimide
+0.4
Phenylbutozone
Absorbance
in water
+0.2
Difference
Spectrum
−0.1
240 260 280 300
Wavelength (nm)
Fig. 5.74 The ultraviolet absorption spectrum of phenylbutazone measured in the presence and absence of cetrimide. Blank samples con-
tain the same solvent but without the drug. The difference spectrum is obtained by placing the cetrimide solution of drug in the sample beam
and an aqueous solution of drug in the reference beam. All the measurements are carried out in sodium phosphate buffer of pH 7.4.
Most of the improvements in Bell’s spectrophone are only due to the improvement in technology. The reference cell
in a double-beam spectrophotometer contains some strongly absorbent material like charcoal, sooty carbon black,
etc., and the signal obtained reproduces the power spectrum of the lamp used. Usually, a high pressure 400 W xenon
lamp is used. It gives all wavelengths between 210−100 nm. The ratio of the sample signal to the reference signal
gives the spectrum of the sample after some standardisation procedure. Since PA spectrum is a power spectrum so it
depends not only on the photon flux but also on the energy of the photon. Thus, the signal at 200 nm would be twice
as strong as that in the 400 nm region. The dependence of spectrum on the energy of the photon helps us in better
spectra in the UV region, where the photon flux for most lamps is low. Factors affecting the signal intensity and
precautions to be observed to minimise these factors have already been discussed in Chapter 3. The cell is the most
important element of a photoacoustic spectrophotometer. Its self-explanatory block diagram is shown in Fig. 5.75 (c).
The two removable plates of sapphire or suprasil for UV-visible, which act as windows, attenuate any external noise
since their material is a good reflector of noise. Plates of NaCl or ZnSe are used for mid-IR. The total volume of the
cell is of the order of 1 mL. The removable sample cup is loaded into the sample block from the top, with the window
sealed over it. Typically, the cell is loaded with a sample layer 1.00 nm thick sample layer. Since the sample mount-
ing is easy, it can be held in a cup by gravity so that even liquids can be studied. The gas atmosphere is controlled by
attaching a gas line not shown in the figure to the cell. Note that the most PA effect is observed when the absorption is
in a gas and that heated gas directly activates the microphone. Solids and liquids are studied indirectly, their surfaces
are heated by absorption and this then heats a transparent gas in contact with it and activates the microphone.
Chopper Photoacousitc
Lens Lens cell
Lamp
Monochromator Preamplifier
Microphone
Reference
Voltage- Lock-in
Frequency Amplifier
Converter
Oscilloscope
Chart
Computer
(a)
Oscilloscope
Silica Windows
Chart
Recorder
Computer
Sample
Scan
Chopper
Microphone
Lamp Ratiometer
Mono- Sample
chromator Holder
Amplifier PSD (Silica)
Microphone Cells
Amplifier PSD Aluminium Block

(b) (c)
Fig. 5.75 Block diagram of (a) single-beam photoacoustic spectrophotometer, (b) double beam photoacoustic spectrophotometer, and
(c) photoacoustic cell.
5.21.2 Applications of UV/V Photoacoustic Spectroscopy

The major advantage of PAS is that it enables us to obtain spectra similar to optical absorption spectra on any
type of solid or semisolid material in the crystalline, powder, amorphous, smears, or gelatin, etc., forms. More-
over, since only light is converted to sound, light scattering which interferes in the study of solid materials by
conventional spectroscopic techniques presents no problem

in PAS. A major advantage of PAS over conventional reflec- 10
tion spectroscopy, in studies of semiconductors and metals,
is that highly reflecting, i.e. ultraclean surfaces, are not pre-
8
requisite for studies. Thus, the new classes of organic and
Optical Absorption Relative Intensity

glassy semiconductor and metallic systems, which are often 6
in powder and amporphous forms, can be studied. Splendid
effects from crystals of bichromate of potash, crystals of sul-
phate of copper, tobacco smoke and from a whole cigar have
10 (a)
been observed. Plain water does not show the effect but when
water is discoloured with ink, a feeble sound is produced.
Biological substances in vivo and vitro such as blood smears, 8
membranes, gels, etc., and even whole leaves, flowers and
(b)
grass can be studied by this technique. Some of the typical
applications of UV-visible PAS are discussed here.
12
(a) Study of Organic and Inorganic Compounds
Absorption spectra of organic and inorganic substances in (c)
4
the liquid, solid and gaseous phase can be studied by PAS.
Absorption spectra of powder and amorphous materials are 100 400 700 1000
much better resolved by PAS than by diffuse reflectance spec- Wavelength (nm)
troscopy. Spectra of Cr2O3 obtained by various methods are Fig. 5.76 Spectra of Cr2O3 (a) PAS of powder, (b) optical
shown in Fig. 5.76. It is seen that the two crystal field bands absorbance for single crystal, and (c) reflectance spectra of
of the Cr3+ ion at 460 and 600 nm are clearly resolved in PAS powder.
spectrum than in the diffuse reflectance spectrum and com-
pares favourably with absorption spectrum of the Cr2O3 crystal.
The conventional absorption spectrum of free iodine shows an absorption maximum at 520 nm. The PAS of
iodine vapours matches with this description very nicely.
In ultramarine blue (synthetic lapislazuli), the blue ion S2− is entrapped in the cavities of the colourless sodium
aluminium silicate matrix. The reflection spectra of this compound in the 400−800 nm region exhibits a broad
maximum at 590−600 nm and a minimum at 450 nm. The PAS spectrum of this compound matches this descrip-
tion very nicely. 1.0
9.49 μg
(b) Charge-Transfer Studies PA spectroscopy can also be
used to study charge-transfer complexes. The PA spectrum of
iodine vapours shows an absorption band at 520 nm and of ethyl
sulphide vapours at ~235 nm. When ethyl sulphide vapours are 0.6
added to a cell containing iodine vapours, the iodine absorption

band disappears altogether and a charge-transfer band appears 0.4
1.11 μg
at 314 nm along with the absorption band due to ethyl sulphide
Relative PAS signal
at ~235 nm.
S=1
(c) Surface Studies The tool of PAS can be used in the study 0.4
of adsorbed and chemisorbed molecular species and compounds
on the surfaces of metals, semiconductors and insulators. This 0.095 μg
tool is sensitive enough to detect and identify a monolayer of 0.6 S=3
adsorbed or chemisorbed compound. Fig. 5.77 shows the sen-
sitivity of PAS in studying the spectra of benzylidene acetone
directly on TLC plates. We note that even when a spot is devel-
oped from a starting drop containing <0.1 mg of benzylidene
0.2
acetone, the main broad absorption band of the benzylidene S=6
acetone at 250−300 nm is visible. From the area of TLC spot
(~0.3 cm2), the amount of material present in the TLC spot 200 300 400
(~0.095 mg) and the molecular weight of the compound, it can Wavelength (nm)
easily be shown that there is only one monolayer of the com- Fig. 5.77 Photoacoustic spectra of benzylidene acetone
pound in the spot. The monolayer has been also detected on spots direct form TLC plates.
both the metallic and nonmetallic surfaces. Such studies provide information regarding the structure, valence, com-
plexing, etc., of the deposited compound, e.g. the PA difference spectrum of the chemisorbed monolayer of benzo-
triazole on Cu surface in the 250–450 nm region obtained by subtracting the PA spectrum of untreated Cu surface
from the spectrum of Cu surface treated with benzotriazole, a known passivating agent is different from the PA spec-
trum of benzotriazole powder. This difference has been attributed to some structural changes in benzotriazole due
to chemisorption into Cu surface. Such structural changes may help us understand the mechanism of passivation of
metal surfaces by passivating agents.
(d) Biological Studies Light scattering causes a major problem while obtaining the absorption spectra of
biological systems, i.e. plant, animal, or microorganism in vivo and vitro by conventional spectroscopic tech-
niques, but PAS is free from such a problem since in PAS, light is converted to sound. The range and usefulness of
PAS in dealing with biochemical compounds in their solid or smear form and in solution is demonstrated here with
the aid of some typical biological examples. The PAS spectrum of smear of whole blood is compared with those of
oxyhaemoglobin freed from plasma and haemoglobin extracted from red blood cells in Fig. 5.78A. The compari-
son of spectra shows that the presence of other protein and lipid material in the whole blood, which poses problems
in conventional spectroscopy from light scattering, does not create any difficulty in PAS. The PAS spectra of blood
smears and some haemoproteins in the solid form and in solution have also been shown in Fig. 5.78 B.
The optical characteristics of chloroplasts in the leaf matter are Soret band at 420 nm, the carotenoid band
structure between 450 and 550 nm and chlorophyll band between 600 and 700 nm. Thus, PAS can be used to study
intact plant matter in vivo and vitro to gather information regarding the normal and abnormal plant processes
and pathology. The spectra of spinach leaf by various methods or treatments are shown in [Fig. 5.79]. The PAS
spectrum of intact spinach leaf [Fig. 5.79 (b)] does not show the characteristic chlorophyll band between 600
and 700 nm. However, the spectrum of a crushed leaf exhibits a band between 600 and 700 nm characteristics
(B)
(A)
(a)
Relative Accoustic Intensity
(a)
Relative Intensity
(b)
(b)
(c)
(c)
(d)
200 400 600 800 200 500 800
Wavelength (nm) Wavelength (nm)

Fig. 5.78 (A) Photoacoustic spectra of (a) whole blood, (b) red blood cells freed from plasma, (c) haemoglobin extracted from red bloood
cells, and B: (a) whole blood smear, (b) cytochrome (20m), (c) peroxidase (20m), and (d) catalase (20 m of 2% suspension in ammonium sul-
phate solution).
of Chlorophyll [Fig. 5.79 (c)]. The difference spectrum [Fig. 5.79 (d)]
obtained from the difference of the spectra (c) and (b) matches with
the description of the absorption spectrum shown in Fig. [5.79 (a)].
The difference between the crushed and the intact leaf is attributed to (a)
Absorbance
the difference in the thermal diffusivity of the surface wax layer and
chlorophyll part below. Mathematically, thermal diffusivity is related to
the phase lag and in the presence of the wax layer the acoustic signal
becomes out of phase with the chopping frequency to which the detector
is locked in. A 90° phase-shifted signal from the intact spinach leaf
matches closely with the spectrum of the crushed leaf. Thus, this method
can be used to study the spectra from two different layers. Unlike apex
extract, the base extract of black-eyed susan flower petals is rich in ultra-
Relative PAS signal

violet absorbing flavanol glucosides, which serve as nectar guides to
(c)
pollinating insects with ultraviolet vision. The absorption spectra of basal
and apical extracted with methanol in the 200–800 nm region exhibit a
carotenoid absorption band structure in the 400–500 nm region (i.e. a (b)
triplet: ≈416, 444 and 472 nm has been observed). The basal extract
also exhibits a strong band at 350 nm and is attributed to the presence
of nectar guide in this part of the petal. The PA spectrum of the whole
petal shows that the carotenoid region becomes more pronounced, espe-
cially the central band, and shifts to some degree to a longer wavelength, (d)
i.e. 416, 463 and 491 nm. This spectrum of the whole petals corresponds
(c) — (b)
to average of the apical and basal extract solution spectra.
The solution spectrum of cyanine extract from red rose petals shows
a strong absorption band at 530 nm and a very weak band at 275 nm. PA
spectrum of whole red rose petals shows two strong absorption bands at
530 and 340 nm. The former is attributed to cyanine, whereas the latter 400 600 800
340 nm is due to some other ultraviolet absorbing compound in the red Wavelength (nm)
rose petals. We can say that the band at 340 nm is functioning as a nectar Fig. 5.79 (a) Solution absorption spectrum
guide in the red rose petals just like the 350 nm band in the black-eyed of acetone extract of fresh spinach leaf,
(b) intact spinach leaf sample, (c) crushed spin-
susan petals.
ach leaf, and (d) difference spectrum (c)–(b).
Substances secreted by female butterflies for attracting males by their
H H H
smell can also be studied by UV/visible PA spectroscopy. Bombykol, an
unsaturated alcohol extracted from the female silkworm has the formula: CH3 CH2 C C C C (CH2)8 CH2OH.
cis trans
Butterflies feel its odour at an enormous distance. H

UV/V photoacoustic spectroscopy has also been used for the study of photochemical processes in solids,
e.g. Cooper blue (2, 2-di-methyl-4-phenyl-6-p-nitrophenyl-l, 3-diazobicyclo [3.1.0] hex-ene), and in medicine in
the study of animal and human tissues, both hard tissues such as teeth and bone and soft tissues such as skin,
muscle, etc, for example the difference PAS of tetrachlorosalicylanilide, i.e. TCSA (a most effective antibacterial
agent but causes photosensitivity and other skin problems, the cause of which is unknown so far) bound within the
epidermis in 250–650 nm region, have been reported. The spectrum shows a strong absorption band at 295 nm and
a shoulder at 350 nm. The difference spectrum is obtained by substracting the spectrum of guinea-pig epidermis
from that of epidermis treated with 2 per cent ethanol solution of TCSA.
A new field of calorimetry known as photoacoustic calorimetry (PAC) has also emerged. Bond dissociation
ethalpies of organic and organo metallic compounds have been determined in different solvents with great suc-
cess, e.g. the average of bond dissociation energy of O–H bond in phenol in different solvents at 298 K and duly
corrected for H-bonding to the solvent is 87.4 kcal mole−1, while in the gas phase this value is 87 kcal mole−1.
5.22 OPTICAL ROTATORY DISPERSION (ORD) AND CIRCULAR

DICHROISM (CD)
k mλ2
The refractive index of a medium is not a constant but depends on the wavelength. Mathematically, n = 1 + ,
λ 2 − λ 02
where km and l0 are constants for the medium. Recall from Chapter 1 on introduction that plane polarised light
is imagined to consist of left and right circularly polarised components (Fig 5.80). When a plane polarised light
is passed through a medium containing asymmetric molecules then the left and right circularly polarised components
are transmitted at different speeds, i.e. they have different refractive indices.
This phenomeaon results in a rotation of the plane of polarised light and E
the compound concerned is said to be optically active. When optical rota-
tion is measured as a function of wavelength then the resultant dispersion
curve is generally plain, i.e. there is no maxima and minima in the curve
and the phenomenon is called optical rotating dispersion. However, in the EL ER
wavelength region where the optically active compound absorbs light, the
optical rotatory dispersion curve becomes anomalous, often changing sign,
i.e. the rotatory power at first increases strongly, then falls off and changes
sign. This anomalous behaviour is known as the Cotton effect and is shown
in Fig. 5.81A.
In the wavelength region, where the optically active chromophore absorbs,
there is also an unequal absorption of the right and left circularly polarised
components of plane polarised light, i.e. aL (eL ) ≠ aR (eR), where aL and aR
have their usual meanings. This effect changes linearly polarised light into Fig. 5.80 Influence of an optically active
elliptically polarised light and is termed circular dichroism. CD is usually absorbance sample on linearly polarised
expressed as a differential dichroic absorption, eL − eR, usually written as Δe. light (incident beam linearly polarised).
When the differential dichroic absorption of a simple Cotton effect is plotted
as a function of wavelength, it will have the form as shown in Fig. 5.81B. The CD maximum and the absorption
maximum will occur at the same wavelength. Since CD results from a difference between eL and eR, it follows that
this phenomenon is not observed outside the wavelength region where the optically active chromophore absorbs
light. Contrary to this, optical rotation does extend into other wavelength regions, which explains why it is possi-
ble to measure optical activity, with the NaD line (D line of sodium), of compounds having no visible absorption.
A B
Optical Rotatory Dispersion Circular Dichroism
+ +
[θ]
or
φ Δε
− −
Absorption Absorption
ε ε
λ λ
Fig. 5.81 Optical rotatory dispersion (A), circular dichroism (B) and ultraviolet absorption (lower) curves associated
with an optically active chrornophore. Dark line-Positive Cotton Effect; Dotted line-Negative Cotton effect.
If one assumes eL > eR then amplitude of the R component will be greater than the L while the reverse will be
true if eL < eR. The resulting elliptically polarised light has been represented in Fig. 5.82.
The angle between the major axis of the ellipse and the plane of the original radiation is the angle of rotation, a.
The ellipticity, i.e. the angle whose tangent is the ratio of the minor axis of the plane, OB, to the major axis, OA, is
known as q and is expressed as
1
(ε − ε ) (5.116)
2
The molecular, or molar, ellipticity, [q], is given by the relationship
θ ×100
[θ ] = (5.117)
l ×C
where, [q] = observed ellipticity (deg), α

θ
l = cell path length (cm) and C = concentration (mole lit−1)
A
and [q] = 3305 or 3300 (eL − eR) (5.118)
= 3305 or 3300 Δe R-Component
L-Component
Here, [q] is in deg cm2 decimole−1.
When comparisons have to be made between CD of optically active
B O
absorption bands with different absorptivities then it is often conve-
nient to use the corresponding ‘dissymmetry factors’ g, calculated
from the relationship
Δε
g= (5.119) Plane of
ε incident beam
where Δe = differential dichroic absorption and e = extinction coef-
ficient (using ordinary light) at the same wavelength. The specific Fig. 5.82 Influence of an optically absorbance
rotation, [a] defined by Drude, below changes with wavelength, and sample on linearly polarised light; transmitted
as already stated, change of specific rotation with wavelength is known light elliptically polarised when nR > nL and eL> eR.
as the ORD.
k
[ ] = ∑ 2 i 2 , i = 1, 2, 3... (5.120)
i λ λi
where l is the wavelength of measurement and ki s are the constants that can be identified with the wavelengths of
maximum absorption of the optically active absorptions.
In the wavelength region far away from an otpically active absorption band, the dispersion is normal and
Eq. (5.120) can be written as
k
[ ]= 2 (5.121)
λ λ 02
where l0 is a constant representing the wavelength of the nearest optically active absorption band. When l >>l0,
Eq. (5.121) further reduces to
k π
[ ]= 2 = ( ) (5.122)
where, λ λ
α ( g)λ ( nm )
nL nR = × 10 −8 (5.123)
π (= )l (dm
d )
The vaiue of a can be measured directly with a polarimeter, but q can only be determined indirectiy through
CD.
α 1800
Specific rotation = [ ] = = (5.124)
C π ( °)
The change of temperature has several effects upon the rotation of a solution or liquid. The length of the tube
increases while, the density of liquid decreases with the growth of temperature. It causes changes in the rotatory
power, of the molecules themselves due to association or dissociation, increased mobility of the atoms and affects
other properties. In general, the effect of temperature may be defined as
[ ]T = [ ]20 + z ( ) (5.125)
where z = temperature coefficient of rotation (substances vary widely in their values of z), T = temperature in
degree Celsius.
[ ]M
Molecular rotation = Φ = (5.126)
100
where a is optical rotation (deg dm−1), C = concentration (g mL−1) and M = molar mass. Note that both molecular
ellipticity and molecular rotation may be either positive or negative depending on the relationships between eR
and eL or nR and eL.
A plot of molecular rotation j, versus wavelength for an organic and a hypothetical compound with single
and double chromophore(s) has been displayed in Figs 5.83 and 5.84. The midpoint between peak and trough
of the ORD curve corresponds to the CD maximum, and the ORD trough coincides in wavelength with the
absorption maximum. The CD curves often aids in locating the hidden absorption maxima, and do so with less
7 CDpositive maximum
C8H17
6
Molecular ellipticity [θ] × 10−3

Molecular rotation ϕ × 10−3
4
HO
S
3 ORD peak
0
UV
maximum
−1 ORD midpoint
2
−2
Log ε
−3
1
−4 ORD
trough
−5
250 300 350 400
λ, nm
Fig. 5.83 Optical rotatory dispersion, circular dichroism, and ultraviolet absorption curves
of 3b –hydrodxycholestan –5a , 6a episulphide.
0.02
Molecular rotation, [ϕ]
Molecular ellipticity, [θ]
0.01
0
200 300 400 500 600 700
−0.01
λ, nm
Fiq. 5.84 Circulardichroism (...) and optical rotatory dispersion (-) for a
hypothetical substance with absorption bands in the region 200–600 nm.
ambiguity than will the corresponding ORD curve. On the other hand, ORD curves are affected by more distant
chromophoric bands and hence are more characteristic of specific compounds than CD. Both ORD and CD can
give essential information about the stereochemical features of optically active materials.
Also we know that when any liquid or solution is placed in an external magnetic field, it rotates the plane of
polarised light. This effect is called Faraday effect and has already been explained in Chapter 1. This effect results
because of the effect of the applied field upon the motion of electrons in the molecules.
The angle of rotation is given by
jF=VHl
where H is the magnetic induction, l is the path length and V is a constant (for the medium) called the Verdict
constant.
The mathematical expression for the Kerr electric-optic effect discussed in Chapter 1 is
n| |− n⊥ = KE2l
The quantity K is called the Kerr constant.
5.22.1 Origin of Absorption and Circular Dichroism

Both the phenomena have their origin in the charge displacements caused by the interaction of a molecule with
the electromagnetic radiation which produces induced electric and magnetic dipoles. The rotational strength Rk of
a chromophore for k electronic transition is expressed as
R k M e M m cos θem
where Me and Mm are the electric and magnetic transition moments and qem is the angle between the directions of
these two moments. If the molecule has either a centre of inversion or a reflection plane of symmetry then we may
come across any one of the following three situations:
(i) Me ≠ 0, Mm = 0; Electric dipole transition permissible but magnetic dipole transition forbidden
(ii) Me = 0, Mm ≠ 0; Magnetic dipole transition permissible but electric dipole transition forbidden.
(iii) Me ⊥ Mm;cosqem = 0
In all cases, Rk = 0 and the molecule is optically inactive.
Problem 5.53: Assume that the initial beam intensity, I0, for R radiation equals IR for IL radiation. Develop the
equation for molar ellipticity using a spectrophotometer.

[θ ] [ε ] (Eq. 5.118)
In order to develop an expression for use in a spectrophotometer, eL and eR must be expressed in terms of AL
and AR respectively. Applying Beer–Lambert law to R and L radiations, we obtain
I
log 0 = ε L C
IL
I0
and log = εR C
IR
Lumping together these expressions, we write
IR
log (ε L − ε R ) C
IL
1 I
or ( )= log R
C IL
Substituting ( ) into Eq. (5.118), we arrive at the desired expression, i.e.
3300 I
[θ ] = log R
lC IL
Problem 5.54: Calculate the circular birefringence for a substance giving a rotation of −12° for light of 500 nm
wavelength in 15 g /150 mL solution. The polarimeter tube length is 2 dm.
Solution Substituting the data from the problem into Eq. (5.123), we get
−12° × 500 × 10 −9 m
nL nR = = −166 × 10 −7
180° × 0 2 m
Problem 5.55: Calculate the molar ellipticity if the solution placed in a 2 dm tube showed spectrophotometer
readings of 20 per cent transmittance for R-circularly polarised light and 21 per cent for L-circularly polarised
light. The incident beams are equal in intentsity. The concentration of the stock solution (in water) is 1 g /10 mL.
Solution Substituting the data in question into the expression deduced in Problem 5.54, we have
3300 20
[θ ] = log 0.70 g −1 ( m 1
)
1(g / ) × 2dm 21
5.22.2 Types of Optically Active Chromophores

They may be classified into two extreme types: (i) the inherently asymmetric chromophore, and (ii) the inherently
symmetric chromophore which is asymmetrically perturbed. The former includes such compounds as hexahelicene
and certain substituted biphenyls and allenes in which the chromophore itself is asymmetric. These structures are
seldom encountered in biological systems. By far, the large majority of chromophores is symmetrical and can only
become optically active when perturbed by an asymmetric centre. When an asymmetric centre induces optically
activity in a chromophore which is a part of the same molecule then the molecule is said to be intrinsically optically
active. Intrinsic optical activity can also occur if the chromophore has an asymmetric arrangement in space, e.g. carbo-
nyl group in the a-helical regions of proteins. Extrinsic optical activity may be observed when a symmetric molecule
binds to a macromolecule, e.g. interaction between the ligand and an asymmetric binding site. But the interaction
may also occur when binding results in an asymmetric spatial arrangement, e.g. the helical arrangement of basic dye
molecules bound to poly-L- glutarnic acid.
The sign and magnitude of an induced Cotton effect (extrinsic or intrinsic) depends on the spatial relationship
between the asymmetric centre and the perturbed chromophore. The space around a chromophore is divided into
regions of positive and negative contribution to a Cotton effect according to well-defined symmetry rules. A given
asymmetric centre may therefore generate either a positive or a negative Cotton effect, depending on its spatial
relationship to the perturbed chromophore. The magnitude of a Cotton effect will increase as the distance between
the asymmetric centre and the perturbed chromophore decreases. Theoretically, the sign of contribution can be
depicted by the so-called ‘octant rule’. The rule states that the sign of the contribution that a given atom at the
point P (x, y, z) makes to anomalous rotatory dispersion will vary as the simple product, x· y· z, of its coordinates.
For example, let us look at the carbonyl chromophore group in the chair form of cyclohexanone molecule in
which the coordinates are defined by the X− Y, X− Z, and Y−Z planes as shown in Fig. 5.85.
Z
a e a
C4 C3R
C3L Ye C3L C4
C3R e
a
Upper left Upper Right
C2L -
+
C1 C2R
O
Lower Left Lower Right
X - +
C1
e e
C2L C2R
a a
(a) (b)
Fig. 5.85 Illustration of the octant rule for cyclohexanone (a) the octants,
and (b) contributions of groups in the far octants.
The carbonyl group lies in the X−Y plane and is bisected by the Y−Z plane. Atoms or substituent atoms can
now be located in one of the eight octants defined by ± x, ± y, and ± z. Atoms or substituents in the + x, + y, + z; +
x, −y, −z; −x, +y, − z; or − x, − y, +z octants have positive contributions to the Cotton effect. Atoms or substituents
in the other four octants, viz., −x, −y − z; −x, +y, + z; +x, −y, + z; or +x, +y, − z have negative contributions to the
Cotton effect. Atoms or substituents which lie on or near the plane have little or no effect. The rule may further be
elaborated as follows. Carbon atoms C1 and C4 lie in one plane. Carbon atoms C1, C2L and C2R lie in another plane
from the observer viewpoint. The substituents in the equatorial positions at C2L and C2R are in this plane. The third
plane bisecting the carbonyl group is perpendicular to the above-said two planes. Thus, the first two planes produce
four octants and the third produces four more octants. The four octants defined by the third plane are usually vacant
and are not considered. Only the four octants defined by the first two planes are considered. The midpoint of C = O
is taken as the origin of coordinate system. This rule is illustrated here by taking the example of 3-methylcyclo-
hexanone molecule for conformation studies. The structure of the compounds must be known in such studies.
When the ketone group is oriented in order to apply the octant rule there are two alternatives: (i) Equatorial
representation with the methyl group on the upper left (at C3L) or (ii) axil representation with the methyl group on
the upper right (at C3R). The fact that a positive Cotton effect is observed confirms that (i) represents the correct
conformation. There is no contribution from other groups since carbon atoms 2, 4, and 6 lie in or near the planes
P1 and P2 and the contribution of the carbon atom 3 is cancelled by that of 5.
a P1
5.23 APPLICATIONS OF ORD AND CD O e
e a
1 H 3C 3 4 5 e
Both these techniques are used for the structure determination 6 2
of optically active substances such as amino acids, polypeptides a
5 e 2 1 6 e P2
and proteins, steriods, antibiotics, terpenes and metal-ligand 3 CH 3
4
complexes. Some general rules are available for some classes of a
a
compounds and are the following: one
The sign of Cotton effect in amino acids reflects the stereochemistry at the asymmetric centre. L−a− amino
acids show a positive Cotton effect around 215 nm whereas their dextroenantiomers show a negative Cotton
effect. In polypeptides, the percent of a-helix structure can be determined from measurements of ORD.
In steroids, different types of ORD curves corresponding to cis and trans forms have been observed. In one
form, the specific rotation increases with decreasing wavelength until it reaches a peak and reverses (a positive
curve) and in the other; the specific rotation decreases with decreasing wavelength until it reaches a trough and
reverses (a negative curve). The sign of the Cotton-effect curve and the wavelength and specific rotation of the
peak or trough also aids in locating the carbonyl groups in steroids.
Further, the large variety of optically active molecules present in biological systems makes ORD and CD two
of the most widely applicable methods available for studying drug interactions. While either techniques may
be employed, CD is probably easier to interpret, since ellipticity bands at any given wavelength are not compli-
cated by contributions from optically active transitions occurring at other wavelengths. Optical rotation can be
employed to monitor the interaction of small molecules with a macromolecule. For example, the examination of
the effect of anesthetics procaine and procaine amide on intrinsic optical activity of adenosinetriphosphate (ATP)
reveals that both these local anesthetics decrease the negative optical rotation of ATP over the range 350–600 nm
suggesting that binding of procaine or procaine amide to ATP causes changes in the orientation of the adenine
ring system of the nucleotide relative to its neighbouring sugar groups. The optically inactive acridine orange dye
becomes optically active on interacting with poly-L-glutamic acid. The vanishing of optical activity of poly-L-
glutamic acid as a result of such interaction is due to the destruction of its helical structure by a helical array of
dye molecules. It has been found that many small symmetric molecules become optically active on binding to
proteins and other macromolecules. In a majority of such cases,
optical activity arises not from a special spatial arrangement of
the ligand but from a perturbation of the ligand chromophore
by an asymmetric locus at the binding site. Since extrinsic 7.5
Cotton effects reflect the three-dimensional characteristics of
specific sites on a macromolecule, the binding of anionic drugs
to serum albumin can be examined by measuring their extrinsic
5.0 (a)
Cotton effects.
The CD spectrum of Human Serum Albumin (HSA) in
[θ] × 10−4
sodium phosphate buffer (pH 7.4) has a small negative elliptic-

ity band at wavelengths below 320 nm (Fig. 5.86). The addition 2.5
of flufenamic acid to HSA gives rise to a large positive ellip-
ticity band at 296 nm and a smaller negative band at 345 nm
(Fig. 5.86). Since ultraviolet absorption spectrum of flufenamic 0
acid has a strong absorption band at 288 nm with a shoulder (b)
at 322 nm, the observed optical activity is extrinsic in origin.
Thus, the acid takes up a preferred spatial relationship to its −2.5
serum albumin binding site, so that an asymmetric locus at or
near that site is in a region of positive contribution to a Cotton
280 300 320 340 360
effect for the electronic transition occurring at 296 nm but in
Wavelength (nm)
a region of negative contribution for the transition of 345 nm.
Fig. 5.86 Extrinsic circular dichroism of (a) flufenamic
The binding of structurally related drugs, mefenamic acid and acid bound to human serum albumin, and (b) humanserum
meclofenamic acid, to HSA also generates biphasic extrinsic alone. All measurements are made in a sodium phosphate
Cotton effects. The spectral data on the interaction of fenamic buffer of pH 7.4.
acids with HSA is listed in Table 5.35.
The dissymmetry factors for the three drugs (Table 5.35) indicate that meclofenamic acid is much less influ-
enced by the asymmetry of the binding site as compared to mefenamic acid or flufenamic acid. However, an
examination of molecular models shows that the bulky chlorine atoms of meclofenamic acid forced the N-phenyl
ring out of coplanarity with the rest of the molecule. If the binding site for the fenamic acids on human serum
albumin is a narrow crevice, meclofenamic acid may have difficulty in entering and hence will be less affected by
the associated asymmetric locus.
The titration of a fixed amount of HSA with flufenamic acid at 295 nm in a sodium phosphate buffer (pH 7.4)
indicates that each mole of protein is bound to 3 moles of drug. (Fig. 5.87).
The molar ellipticities of flufenamic acids bound to different serum albumins are recorded in Table 5.36. Just
like HSA, some of the albumins exhibit biphasic Cotton effects while the others generate a single positive Cotton
effect. This suggests that the serum albumins generating single positive Cotton effect have different drug-binding
Table 5.35 Molar ellipticities of complexes between fenamic acids and human serum albumin.
Drug R3 UV-absorption maximum CD maximum g × 104

COOH
NH
l (nm) [q] × 10-4
R1 R2 (deg cm2
dmole-1)
R1 R2 R3 lmax (nm) e × 10-4 (mole-1 cm-1)
288 1.24 +5.70
296 +2.01
Flufaenamic acid H CF3 H
345 −0.69
322 0.54 −7.04
284 1.40 292 +2.24 +4.70
Mefenamic acid CH3 CH3 H
332 0.64 340 −1.44 −6.54
277 0.88 302 +0.44 +2.40
Meclofenamic
Cl CH3 Cl
acid 315 0.56 332 −0.30 −2.80
sites on these proteins than those on human serum albumin seru- n

1 2 3 4 5
malbumin, and (b) human serum alone. All measurements are made
in a sodium phosphate buffer of pH 7.4. 1.0
The binding of phenylbutazone to HSA produces a positively
ellipticity bandwith, a maximum at 287 nm. The UV spectrum of
the drug exhibits an absorption maximum at 265 nm and is attrib-
uted to p–p * transitions in the phenyl group. The n–p * transitions
[θ] × 10−5
of the carbonyl group are observed around 280 nm while p–p *

0.5
transitions are observed around 185 nm. Thus, it is the n–p * tran-
sitions of the pyrazolidinedione ring carbonyl groups which are
being perturbed by the asymmetric locus at the drug-binding site.
The n–p * transitions give rise to relativity weak absorption bands
which are hidden under the much stronger p–p * bands.
The substitution of Zn by Co in human carbonic anhydrase is
0 2 4 6 8
accompanied by multiple absorption maxima in the region 500–600 Bound Flufenamic Acid (M × 105)
nm. They may be attributed to d–d electronic transitions of the cobalt Fig. 5.87 Relationship between molar ellipticity [q]
ion. The cobaltenzyme is optically inactive in this wavelength (deg cm2 dmole-1) measured at 295 nm and the con-
region. However, the interaction of sulfonamide inhibtor acetazo- centration of flufenamic acid bound to human serum
lamide, produces an asymmetrical anomalous rotatory dispersion albumin. n = number of acid molecules bound per
with a peak at 590 nm, a crossover point at 554 nm and a trough at molecule of HSA; pH = 7.4 (sodium phosphate).
525 nm. These results suggest that the binding of acetazolamide to human carbonic anhydrase is accompanied by
protein conformational changes at the active site. Such studies will help to elucidate the mechanism of sulfonamide
inhibition of carbonic anhydrase.
Extrinsic Cotton effects are often generated when drugs bind to nucleic acids, for example, when proflavine
and acridine complexes with calfthymus DNA optical activity is observed. The dependence of the magnitude of
Table 5.36 Molar ellipticities of the complexes of flufenamic acid with different serum albumins.
Serum albumin Wavelength (nm) q × 10−1 g × 104 Remarks
(deg cm2 mole−1)
Human 296 +2.01 +5.70 ⎫
345 −0.61 −7.04 ⎪ Biphasic
Porcine 292 +1.90 +4.43
⎪⎪ Cotton
⎬
352 −0.31 −3.92 ⎪ Effects
Bovine 305 +0.77 +2.65 ⎪
348 −0.26 −2.63
⎪⎭
Canine 295 +1.08 +2.84 Single positive
Ovine 302 +0.90 +3.22 Cotton effect
the extrinsic Cotton effect on the DNA/proflavine ratio suggests that a nearest-neighbour interaction between
bound dye molecules is necessary for optical activity.
Quinacrine produces an extrinsic Cotton effect at 345 nm when bound to single-stranded DNA or t-RNA. On
the other hand, binding of quinacrine to double helical DNA produces Cotton effect at 333 nm. These differences
are related to the bimodal binding of acridine to DNA. A weak positive Cotton effect also appears at 350 nm when
the antimalarial drug berberine binds to yeast RNA or calf thymus DNA.
When optically active chromophore participates in a reaction, the extent of that reaction can be followed by
observing the reduction of the Cotton effect (if there is one). Thus when hydrochloric acid is added to a solution
of (+) 3-methoxy cyclohexanone in methanol, a Cotton effect is reduced by OCH3
93 per cent as a result of the dimethyl ketal formation. In ethanol, the reduc- C O C
tion is 33 per cent and in isopropanol there is no reduction.
OCH3
5.24 FLUORESCENCE AND PHOSPHORESCENCE

The molecule from the lowest vibrational level (V′ = 0) of a ground electronic state on absorption of light in the
ultraviolet and visible region of the spectrum (200–800 nm) crosses over to one or more of the subvibrational levels
of the first excited singlet state. The time taken for this process is 10–15 second For any given molecule, the strong
absorption bands which lie at the lowest wavelengths usually correspond to electronic transitions from the ground
state to one or more of the vibrational sublevels of the first excited singlet state (Fig. 5.88). In liquids/solutions, vibra-
tion energy is dissipated as heat energy to the surroundings so that the excited molecule rapidly (within 10−13−10−12
second) reaches the lowest vibrational level (V ″ = 0) of the first excited state via ladder mechanism. This dissipa-
tion of energy is due to collision between the solute molecules or between the solute and solvent molecules. The
molecule from V″ = 0 vibrational level may then return directly to the ground state either by radiationless transition
(in which excitation energy is transformed into the thermal energy) called internal conversion (ic); or by emission of
light. Both the transitions occur between electronic states of same spin multiplicity, i.e. S1→ S0. The light emitted by
a V″ = 0 of the S1 state when electron crosses over directly to one or more of the vibrational sublevels of the ground
electronic state is called fluorescence ( f ). The time of fluorescence is approximately 10–8 second while that of inter-
nal conversion is 10−11−10−13 second. In the gas phase, when a molecule from the vibrational sublevels of the excited
singlet state directly crosses over to the ground electronic state, the energy of emitted radiation is equal to the energy
of the initial excitation energy. This process is called resonance fluorescence. Since atoms do not posses vibrational
energy, they show resonance fluorescence. For example mercury vapours in the normal state when exposed to ultra-
violet light of 253.7 nm wavelength, emit fluorescent light of 253.7 nm wavelength, corresponding to the transition
3
P1 → 1S0 + hvf . Further, fluorescence can be classified as either prompt or delayed depending on the time of fluo-
rescence and the mechanism by which the fluorescence results. Prompt radiation is emitted within 10–8 second after
the excitation radiation is switched off and results from the direct radiative S1 → S0 transition. Radiation emission
in this process occurs either from (a) an S1 state molecule, or from (b) an eximer formed by the interaction of an
Vibrational Relaxation from

V′′ = 3 V′′ = 0, Ladder Mechanism
3
10–13 –10–12 sec
2 Intersystem Crossing (10-8 - 10-7 sec)
3
1
V″ = 0 2
S1(10-8 - 10-7 sec)
1
Fluorescence
V′′ = 0
Photoexcitation
Resonance
Internal Conversion
Fluorescence
T1(10–5 sec)
Phosphorescence
10–11 –10–13 sec
10–15 sec
Intersystem crossing
10–8 sec
10–2 –10+2 sec
2
1
V′ = 0
S0
Fig. 5.88 Sematic diagram showing radiative, viz., fluorescence, phosphorescence and nonradiative intramolecular processes. The wavy
arrows stand for radiationless transitions.
S1 state molecule with an S0 state molecule. By eximer we mean an excited-state dimeric species. The two processes
resulting in prompt fluorescence are shown below:
(a) S1 → S0 + hvf (b) S 1 S0
fast
(S1S 0 )* → 2S
2S 0 h f (ex )
hv
The eximer fluorescence emission may or may not be similar to that of the free S1 state molecule. The efficiency
of either of these processes depends on the concentration of solute molecules. As the concentration of solute
molecules is increased, the probability of S0 → S1 transition increases and hence the probability of formation
of (S1 S0)* eximer also increases. Thus, if there is a difference between the fluorescence spectrum of a molecule
at two different concentrations, it is likely that eximers are formed in the system, e.g. dilute solutions of pyrene
(∼10−6 mole lit−1) show violet fluorescence and have a structured spectrum, while concentrated solutions of it
(∼10−3 mole lit−1) exhibit blue fluorescence and have a structureless spectrum.
The emission time of delayed fluorescence is longer (>10−3 second) than that of prompt fluorescence
(10−8 second) and this fluorescence arises as a result of transformation of S1 state molecules formed from the corre-
sponding low-energy T1 state molecules to the S0 state molecules. This type of fluorescence is emitted throughout
the lifetime of the T1 state of the molecule.
The S1 state molecules giving rise to delayed fluorescence are formed either by (a) thermal excitation of a
low-energy T1 state molecules or by (b) collision of two T1 state molecules. The processes resulting in delayed
fluorescence are shown below:
(a) 1
Thermal
S1 → S 0 hv f (b) T1 + T1 → S1 + S0 (energy-pooling step)
S1 → S0 + hvf
The first step in the scheme (b) is called energy-pooling step or triplet–triplet annhilation since two low-energy
states combine their energy to form a third state of higher energy than either of the initial two states, i.e. 2T1.
The fluorescence emission by the scheme (a) is temperature dependent while that by the scheme (b) is tempera-
ture independent. Normally, the form of prompt fluorescence arising from the free S1 state molecule is similar to
that of delayed fluorescence. The intensity of delayed fluorescence rarely exceeds a few per cent of that of prompt
fluorescence.
Further, the molecule in the excited state can emit radiation by another process called phosphorescence ( p).
The molecules in the lowest vibrational level (V″ = 0) of the S1 state could cross over to one or more of the vibra-
tional sublevels of the first triplet state T1 corresponding to the singlet state S1 of the molecule. This radiationless
process occurring between states of different multiplicities is called intersystem crossing (ISC). The time for such
a process is approximately 10−8 to 10−7 second which is the same order of magnitude of the lifetime of the singlet
excited state. Hence, this isc process and fluorescence are mutually competitive. The lifetime of the triplet state
being longer (≈10−5 second), the chances of losing energy along radiationless processes are great. The molecules
in the vibrational sublevels (V″ = 1, 2, 3) of T1 state lose energy to the surroundings as thermal energy and moves
to the lowest vibrational level (V″ = 0) of the T1 state via ladder mechanism.
The molecules in the vibrational level V″ = 0 of the T1 state can now directly return to the ground state S0
with the emission of radiation. This process is called phosphorescence (or sometimes b-phosphorescence) and is
independent of temperature. The time of phosphorescence is 10−2−10+2 second. The other process by which the
molecules can move from T1 (V″ = 0) state to the S0 state is the intersystem crossing. Phosphorescence occurs at
longer wavelengths compared to fluorescence because triplet state has lower energy relative to the corresponding
singlet state. Further, since phosphorescence is exhibited mainly by solids (i.e. in a rigid solvent), the extent of
fluorescence (S1 → S0) decreases and under certain conditions, the molecule moves from T1 state to the S1 state.
The activation energy is small (i.e. 7–9 kcal which is almost identical with the calculated energy difference for
blue and yellow phosphorescent radiations for the fluorescein) and corresponds to radiation in the far infrared
region of the spectrum. The molecule may now directly cross over to the ground state (S0) from the S1 state with
emission of radiation. Such a process is called a-phosphorescence and is temperature dependent. The frequency
of a-phosphorescence will of course be the same as that of fluorescence. In phosphorescence, the emission can
occur provided the energy is taken by the molecule in the T1 state from the surroundings. This is confirmed by the
fact that lowering of temperature as in the case of freezing of liquids, is helpful in producing phosphorescence.
However, excessive cooling slows down the emission and at liquid air temperature, the emission may be com-
pletely cut off. The two processes giving rise to a and b phosphorescence are shown below:
(a) S1 → T1→ S0 + hvp (b) T1 → S1→ S0 + hvp (= hvf)
Phosphorescence spectra of dyes exhibit two distinct bands. As an example, fluorescein in boric acid shows
phosphorescence in the blue (∼570 nm) and yellow (∼480 nm) regions. The former is known as a-phosphorescence
and is identical with fluorescence and becomes less marked as the temperature is lowered and disappears com-
pletely at 0°C. The yellow or b-phosphorescence is independent of temperature and has been observed at 23 K.
5.24.1 Quantum Yield or Quantum Efficiency of Photophysical Processes

Note that all the activated molecules convert rapidly to the V″ = 0 vibrational levels of the S1 and T1 states via radia-
tionless ladder transitions and intersystem crossing-ladder mechanisms respectively. When the molecules in the
excited state do not take part in any chemical reaction, they return back unchanged from V″ = 0 levels of the S1 and
T1 states to the initial ground state of the molecule. The efficiency for the reformation of the ground state via this
mechanism is expressed in terms of a parameter called quantum yield or quantum efficiency, and is defined as
number of ground state molecules reformed
ϕ= (5.127)
number of ground
d state molecules initially excited
The value of j for the reformation of the ground state from the excited
states by the above mechanism will thus be equal to unity. But on the other S1
hand, the molecules in the V″ = 0 level of the activated singlet molecules isc
may cross over to the initial ground state via number of processes such as
radiative (phosphorescence, T1 → S0; and fluorescence, S1 → S0) and non- T1
radiactive (interconversion (ic), S1 S0 and intersystem crossing (isc), f
S1 T1 S0); Fig. 5.89. ic
p
The quantum efficiency of each of these processes is defined as isc
number of molecules deactivated by the process

= (5.128)
number of g
ground state molecules initially activated
S0
rate of process Fig. 5.89 Flowsheet diagram representing
or j process = (5.129)
rate of absorption of radiation the radiative and nonradiative intramolecu-
lar transitions in an excited molecule.
Thus, for fluorescence
ff (S1 → S0) + fic (S1 S0) + fisc (S1 T1) = 1.00 (5.130)
provided there is no photochemical reaction or there is no loss of energy from excited states by intermolecular
quenching.
Similarly, we can write for phosphorescence
fp (S1 → S0) + fic (T1 S0) + fisc (S1 T1) = 1.00 (5.131)
The quantum efficiency increases in proportion to the wavelength (l) of the absorbed radiation. After attaining
a maximum value in a certain interval of lmax, the quantum efficiency drops rapidly to zero on further increase in
wavelength.
Further, if kf, kic, kisc, (S1 T1), kp and kisc (T1 S0 ) represent the unimolecular rate constants for each of
the processes shown in Fig. 5.89 then the quantum yield for each of these processes is related to their respective
rate constants by the expression
k process
φ process = (5.132)
∑ki
where Ski, is the sum of the rate constants for each of the processes. The observed radiative lifetimes of the S1 and T1
states are determined from the rates of the deactivation processes originating from the respective states, and are given
by the expression
1
τ= (5.133)
∑ki
Factors Affecting Rate Constants of Photophysical Processes The energy gap ΔE (S1 − T1) determines
the relative magnitudes of kisc, kf , kp and kic. It is found that as ΔE decreases, the ratio of ϕp/ϕf increases. The rate
of internal coversion and hence rate constant kic (S1 S0)/intersystem crossing and hence kisc (T1 S0)
decreases as the energy gap ΔE (S1 − S0)/ΔE (T1 − S0) decreases.
Fluorescence is very sensitive to temperature, so that rate of fluorescence can be increased relative to other
processes by deactivating the S1 state by lowering the temperature of the system. The fluorescence quantum yield
ff varies with temperature according to the following equation:
1
= α + e β /T
φf (5.134a)
⎡1 ⎤ β
or 1n ⎢ − α ⎥ = (5.134b)
⎣φ f ⎦ T
where, b is a constant, a = (1 + Σ ki′ τ), τ is the observed radiative lifetime of the S1 state and Σ ki′ is the sum of the
rate constants for deactivation of the S1 state by processes other than fluorescence. The rate of isc (S1 T1) can
be enhanced relative to the rate of fluorescence (S1→ S0) and ic (S1 S0) by substituting a heavy atom in the
molecule or by using solvents containing a heavy atom. This results in higher value of jp and the lower values of jf
in the system with a heavy atom present. The heavy atom causes spin-orbit decoupling due to which two electrons
are present in two orbits. The presence of heavy atoms also increases the rate of (T1 S0) intersystem crossing
and reduces phosphorescence lifetime of the T1 state.
Relative rates of various deactivation processes also depend on the nature of S1 and T1 states. The relative rates
depend on whether the excited state is (n – p*) or (p – p *). Some of the criteria fixed for identifying S1 and T1
states of the (n, p*) and (p, p*) type are summarised in Table 5.37.
Table 5.37 Some of the criteria used for identifying the S1 and T1 states of the (n–p*) and (p–p*) type.
Spectra n–π* π–π∗

Absorption spectra
emax values S0 → S1, emax < 10 S0→S1, εmax >100
O2 perturbation S0 → T1, not normally enhanced S0→T1, normally enhanced
ΔE (S1 − T1) range 1500–15,000 cm−1 10,000–20,000 cm−1
Solvent polarity Blue shift of n–p * bands, vibrational fine struc- Red shift of p–p * transitions, vibrational fine structure
ture broadens on increasing solvent polarity. remains unchanged on increasing solvent polarity.
Emission spectra
Fluorescence τf > 10−6 second τf ∼10−9−10−7 second
jf < 0.01 jf ∼ 0.5–0.05
Phosphorescence τp ∼ 10 −10 second
−3 −2
τp ∼ 0.1–100 second
jp ∼ 0.5–0.05 jp ∼ 0.5–0.05
5.24.2 Relationship Between Luminescence (Fluorescence/

Phosphorescence) Intensity and Concentration
If I0 is the intensity of the incident radiation and It that of the transmitted radiation then the intensity of the radia-
tion absorbed by the sample will be (I0−It).
Fluorescence intensity is related to the absorbed radiation intensity by
If φ (I − I t ) (5.135)
According to Beer–Lambert’s law
It I 0e − εCCI (5.136)
Lumping together Eqs (5.135) and (5.136),
If φI 0 (1 − e − εCCI ) (5.137)
For dilute solutions, i.e. when C (absorbance, A) is small, we can write Eq. (5.137) as
If ϕ I [1 − (1 − ε Cl )]
(5.138)
= ϕ I ε Cl
Thus, the fluorescence intensity linearly varies with the concentration provided only a single absorbing and
fluorescencing species is present in a solution.
At higher concentration, i.e. until quenching starts, even if intensity may not be linearly varying with con-
centration, we can plot a graph of intensity against concentration and use it as a calibration curve for unknown
concentrations. At higher concentrations, If drops off resulting in a curve, which resembles negative deviation
from Beer’s law. The decrease in value of If at higher concentrations is governed by the expression
⎡ (εC l ) 2 (εC l )3 …..⎤
If ϕ I ⎢ εC
Cl − + − ⎥ (5.139)
⎣ 2 3 ⎦
The disadvantage of fluorescence is that it is highly temperature dependent and the intensity of fluorescence
depends on the excitation intensity. Neither of these is true for spectrophotometry.
The fluorescence efficiency of the substances under study can be found by comparing the fluorescence intensity
with that of the standard. Some of the standard materials whose fluorescence efficiencies are known are quinine
sulphate, fluorescein, and 2-aminopyridine.
The total fluorescence yield ϕi arises from contribution of light polarised parallel and perpendicular to the
direction of the excitation light vector.
I f = I ||f + 2I ⊥f (5.140)
Further, information on the flexibility of the fluorescent chromophore and rotational movement of its carrier is
obtained from measurements of the fluorescence anisotropy, ri which is defined as
I ||f I ⊥f ∑ ε i C i ϕ i ri
ri = = (5.141)
I ||f I⊥ f
∑ ε i C i φi
where (ei) denotes the molar decadic coefficients of species i
at the excitation wavelength, ji is the integrated quantum yield If
of the emission spectrum, which is observed at a space angle qf
(Fig. 5.90), and ri denotes the anisotropy of the ith species. qf
Problem 5.56: The calibration graph for luminescence

determination of uranium in a solution of tributylphosphate is
expressed by the equation
5 4
C l gI I0
where C is the concentration (in g/mL) and I, the luminescent inten-
sity expressed in arbitrary units. Determine the concentration of
uranium (in g/mL) if its luminescent intensity is 50 arbitrary units.
Solution Substituting the value of I = 50 into the equation of l′

the problem, we get
C = 2 × 10−5 + 5 ×10−4 log 50 l
or C = 2 × 10 + 5 ×1.698
−5
Fig. 5.90 Schematic diagram of optical path for the
or C = 10−4 [0.2 + 8.490] measurement of fluorescence observed at space angle qf .
or finally C = 8.690 ×10−4 g/mL
Problem 5.57: The luminescence (or fluorescence) of triazinyl stilbexone is quenched by Cr3+ ion and the
results of observations are as follows.
Concentration of Cr3+ (C×107); 0.0 0.4 1.2 2.3 4.4 8.3 12.0 16.0
g ion lit-1
Luminescence intensity arbitrary units 100 95 87 79 68 56 49 44
By using the method of least squares, derive an equation for the calibration curve; assuming the following
linear relation:
1 1
− = kC
I 2 I 02
where I is the luminescence intensity of the solution and I0 is the luminescence intensity of the blank sample.
Determine the concentration of Cr3+ (in g-ion lit−1) from this equation if the luminescence intensity is equal to
75 arbitrary units.
Solution Applying the method of least squares (as discussed in Chapter 1), on the data deduced from the equation
⎛ 1 1⎞
⎜⎝ I 2 − I 2 ⎟⎠ = k C , the value of ‘k’ comes out to be 2.6 × 10 .
2
Thus the equation for the calibration curve is

1 1
− = 2 6 × 10 2 C
I 2 I 02
In order to obtain ‘C’, put the values of I (= 75 units) and I0 (= 100 units) into the above equation
1 1
2
− 2
= 2 6 × 10 2 C
(75) (100)
1
or C= [1.777 × 10 −4 − 10 −4 ]
2 6 × 10 2
0.777 × 10 −4
or C= = 2.98 × 10 −7 g ion lit −1
2 6 × 10 2
Problem 5.58: The luminescence (fluorescence) intensity of an unknown solution of aluminium in
8-hydroxyquinoline is 0.99 arbitrary units. Determine the concentration of aluminium in this solution in (mg/mL)
if the luminescence intensity of the standard solution containing 5 mg of aluminium in 1 mL is 0.45 arbitrary
units under the same circumstances. The luminescence under the conditions of determination is proportional to the
concentration of aluminium.
Solution Since I a C, therefore

I un C un
=
I std C std
C std I un 0.99 × 5( μg/mL)
or C un = = = 11μg/mL
I std 0.45
5.24.3 Apparatus for Detecting Fluorescence (Fluoremeter)

The schematic diagram for the three possible geometries of an instrument for detecting fluorescence is shown in
Fig. 5.91. The apparatus consists of a source of light giving radiation in the ultraviolet and visible regions of the
spectrum, a primary monochromator to select the required wavelength for excitation, a sample cell, a secondary
monochromator to remove the unwanted fluorescence due to impurities, luminescence from the cell and surround-
ings, etc., photomultiplier and the recorder (Fig. 5.92).
Mercury and xenon arc lamps are the most commonly used sources of radiation in fluoremetry. Mercury arcs
of low and high pressures have energy outputs varying from 100 watts to 2 or 3 watts. Low-pressure Hg arcs are
strong line sources at specific wavelengths, but high-pressure arcs give a continuum due to overlapping of lines.
Xe arcs provide a strong continuum with an intensity maximum between 400–600 nm. These days, strong radiation
l
l2
I1 I0
I0
I0
l
— f
2
f
f f I0 f
Sample f
Cell
(a) (b) (c) (d)

Fig. 5.91 Possible illumination and viewing geometriesin a fluoremeter. Observation (a) and (b) at right angles, (c) at a small angle (frontal
method), and (d) in a straight line (in-line method).
Primary
Secondary
Monochro Sample Cell
Monochromator
mator
Ultraviolet
or
Visible Lamp
Photomultiplier
RMS
Chart
Computer Osciloscope
recorder
Fig. 5.92 Schematic diagram of a fluoremeter.
sources up to a few j watts are available.The output of the source is often found with the help of a ferrioxlate/uranyl
oxalate actinometer and is often counted in quanta/second. In a chemical actinometer, the intensity is measured by
the amount of reaction produced by the radiation. A uranyl oxalate actinometer consists of a 0.05 mole lit−1 solu-
tion of oxalic acid mixed with an equal volume of 0.01 mole lit−1 of uranyl nitrate or sulphate as sensitiser. The
actinometer, when exposed to violet or ultraviolet light within a wavelength range of 254 to 435 nm, decomposes
oxalic acid. The extent of decomposition during a given period of time is determined by titration with standard
KMnO4 at the beginning and at the end of the exposure.
Sensitiser UO2++ + hv → (UO2++)*
(UO2++)* + (COOH)2 → CO2 + CO + H2O + UO2++
Quantum yield in the said wavelength range lies in the range 0.49–0.60. In the ferrioxalate actinometer, a trace
of FeCl3 is used as a sensitiser.
2HgCl 2 + ( NH 4 )2 C2 O 4 ⎯FeCl3 → 2NH 4 Cl+2CO 2 +Hg 2 Cl 2 ( precipitate )
Choice of the geometry of the apparatus depends on inner filter effect and the possibility of scattering and
fluorescence of the cell walls. When monochromatic radiation passes through a solution over a certain distance,
it is absorbed and consequently cannot excite fluorescence over subsequent layers of the solution. This is called
inner filter effect. This effect does not cause any serious problem in dilute solutions but in concentrated solu-
tions, this effect cannot be ignored. On the other hand, the major culprit in dilute solutions is fluorescence blank,
which is due to scattering and fluorescence by the walls of the sample cell. In such a case, right-angle geometry is
used. In case of concentrated solutions, one has to choose between the frontal method or the inline method. The
frontal method is the obvious choice if the solution is turbid or if it absorbs appreciably at the wavelengths of the
emitted light possibly due to other substances present and for solids. The fluorescence emission will occur from
the surface of the sample where the absorption of the incident ultraviolet radiation will be greatest and will have
penetrated through a minimum thickness of the sample.
The inline method is useful if the theoretical parameters such as quantum yield are to be estimated from the exper-
imental observations because the theoretical equations correlating the experimental data with the theoretical param-
eters are not based on a fewer number of assumptions. The other most important parameter that needs be stated is the
sensitivity. The term sensitivity in fluorescence and phosphorescence spectroscopy is used in three distinct ways:
(a) Instrumental Sensitivity It is the minimum detectable signal-to-noise ratio in terms of the corresponding
concentration of some standard substance observed under certain precisely specified conditions. The instrumental
sensitivity is governed by the light gathering power of monochromators, sensitivity of photomultiplier–amplifier–
recorder combination, efficiency of various entrance- and exit-slit optics (not shown in figure), etc.
(b) Absolute Sensitivity It is the luminescence sensitivity of a particular substance in a stated solvent and is inde-
pendent of the instrument used to measure it. It depends on molar decadic coefficient and the fluorescence efficiency.
(c) Method Sensitivity This is the overall sensitivity of a particular method of analysis. This type of sensitiv-
ity is determined by one or more of the following factors: Raman emission from the solvent; luminescence of the
sample cell and surroundings, scattered light, luminescent impurities in solvent and in reagents, luminescence of
other components of the specimen.
Problem 5.59(a): An uranyl oxalate actinometer is exposed to light of 435 nm wavelength for 10 minutes.
At the end of 10 minutes, it is found that 0.001 mole of oxalic acid has been decomposed by light. At this wave-
length, the quantum efficiency of the actinometer is 0.5. (a) What might be the average intensity of the light
used? (b) An uranyl oxalate actinometer is exposed to light of 435 nm wavelength for 15 minutes. At the end of
15 minutes, it is found that oxalic acid equlvalent to 12 mL of 0.001 molar KMnO4 has undergone photochemical
decomposition. At this wavelength, the quantum efficiency of the actinometer is 0.58. Find the average intensity
of the light used. (c) The fluorescence and phosphorescence of naphthalene (at 77 K) has been excited by radiation
of wavelengths 274 and 310 nm respectively. The peaks in the former have been observed at 310, 320, 335, 355
nm while in the latter at 475, 490, 515, 565 nm. Draw the energy-level diagram for the fluorescence and phospho-
rescence transitions to the vibrational levels in the ground electronic state.
Solution (a) Number of molecules of oxalic acid decomposed = 0.001 × 6.023 × 1023
= 6.023 × 1020
6.62 × 10 −34
34
(J s)) × 3 × 108 ( / s)
Energy of one quantum of radiation =
435 × 10 −19 m
= 4.56 × 1
100 −19 J
Number of molecules decomposing

Quantum efficiency =
Number of quanta absorbed
6.023 × 10 20
Number of quanta absorbed = = 1.20 × 10 21
05
We know that
Total energy of radiation absorbed
= Energy of one quantum of radiation
Number of quanta absorbed d
Therefore, total energy of radiation absorbed = Number of quanta absorbed × Energy of one quantum of
radiation
= 1.20 × 1021 × 4.56 × 10−19 J = 5.47 × 102 J
5 47 × 10 2 0 91
Intensity of radiation = J = 0.91J / s = × 1019 quanta /s
10 × 60(s) 4 56
= 2 1018 quanta /s
12 × 0.001
(b) Amount of KMnO4 in 12 mL of solution = = 12 × 10 −6 mole
1000
Oxidation of oxalic acid by KMnO4 in presence of sulphuric acid shows that 2 moles of KMnO4 = 5 moles of
(COOH)2
Thus, the amount of oxalic acid that undergoes decomposition in light
= 12 × 10−6 × 5/2 = 30 × 10−6 = 0.30 × 10−4 mole.
Number of molecules of oxalic acid that have undergone decomposition
= 0.30 × 10−4 × 6.023 × 1023 = 1.806 × 1019
Now, proceeding as in the above problem, we obtain
1.806 × 1019
Number of quanta absorbed = = 3 × 1019
0 58
Therefore, total energy of radiation absorbed = 3 × 1019 × 4.56 ×10−19 J = 13.68 J
13.68 J
The intensity of radiation absorbed = = 1.52 × 10 −2 J / s
15 × 60(s)
1 52 × 10 −2 × 1019
= quanta///s = 3.33 × 1016 q /s
4 56
5.24.4 Measurements of Radiative Lifetimes of Excited States

from the Absorption Spectra
The luminescence (phosphorescence/fluorescence) decay of a molecule is accompanied by the depopulation of
the excited state. If the depopulation of the excited state occurs by emission process only, i.e. no radiationless
processes contribute to its depopulation, then the emission lifetime of that state is called the natural radiative
lifetime τ0 of the excited state.
The natural radiative lifetimes of S1 and T1 states can sometimes be determined from absorption spectra by
making use of the relationship
3 47 108 g u
τ0 = v2
⋅ (5.142)
gl
v ∫ ε dv
2
v1
where gu and gl are respectively the degeneracies of the upper and lower electron states involved in the transition.
The degeneracy terms are equal to one for a singlet state and three for a triplet state; ν is the mean frequency
(in cm−1) of the S0 → S1 or S0 → T1 absorption band and ∫ ε dv is the integrated intensity of the absorption band;
∫ = Δ 1/ 2 × ε max , Δv 1/ 2 is in cm−1.
The numerical values of τ0 as determined from absorption spectra of anthracene are τ0(S1) = 13.5 ns (16.7 ns)
and τ0(T1) = 90.0 ms (0.1 ms). The quantities within bras are from radiative emission.
The natural radiative lifetimes τ 0f and τ p0, may be determined from the relationships
τf τp
τ 0f = ; τ p0 = (5.143)
φf φp
where τ′s and ϕ′s are the observed lifetimes and quantum yields respectively.
But it is not always true. The other deactivation processes also compete with the emission process and the life-
time of a state is then determined by the sum of all the processes depopulating the state. Consequently, the observed
radiative lifetime τ will be less than the natural radiative lifetime of the excited state, i.e. τ < τ0. Since the states which
are populated readily are also depopulated readily, the radiative lifetime of a state will be inversely proportional to
the probability of populating that state. Thus, for states which are depopulated by allowed transitions, the radiative
lifetime will be short. The relationship between the radiative lifetime of an excited state and the allowedness of a
transition is approximated by the equation
10 −5
τ≅ (5.144)
ε max
where emax is the molar extinction absorption coefficient at the absorption band maximum. In general, emax values
for S0 → S1 transitions in organic molecules are much greater than for S 0 S T1 transitions. The lifetimes of the
allowed and spin-forbidden transitions approximated by this equation are listed here:
Transition emax, m2 mole–1 Lifetime of excited state; τ, s

S0 → S1 103
10–8(S1)
S0 → T1 10−3 10–2(T1)
The data indicate that the triplet-state molecules will decay slowly as compared to the excited singlet-state
molecules and, therefore, the triplet-state molecules will reside in the system for longer periods than the excited
singlet-state. Consequently, the probability of collision of triplet-state molecules with other solute species will
be high and hence, the chances of participation of triplet-state molecules in photochemical reactions with other
molecules will be high as compared to singlet-state molecules.
5.24.5 Measurements of Lifetimes of Excited States from Fluorescence

or Phosphorescence Intensity
Measurements of the lifetimes of Sl and Tl states will give information on the rates of the photophysical pro-
cesses undergone by excited-state molecules. When the molecules emit fluorescence or phosphorescence, the
lifetimes can be determined from the measurement of decay of luminescence as a function of time. The decay of
luminescence is normally first order and follows the rate law:
−t /τ
Il I lo e e (5.145a)
t
or InI
n l nI l0 −
In (5.145b)
τ
Thus, from the slope of the plot of ln Il against t, we can determine the lifetime of the excited state. Since
compounds which show phosphorescence are likely to show fluorescence also, a phosphoremeter must be able to
differentiate between the two. This is achieved by using a rotating shutter or cylinder such that definite time lag is
introduced between times during which the sample is excited and observed.
The schematic diagram of a single-beam phospho-
remeter used to monitor Ip as a function of time is shown Revolving Shutter
in Fig. 5.93. Monochromatic radiation is made to fall on
a sample through a slit in a rotating shutter. The incident
radiation is cut off from the sample as the cylinder rotates
and there is a preset time lag during which fluorescence is Ultraviolet Monochromator
emitted before the slit in the cylinder arrives in front of the Lamp
Sample Cell
photomultiplier. The output from the photomultiplier is fed Emitted
Phosphorescence Photomultiplier
to an oscilloscope or computer or pen recorder and a curve
of phosphorescence intensity versus time is recorded.
Since fluorescence lifetimes are much shorter than phos-
phorescence lifetimes, mechanical devices shown in Fig.
5.93 can’t be operated rapidly enough to record the fluores-
cence decay. A pulse of short duration (∼10−9 second) from
Chart Oscilloscope
the discharge lamp is used to populate the S1 state. The excit- Computer
Recorder
ing flash is switched off before the emission of fluorescence Fig. 5. 93 Flow sheet diagram of single-beam phos-
is over and the fluorescence decay is monitored using a pho- phoremeter.
tomultiplier and oscilloscope or computer or a pen recorder.
In case of photo-excited molecules which do not emit fluorescence or phosphorescence, the lifetimes of excited
states of such molecules are measured using the technique of flash photolysis. The sample is subjected to a high-
intensity flash of short duration (∼10−9−10−6 second) to populate the S1 state. If ϕ 1 ϕT1 for intersystem crossing is
nonzero then within a period of ∼10−12−10−14 second the T1 state of the molecule will be populated.
Typical flash photolysis apparatus and the electronic excitation processes occurring in a flash photolysis experi-
ments are shown in Fig. 5.94.
Light pulses as short as 1 × 10−14 second are routinely generated these days in the laboratories. The laser output
is split into two parts by a beam splitter. Consider the path travelled by each beam from the beam splitter to the
sample. The path lengths are made different so that the two pulses arrive at the sample at different times. In a
time-resolved laser experiment, some property of the sample is measured as a function of delay time between
these two laser pulses. The laser pulse that initiates the photochemistry is called the pump pulse. The laser pulse
that is used to record changes in the sample since the pump pulse arrived is called the probe pulse. Depending on
the nature of studies under consideration, the pump and probe pulses can be at the same or different wavelengths.
For example, in case of ICN, lprobe = 388 nm and lpump = 360 nm. If the probe laser pulse travels a pathlength that
Trigger
Trigger
Flash Lamp S2
Sample Cell
T2
Spectrograph Spectroscopic
S1
Excitation
Spectroscopic T1
MgO Coated Flash
Lamp
Reflector Excitation
Photographic Plate S0
(a) (b)
Fig. 5.94 (a) Schematic diagram of a typical flash photolysis apparatus (b) Excitation processes taking place in a
flash photolysis experiment.
is 20 cm longer than that travelled by the pump laser pulse, the difference in arrival times of the two pulses of
light at the sample will be
( )
10
= 6.666 × 10 −10 s
3 10 cm / s
The potential-energy curve depicting the time-resolved laser photochemical reaction for the ground and first
excited state of AB (g) molecule is shown in Fig. 5.95
B∗(g)
λprobe
A (g) + B (g)
U (A–B)
λpump
AB (g)
A–B bond distance

Fig. 5.95 Potential energy curve depicting the time resolved laser photochemical reaction.
The intensity of fluorescence caused by the probe pulse is a measure of the concentration of B* (g) free radical
present in the sample. The plot between If and delay time ‘t’ fits into the equation If = 1 − exp (t/τ1/2) where τ1/2 is
the half-life of the reaction.
Since the triplet state has a relatively long lifespan (∼10−5−107 second), so during its lifetime, radiation from a
low-intensity flash from the spectroscopic lamp is passed through the sample in order to measure the absorption
spectra of the triplet–triplet (i.e. T1→T2, T1→T3, etc.) transitions. Thus, from the absorption transition spectra of
S0 → S1 and T1 → T2 transitions, we can determine the lifetimes of the S1 and T2 states respectively. Moreover, from
the absorption spectra of triplet–triplet transitions, the wavelength positions of such transitions can be accom-
plished. The lifetime of the T1 state can then be determined by measuring the decay in intensity of absorption
with time. In order to do this, a flash of monochromatic radiation corresponding to the wavelength of maximum
absorption of T1 → T2 band from the xenon or tungeston lamp is used to excite the molecule and the immediate
decrease in intensity of radiation of the xenon or tungeston lamp is monitored as a function of time. Lifetimes
of excited states of pyrene and phenanthrene as measured by flash photolysis and by emission methods are listed
below:
Compound S1 state a (ns) T1 state b (ns)

Emission FP Emission FP
Pyrene 261 296 0.2 0.7
Phenanthrene 67.2 c
65.0 c
3.3 3.3
a
-measurements made in cyclohexane solution; -in ether - isopentane ethanol at 77 K and, - in rigid potymethyl methacrylate
b c
5.24.6 Rate Constants of Fluorescence and Phosphorescence

The fluorescence emission or phosphorescence emission from an excited-state molecule follows first-order rate
law, i.e.
rate of fluorescence = kf [S1] (5.146)
rate of phosphorescence = kp [T1] (5.147)
where kf and kp are respectively, the specific rate constants for the fluorescence and phosphorescence processes
and [S1] and [T1] are the concentrations of the excited-state molecules giving rise to the radiative emission.
The rate constants for the fluorescence and phosphorescence emissions can be derived from the experimental
values of τf , jf and τp, jp respectively from the following relationships.
1 φf
kf = = (5.148)
τ 0f τ f
1 φp
kp = = (5.149)
τ p0 τ p
The values of kf and kp (in s−1) for benzene and naphthalene have been found to be 1 × 106 and 1.6 × 10+2; and
1 × 106 and 1.6 × 10−2 respectively. For a given molecule kf >> kp or τf << τp.
Rate Constants for Intersystem Crossing The rate constants for the S1 T1 and T1 S0 intersystem
crossings can be determined from the experimentally observed values of the rate constants and quantum yields of
fluorescence and phosphorescence emissions.
Applying the steady-state hypothesis respectively to the S1 and T1 states, we write:
Rate of formation of S1 = Rate of removal of S1
k abs k ic [S
[S ] k f [S
[S ] k isc [S ] (5.150)
Rate of formation of T1 = Rate of removal of T1
k isc [S ] = k isc [T ] + k p [T ]
i′sc [T (5.151)
Equations (5.150) and (5.151) respectively yield
k abs
[ ]= (5.152)
k ic k f + k isc
k isc [S ]
and [ ]= (5.153)
′ kp
k isc
Substituting the value of [S1] from Eq. (5.152) into Eq. (5.153), we write
k isc k abs
[ ]= (5.154)
(k ′
isc )
k p ( k ic k + k isc )
The quantum yields for fluorescence and phosphorescence are given by the expressions:
rate of fluorescence emission k f [S ]
ϕf = = (5.155)
rate of absorption of radiatio
a n k abs
rate of phosphorescence emission k p [T ]

ϕp = = (5.156)
rate of absorption of radiatio
a n k abs
Thus,
ϕ p k p [T ]
= (5.157)
ϕ f k f [S ]
Elimination of [T1]/[ S1] from Eq. (5.157) yields

φ p k p ⎡ k isc ⎤
= ⎢ ⎥ (5.158)
φf k f ⎣ k isc
′ kp ⎦
If total emission yield is high, i.e. jf + jp → 1, then T1 S0 intersystem crossing will be minimal. As a
consequence there of,
kp k i′sc
In such a situation, Eq. (5.158) reduces to
ϕ p k isc
= (5.159)
ϕf kf
Thus, the rate constant kisc for the S1 T1 intersystem crossing process can be determined from .Eq. (5.159)
provided the measured values of the parameters jp, jf and kf are known for the molecule understudy.
Now, in order to determine the rate constant k′isc for the T1 S0 intersystem crossing, we proceed
mathematically as follows:
ϕ f + ϕ p + ϕ isc (T1 S0) = 1 (5.160)
The quantum yield for the T1 S0 intersystem crossing process is given by
rate of T1 ′ [T ]
S 0 intersystem crossing k isc
fisc (T1 S0) = (5.161)
rate of absorption of radiationn k abs
Lumping together Eqs (5.156) and (5.161), we get
kp
fp /[fisc (T1 S0)] = (5.162a)
′
k isc
φ p k i′sc
or fisc (T1 S0) = (5.162b)
kp
Now combining Eqs (5.160) and (5.162b), we write
ϕ p k i′sc
ϕf + ϕ p + =1 (5.163a)
kp
1 − (ϕ f + ϕ p )
′ = kp
k isc (5.163b)
ϕp
Thus, one can calculate the value of k′isc from Eq. (5.163b) provided the observed values of quantities kp, jp
and jf are known. Once the value of k′isc is determined, the value of kisc can be determined by making use of
Eq. (5.158) for those molecules which do not have a high emission yield. Rate constants for S1 T1 (i.e. kisc)
and T1 S0 (i.e. k ′isc) intersystem crossings processes in case of benzene have been found to be 8 × 106 s−1 and
0.10 s−1 respectively. In case of acetophenone, kisc (S1 T1) = 5 × 109 s−1 and k′isc = 50.0 s−1.
For aromatic compounds, the range of magnitude of rate constants for each of the intramolecular processes
originating from S1 and T1 states are given in the self-explanatory Fig. 5.96
kisc = 10 4–1012
S1
T1
kf = 106 –10 9
hν0 kp = 10–2–10 4
6 12
kic =10 –10
k ′isc = 10–2–105
S0
Fig. 5.96 The range of magnitude of rate constants (in s−1) for each of the intramolecular processes
originating from S1 and T1 states for aromatic molecules.
5.24.7 Comparison of Absorption, Fluorescence

and Phosphorescence Spectra
Absorption is shown by solids, gases and liquids. Fluorescence is also known for gases, liquids and solids but
gases will show fluorescence only at low pressure. Different substances fluoresce with light of different wave-
lengths, e.g. fluorspar fluoresces with blue light, chlorophyll with red light, oranges and greens of fluorescent
dyes absorb in the ultraviolet and blue, and fluoresce in the visible, fluorescence which is very marked with NO2
in blue light (492–455 nm) becomes weaker in violet light (455–365 nm) and practically vanishes at or beyond
365 nm, acetone absorbs at 270 nm (corresponding to C=O group) but fluoresces in blue, etc. Phosphorescence
is exhibited by solids but is virtually unknown for dissolved substances and is only observed when phosphor is
frozen into glasses at liquid nitrogen temperature (rather than freezing to a crystalline state) so that collisional
deactivation is prevented or at least severely restricted, i.e. the reduction in temperature enhances the probability
of S1 T1 transition, the required condition for phosphorescence and also to diminish competing processes
for nonradiative return to the ground state. When fluorescent substances are mixed by fusion with boric acid and
are cooled, the resultant mass so produced exhibits phosphorescence, e.g. alkaline earth sulphides and a trace of
heavy metal enhances phosphorescence. A mixture of ethylether, isopentane and ethanol (5:3:2 by volume) called
EPA is the best solvent. The Franck–Condon principle is valid both for the emission and absorption processes,
whereas the absorption spectrum results when the molecule moves from the lowest vibrational level (V′ = 0) of the
ground electronic state (S0) to one or more of the vibrational sublevels (V″) of the excited singlet state (S1). The
fluorescence and phosphorescence results respectively from the lowest vibrational level (V″ = 0) of the excited
singlet (S1) and triplet (T1) states to one or more of the vibrational sublevels (V′) of the molecule (Fig. 5.97).
S1
Dissipation of Vibrational
Energy by Collision Process
2
1
V″= 0 T1
Potential Energy
2
1
V″= 0
Phosphorescence
Fluorescence
Absorption
S0
3
2
1
V ′= 0
Interatomic Distance
Fig. 5.97 Schematic diagram showing Franck–Condon principle in absorption,
fluorescence and phosphorescence processes.
The relative spectral positions of absorption, fluorescence and phosphorescence follow the wavelength order:
− λabs < − λ f < − λ pmax , and that of wave number
−ν abs > −ν f > −ν p .

The intensities of vibrational bands in fluorescence spectra of diatomics follow the order:
I (V ) < I( ) > I( ) > I( ) > I( ).
Further, at a fixed value of wavelength (i.e., λ abs

max
), the linear relationship between the absorption intensity
and concentration of the absorbing substance holds good over a fairly wide range of concentrations. The same is
not true for fluorescence intensity since it is dependent on other competitive processes, viz. internal conversion
and intrasystem crossing available. Thus, as we increase the concentration, collision among solute molecules
will result in quenching of fluorescence intensity. Only in very dilute solutions a linear relationship between the
concentration and the fluorescence intensity is found.
Furthermore, the fluorescence and phosphorescence spectra vary in a similar manner to bands in absorption
spectra, i.e. fluorescence and phosphorescence spectra generally consist of many lines, mostly in the visible
region. In the solution phase, lines are broadened and fuse together to give a structured spectrum having an
appearance more or less similar to that of an ultraviolet or visible absorption spectrum. The short-wavelength end
of a fluorescence spectrum usually overlaps slightly with the long-wavelength end of the absorption spectrum
which gives rise to the excitation, but this overlap may not occur in phosphorescence (Fig. 5.98).
5.24.8 Factors Affecting Fluorescence and Phosphorescence

All molecules can’t show fluorescence and phosphorescence. Only those molecules which absorb ultraviolet or
visible radiation fluoresce. In general, greater the absorption of radiation, greater the luminescence. Some of the
factors which influence fluorescence are the following:
Prompt and
Absorphon Phosphorescence
Intensity
Delayed
fluorescence
Wave number
Wavelength
Fig. 5.98 Comparison of absorption, fluorescence and phosphorescence spectra.
(a) Effect of Substituent Only those organic molecules will behave as fluorescers that contain multiple conju-
gated double bonds with high degree of resonance stability, e.g. benzene and polycyclic aromatics, etc. Moreover,
substituents often affect fluorescence. Though there are no unique set of rules which can predict the effect of
substituent on the fluorescence of the parent fluorescer, some generalities have been made:
(i) Substituents that cause delocalisation of electrons enhance fluorescence of the parent fluorescer, e.g. NH2,
OH, F, OCH3, NHCH3 and N(CH3)2.
(ii) Substituents that promote localisation of electrons decrease or quench fluorescence completely, e.g. Cl, Br, I,
NHCOCH3 or COOH.
(iii) In heterocyclics, pyridine is nonfluorescent but 3-hydroxypyridine behaves as a fluorescer because of the
electron-delocalising substituent, i.e. OH.
(iv) Ring closure in aromatics also enhances fluorescence, e.g. phenophthalene is nonfluorescent but fluorscein
is fluorescent. The bridging oxygen in the latter induces rigidity and hence fluorescence. The higher the
rigidity, the higher will be the fluorescence. In general, we can say that fluorescence in multicyclic aromat-
ics is due to rigidity.
(v) Effect of geometry of the fluorescer—when the fluorescer structure is coplanar relative to the chromophore,
the fluorescence will be enhanced. The higher the degree of coplanarity, the higher will be the fluorescence.
On the other hand, nonfluorescent structures interfere with coplanarity.
(b) Effect of Chelation Metal-chelate formation

often promotes fluorescence. The fluorescence in such Chelating agents Fluorescent Agents
complexes is due to certain structural features such

as rigidity and coplanarity. Minor modification of the
N N
ligand structure may markedly affect the intensity of flu-
O H O
orescence as this may offer alternative paths for energy M
transfer. The fluorescence is stronger in an inert solvent 8-Hydroxyquiniloine
than in a polar solvent. As examples, some chelating
agents and fluorescent chelates are shown below.
Generally, only those cations which are diamagnetic
when coordinated and are nonreducible will form fluo- N
N N
rescent metal-chelate compounds, e.g. magnesium and N
zinc compounds behave as strong fluorescers but are non- OH Ga
O
O
phosphorescers. The transition metals with unfilled outer OH
OH
d-orbitals quench fluorescence completely. On the other 2, 2⬘ - dihydroxyazobenzene
hand, paramagnetic metal ions such as Cu (II) and Ni (II)
on complexation, though quench fluorescence, promote O
OH
intersystem crossing. Consequently, such complexes are
strong phosphorescers but are nonfluorescers. C
C O
O
Be
(c) Effect of Substituents on Phosphorescence O
H
O
The parent cyclic and polycyclic hydrocarbons and their
CH3, NH2, OH, COOH and OCH3 substituent derivatives 2-Hydroxy-3-naphthoic acid
have lifetimes in the range of 5 to 10 seconds and 1 to 4 seconds for many naphthalene derivatives. The nitro group
decreases phosphorescence intensity and shortens the lifetime of the triplet state to about 0.2 second. Carbonyl
groups in ketones and aldehydes shorten the lifetime of the triplet state to about 0.01 second. Bulky substituents
introduce nonplanarity in the parent molecule and hence profoundly decrease the lifetime of the excited triplet state;
consequently the phosphorescence intensity will also decrease markedly.
(d) Effect of Solvent The effect of solvent on fluorescence and phosphorescence is quite different from that
on absorption since in the former, the electron moves from the excited state to the ground state while in the latter,
quite the opposite takes place. Accordingly in solutions, the interaction takes place between solute in the excited
state and solvent both in fluorescence and phosphorescence.
Fluorescence intensity and wavelength often vary with solvent. For nonpolar solutes and solvents, the weak
Van der Waal’s type interaction (i.e. electrostatic interaction) between the solute and solvent does not produce
significant variations in fluorescent yields; whereas for polar solute or solvent, polar solvent or solute and polar
solute and solvent, the Van der Waal’s type binding (i.e. solute-dipoles–induced-dipoles in the solvent or sol-
vent dipoles–induced dipoles in the solute and dipole–dipole), the solvent promotes the lifetime of a collisional
encounter and favours deactivation. Consequently, the intensity variations are significant but small compared to
those produced in H-bonding. For excitation wavelength of 285 nm in each solvent, the λ max f
for indole = 297 nm
in cyclohexane, 305 nm in benzene, 310 in dioxane, 330 in ethanol and 350 nm in water. Fluorescence of chloro-
phyll is markedly influenced upon addition of a polar solvent to dry nonpolar solvent. Moreover, the presence of
heavy substituents such as thiocyanate, Cl, Br, I (i.e. iodine, in ethyliodide), NO2, or —N = N– in solvent cause
an increase in efficiency of S1 T1 intersystem crossing due to spin-orbit decoupling. Consequently, the effi-
ciency of fluorescence decreases while that of phosphorescence increases. The dissolved impurities in solution
may also influence fluorescence and phosphorescence. For instance, dissolved oxygen quenches excited triplet
and singlet states of many molecules, especially of aromatic hydrocarbons, by diffusion controlled process. So,
the solutions should be degassed and freezed before use, since the freezing of a solution reduces quenching. It is
to be noted that the shift in the fluorescence spectrum of a substance in solution compared to that in the gas phase
is because of solute–solvent interactions involved in the former.
(e) Effect of Viscosity The fluorescence increases as the viscosity of the medium increases and this increase
is due to the decrease in the number of deactivating collisions in the viscous medium. For example, the sodium
salt of phenolphthalene which does not show visible fluorescence in aqueous solution, fluoreces when dissolved
in solid gelatin, sucrose, succinic acid or benzamide.
(f) Effect of pH The effect of pH on the fluorescence of an organic compound containing acidic or basic group
is often quite different from that observed for its absorption spectrum. pH changes in the system if it affects the
charge on the chromophore will influence fluorescence: Neutral (pH = 7) or alkaline (pH = 12) aqueous solution
of analine fluoresces in the visible but in acidic solution (pH = 2) the visible fluorescence vanishes due to the
formation of anilinium ion (Fig. 5.99). Aniline also fluoresces in ultraviolet, but this fluorescence is temperature
independent. Both phenol and anisole show fluorescence at 300 and 310 nm at pH 7 but at pH 12 phenol becomes
nonfluorescent due to the formation of phenoxide ion whereas anisole remains fluorescent. Thus, pH sensitive
fluorescent substances can be used as indicators in acidbase titrations. A few are cited below.
100
Fluorescence Intensity
80
60
40
20
2 4 6 8 10 12 14
pH
Fig. 5.99 Behaviour of pH with fluorescence intensity of aniline in aqueous solution.
System pH range Colour change

Erythrosin 2.5–4.0 Colourless to green
b-naphthoquinoline 4.4–6.3 Blue to colourless
o-coumaric acid 7.2–9.0 Colourless to green
(g) Hydrogen Bonding Hydrogen bonding may bring about changes in the position of excited states, n–p *
states which are normally low-energy states as compared to p–p * states, are raised in energy due to intermolecular
H-bonding. Consequently, the nonfluorescent substances become fluorescent in strong polar solvents. Intramo-
lecular H-bonding may inhibit fluorescence because they offer alternative paths for the loss of energy.
(h) Effect of Temperature Temperature and viscosity of a solvent has a marked effect on fluorescence. The
lowering of temperature will prolong the lifetime of a collissional encounter between the solute and solvent
molecule and favour deactivation of the S1 state. Thus, substances fluoresce more brightly in rigid state or in
viscous solutions. The probability of intermolecular energy transfer between fluorescent molecules and other
molecules reduces at low temperature and in a medium of high viscosity in which rotational relaxation time of the
fluorescent substance is much longer than the lifetime of the excited state.
All these environmental factors must be kept in mind while devising a fluoremetric method of analysis.
5.24.9 Intermolecular Deactivation (or Quenching) of Excited States

The deactivation of excited-state molecules by other molecular species present in the system occurs via biomo-
lecular processes. The loss of excess energy of the excited-state molecules by such processes depromotes the acti-
vated molecules to the original ground state. Such intermolecular processes are additional to the intramolecular
processes described earlier by which the decay of the excited state occurs. Thus, the overall decay of the state is
the sum total of the decay rates for the intermolecular and intramolecular paths. As an example, the bimolecular
processes involving in the decay of triplet-state molecule are
(a) Chemical reaction: D* (T1) + Reactant → Prodcuts (radicals or stable products)
(C6H5)2C= O∗ (T1)+(CH3)2CHOH (C6H5)2COH + (CH3)2COH
Benzophenone Isopropanaol
(b) Collision deactivation: D* (Tl) + substrate molecule → D (S0) + substrate molecule
The collision process results in the transfer of energy as heat energy from the triplet- state molecule to the
surroundings. The rate of such collision processes can’t be greater than the rate of diffusion of triplet-state
molecules into the solvent molecules. The rate of diffusion is given by the modified Debye equation:
8RT
k dif f = × 103 dm3 mole −1 s −1 (5.164)
2η
provided the temperature T is expressed in K, the viscosity η of the solvent is expressed in N s m−2 and R, the
gas constant, is expressed in J K−1 mole−1, R = 8.3 J K−1mole−1. For organic solvents at room temperature, the
value of kdiff lies in the range 1 × 1010 − 4 × 1010 dm3 mole−1 s−1.
Since the lifetime of the excited triplet state is long, so such collisions often occur. However, when such col-
lisions take place, they result in the emission of delayed fluorescence.
(c) Energy transfer D*(T1) + A …→ D (S0) + A*
Such type of collisional processes are very common in photochemical systems and provide an efficient
method for the deactivation of excited-state molecules even by very minute quantity of the quencher. O2 is an
effective quencher for triplet-state molecules. So the system must be degassed in order to remove dissolved
oxygen from the system. On the other hand, the decay of the triplet state in different concentrations of dis-
solved oxygen can be used to determine the amount of dissolved oxygen in unknown samples. The details of
the method are deferred here but will be described in detail later on under the title fluorescence quenching.
The various mechanisms by which the donor triplet-state molecule transfers energy to the acceptor molecule
are the following:
Radiative Transfer The general mechanism for radiative transfer can be written as
Donor emission D* → D + hν
Acceptor absorption A + hn → A*
The efficiency of the mechanism increases provided the lmax and intensity of the emission band of the donor
is similar to that of the absorption band. The larger the degree of similarity, the greater will be the absorption of
emitted radiation by the acceptor.
(i) Short-Range Radiative Transfer Such type of radiative transfer occurs provided the intermolecular sepa-
ration between the donor and acceptor molecules approaches the collision diameter. Energy transfer can take
place even without the collision process since energy transfer can take place at distances slightly greater than the
collision diameter.
Short-range energy processes are possible provided the Wigner spin-conservation rule is obeyed, i.e. S–S and
T–T energy transfer are possible
D*(S1) + A(S0) ....→ D(S0) + A*(S1) singlet–singlet energy transfer
D*(T1) + A(S0) ....→ D(S0) + A*(T1) triplet–triplet energy transfer

As an example, short-range energy transfer occurs in the
S1 (n,π*)
benzophenone (D) –naphthalene (A) system by the following isc
T1(n,p∗)
process: Energy Transfer
Short-range intramolecular energy transfer can also take place T1(p,p∗)

between the benzophenone moiety and the naphthalene moiety Absorption
when two groups are present in the same molecule, e.g. Phosphoresence
O S0 S0
Donor Acceptor
C (CH 2) n
Benzophenone Naphthalene
where n = 1, 2, 3, …….
The absorption spectrum of such a molecule is composite absorption spectra of 4-methyl-benzophenone (D)
and 1-methyl-naphthalene (A). Phosphorescence emission characteristic of a p–p* state will be emitted from the
naphthalene triplet state.
(ii) Long-Range Radiative Transfer Such type of radiative transfer occurs even if the distance between donor
and acceptor molecules is much greater than the donor–acceptor collision diameter, provided the energy differ-
ences between the vibrational levels of the ground and first excited states of the donor correspond to the energy
differences between the vibrational levels of the ground and first excited states of the acceptor, e.g. in anthracene
(S1)–perylene (S0) system r = 5.4 (3.1) nm, k × 10−10 = 12(2.3) dm3 mole−1 s−1; perylene (S1)–rubrene (S0) system,
r = 6.5 (3.8) nm; k ×10−10 = 13(2.8) dm3 mole−1 s−1. The quantities within bras are the calculated values.
5.24.10 Kinetics of Intermolecular Processes

Deactivation of the photo-excited molecular species via intermolecular interactions takes place with the loss of
energy through the following two pathways: (a) energy transfer from excited donor (D*) to the acceptor (A) mol-
ecule, and (b) chemical reaction with a suitable reactant. Since inter- and intramolecular interactions go side by
side, the deactivation steps involved in intramolecular processes must be taken into account while deducing the
rate constant expressions for the intermolecular pracesses. The possible inter and intramolecular pathways which
contribute to the deactivation of the excited-state molecules are shown in Fig. 5.100.
The rate constants for the deactivation of photo-excited state via intramolecular processes are evaluated under
experimental conditions such that the intermolecular processes are eliminated but the reverse is not possible. In
this section, we shall be dealing with the kinetics of the deactivation of the excited state by the energy-transfer
process. The energy-transfer process is called quenching if the rate of deactivation by the energy transfer process
is so high that deactivation by fluorescence or phosphorescence process is quenched.
Energy transfer from the photo-excited D* to the ground state of an acceptor A takes place from either the S1 or
the Tl state of the donor with the formation of an Sl or a Tl state of the acceptor, i.e. the energy transfer may take
place via. S1 – S1, S1 − T1 and T1 − T1 transitions. Here, we shall be mainly treating with the kinetics of deactiva-
tion via both spin and energy allowed S1 − S1 and T1 – T1 energy-transfer transitions, It is assumed that the photo-
excited donor D* does not participate in any chemical reaction.
(a) Kinetics of Triplet–Triplet Energy Transfer The various intramolecular processes which compete with the
deactivation of photo excited donor molecule via triplet-triplet energy transfer process are shown in Fig. 5.101.
Excited State
Radiative (f, p) Internal Intersystem Energy Chemical

Emission Conversion Crossing Transfer Reaction
Intramolecular Intermolecular
Processes Processes
Fig. 5.100 Possible pathways for the deactivation of an excited state. Energy transfer from donor to
acceptor also occurs via radiationless process and but is shown by a dotted arrow.
S1
S1 isc
T1
Energy Transfer
T1
ic f
h ν0 p
Abs isc
S0 S0
Donor Acceptor
Fig. 5.101 Energy level diagram representing triplet-triplet energy transfer in
donor-acceptor system.
The reaction scheme showing the inter and intramolecular processes involved in the deactivation of excited
donor D* indicated in Fig. 5.101 can be written as follows:
Process Representation Rate

Excitation S0(D) + hv0 → S1(D) kabs
Internal conversion S1(D) S0(D) kic[S1]
Fluorescence S1(D) → S0(D) + hvf kf[S1]
Intersystem crossing S1(D) T1(D) kisc[S1]
Phosphorescence T1(D) → S0(D) + hvp kp[T1]
Intersystem crossing T1(D) S0(D) ′ [T1]
k isc
Energy transfer (quenching) T1(D) + S0(A) ....→ S0(D) + T1(A) ke[T1][A]
where ke is the rate constant for triplet–triplet energy transfer and kabs is the rate constant for the absorption of
incident radiation.
Applying the steady-state hypothesis respectively to the intermediate states S1 and T1 of the donor we get
kabs = kic[S1] + kf[S1] + kisc[S1] (5.165)
kisc[S1] = kP[T1] + k′isc[T1] + ke[T1][A] (5.166)
Rearrangement of Eq. (5.165) and Eq. (5.166) respectively yields

k abs
[S1 ] = k k f + k isc
(5.167)
ic
and
k isc [S ]
[T1 ] = k ′
k e [A ] k isc
(5.168)
p
The expressions for the quantum yields of phosphorescence in the presence and absence of the energy acceptor
A can be written as
k p [T ]
ϕp = (5.169)
k abs
k p [T1 ] ′
ϕ p0 = (5.170)
k abs
where [T1]′ is the steady-state concentration of the donor in its triplet-state in the absence of the energy acceptor A.
The combination of Eq. (5.169) and Eq. (5.170) yields
ϕ p0 [T1 ] ′
= (5.171)
ϕ p [T1 ]
Now, substituting the value of [T1] from Eq. (5.153) and [T1]′ from Eq. (5.168) respectively into Eq. (5.171),
we get
′ + k e [A ]
φ p0 k p k isc
= (5.172a)
φp ′
k p k isc
ϕ p0 k e [A ]
or = 1+ (5.172b)
ϕp k p k i′sc
If the ratio of the intensities of phosphorescence emission in the absence and presence of the energy acceptor A
ϕ p0
is measured instead of the ratio of then Eq. (5.172b) can be written as
ϕp
I p0 k e [A ]
= 1+ (5.173)
Ip k p k i′sc
Equations (5.172 b) and (5.173) are known as Stern–Volmer equations (See also Eq. 5.197). According to
I p0 ke
Eq. (5.173), the plot of against [A] will be a straight line with slope . From the slope, the value of ke
Ip k p k i′sc
can be determined provided the values of kp and k′isc are known for the donor.
Let us now examine the case in which the deactivation of the donor T1 state proceeds solely via the intra-
molecular processes of phosphorescence (T1→ S0) and T1 S0 intersystem crossing. The lifetime of the T1
state in such a case is given by
1
τ= (5.174)
k p + k isc
′
If deactivation of the T1 state of the donor T1 → T1 via energy-transfer process is also considered in addition
to the T1 S0 and T1 S0 intramolecular processes, then the lifetime of the T1 state will be given by
1
τe = (5.175)
kp ′ + k e [A ]
k isc
Lumping together Eqs. (5.174) and (5.175), we obtain
1 1
= + k e [A ] (5.176)
τe τ
Equation (5.176) indicates that the plot between the reciprocal of the lifetime τe of the T1 state and the concentra-
tion [A] of the acceptor will be a straight line of slope equal to ke and intercept = 1/τ. Biacetyl acts both as a donor
and an acceptor. In the biacetyl–naphthalene system, the rate constants for the triplet–triplet energy transfer to and
from the T1 state of biacetyl (ET = 234 kJ mole−1) in benzene solution at 293 K as determined from Eq. (5.175) have
been found to be 1 × 1010 and 2 × 106 dm3 mole−1 s−1 respec- S1

Energy Transfer
tively. The value of ET for naphthalene in 255 kJ mole−1. isc
The rate constants for triplet–triplet energy transfer between T1 S1
donors and acceptors determined by the above method depend isc
upon the observation and measurement of emitted phospho- hν0 ic T1

f(hνf )
rescence. Since most of substances do not show phosphores- ic f(hν′f )
cence in fluid solution, the rate constants for the triplet–triplet
energy transfer in such D–A systems is determined by flash
photolysis technique.
S0 S0
(b) Singlet-Singlet Energy Transfer Rate constants for Donor Acceptor
singlet–singlet energy transfer can be determined from the Fig. 5.102 Energy-level diagram showing the singlet–
fluorescence emitted by the donor or acceptor species. The singlet energy transfer in donor-acceptor in systems.
hv0 > hvf > hvf′.
energy-level diagram showing the relative energies of the lowest
excited states of donor and acceptor is shown in Fig. 5.102.
The reaction scheme corresponding to the inter and intramelecular processes shown in Fig. 5.102 is as follows:

Excitation S0(D) + hv0→ S1(D) kabs
Internal conversion S1(D) S0(D) kic[S1]D
Fluorescence S1(D)→ S0(D) + hvf kf[S1]D
Intersystem crossing S1(D) T1(D) kisc[S1]D
Energy transfer (quenching) S1(D) + S0(A) ....→ S0(D) + S1(A) ke[S1]D [A]
Internal conversion S1(A) S0(A) k′ic[S1]A
Phosphorescence S1(A)→ S0(A) +hv′f k′f [S1]A
Intersystem crossing S1(A) T1(A) k′isc [S1]A
Applying the steady-state hypothesis to the S1 state of the donor.

kabs = kic[S1]D + kf[S1]D + kisc[S1]D + ke[S1]D [A] (5.177)
Rearrangement of Eq. (5.177) yields
k abs
[S1 ]D = k k f + k isc k e [A ]
(5.178)
ic
The quantum yield of fluorescence in the presence of an energy acceptor A is given by

k f [S ]D
ϕ f (D ) = (5.179)
k abs
kf
ϕ f (D ) = (5.180)
k ic k f + k isc k e [A ]
In the absence of an acceptor, Eq. (5.180) reduces to
kf
ϕ 0f ( D ) = (5.181)
k ic k f + k isc
Thus, the ratio of fluorescence quantum yields in the absence and presence of energy transfer acceptor is
given by
ϕ 0f ( D ) k ic k f + k isc k e [ A ]
= (5.182a)
ϕ f (D ) k ic k f + k isc
or
ϕ 0f ( D ) k e [A ]
= 1+ (5.182b)
ϕ f (D ) k k f + k isc
ϕ 0f ( D )
Equation (5.182b) shows that the plot of fluorescence quantum yield ratio against the concentration of
ke ϕ f (D )
acceptor A will give a straight line of slope equal to .
k ic k f + k isc
From the slope one can determine the value of ke, the rate constant for the energy transfer from the donor sin-
glet state to the acceptor provided the values of the rate constants kic, kf and kisc are known for the donor. The rate
constants for singlet–singlet energy transfer from aromatic hydrocarbons to biacetyl have been determined by this
procedure. For biphenyl (ES1 = 400 kJ mole−1) – biacetyl (ES1 = 277 kJ mole−1) system in aerated hexane solution
at 298 K, the value of ke is equal to 2.2 × 1010 dm3 mole−1 s−1. The high value for the measured rate constant is
because of the large energy gap, i.e. 123 kJ mole−1 between the S1 state of the donor biphenyl and the S1 state of
the acceptor biacetyl.
The experimental values for the rate constants of singlet–singlet and triplet–triplet energy transfer helps in
deciding whether the transfer takes place via the short range or long-range mechanism.
In long-range interactions, the value of rate constant ke should be independent of the viscosity of the solvent
or the temperature of the system. The rate constants for a diffusion controlled process should change when the
viscosity of the solvent or the temperature of the system is changed. The diffusion controlled rate constants indi-
cate that the mechanism involves collision or near collision of the donor and acceptor molecules, i.e. short-range
interaction.
5.24.11 Chemical Reaction

Till now we have discussed the kinetics of triplet–triplet and singlet–singlet energy transfer processes in the D-A
systems on the assumption that there is no chemical reaction between photo-excited molecules and the reactant.
Let us now consider the case in which the photo-excited molecule gets deactivated by a chemical reaction with
some suitable reactant. It is assumed that (a) the photo-excited molecule M* reacts with a reactant R in a single
bimolecular step to give radicals or stable products M* + R→ radicals or stable products; (b) the products of
bimolecular reaction do not react with either M or M*, and (c) only the triplet state of the photo-excited molecule
participates in the reaction though the reaction via singlet state is also possible.
The deactivation of the photo-excited molecules via photophysical and photochemical processes on the above
assumptions involve the following steps:

Rate excitation S0(M) + hv0→ S1(M*) kabs
Internal conversion S1 S0 kic[S1]
Fluorescence S1→ S0 + hvf kf [S1]
Intersystem crossing S1 T1 kisc[S1]
Phosphorescence T1→ S0 + hvp kp[T1]
Intersystem crossing T1 S0 k ′isc[T1]
T1(M*) + R radical or
Chemical reaction kr[T1][R]
Stable products
Energy transfer (quenching) T1(M*)+ S0(A) …→ S0(M) + T1(A) ke[T1][A]
where kr denotes the rate constant for the chemical-reaction steps.
The last step in the above scheme is due to the impurities present in the system or added energy-transfer accep-
tor A; the acceptor quenches the T1 state of M. The energy state T1 of the acceptor should be below that of the T1
state of the photoexcited reactant. Consequently, the rate constant ke for the triplet–triplet energy transfer will be
equal to the diffusion- controlled rate constant, whose value can be calculated from Eq. (5.176).
Applying the steady-state hypothesis to the T1 state of M, we obtain
k isc [S ] = k p [T ] + k ′ [T1 ] + k r [ R ][T1 ] k e [ A ][T1 ] (5.183)
Rearrangement of Eq. (5.183) yields:

k isc [S 1 ]
[T1 ] = k ′ + k r [R ] + k e [A ]
k isc
(5.184)
p
The expression for the quantum yield of the chemical reaction of M* in the triplet state with the reactant R is
kr [ ][ ]
ϕ= (5.185)
k abs
Eliminating [T1] from Eqs (5.184) and (5.185) and rearranging the resultant equation, we get
1 k abs ⎡ kd ke [ ]⎤
= ⎢1 + + (5.186)
ϕ k isc [ 1 ] ⎣ k r [ ] kr [ ] ⎥⎦
where, kd = kp + k′isc
In the absence of acceptor A, Eq. (5.186) reduces to
1 k abs ⎡ kd ⎤
= ⎢1 + ⎥ (5.187)
ϕ 0
k isc [ 1 ] ⎣ k r [ ] ⎦
ϕ0 ke [ ]
= 1+ (5.188)
ϕ kd + kr [ ]
Now, the lifetime of the T1 state of M is equal to the reciprocal of the sum of the rates of the processes involved
in the deactivation of T1 state. In the absence of a quencher, the lifetime of the triplet state, i.e. T1, is given by
1
τ= (5.189a)
kd + kr [ ]
or
1
= kd + kr [ ] (5.189b)
τ
Combining Eqs (5.188) and (5.189b), we get
ϕ0
= 1 + k eτ [ ] (5.190)
ϕ
ϕ0
Equation (5.190) shows that the plot of measured ratio of quantum yields of the reaction in the absence and
ϕ
presence of a quencher against concentration of the quencher and a fixed concentration of R should be a straight
line with a slope equal to keτ. The value of τ at different concentrations of acceptor can be obtained provided ke
is known. Similarly, we can calculate the value of τ at various fixed concentrations of R. Now, once the values
of τ are obtained at different concentrations of the reactant R, we can obtain the value of rate constants kr and kd
by making use of Eq. (5.189 b). The slope of the plot between the reciprocal of the lifetime τ of the T1 state and
the concentration of the reactant will be equal to kr and the intercept of the plot will give the value of kd. The rate
constants for the photoreaction of thymine using cis-1,3-pentadiene as an acceptor are ke (T1 – S0) = 1.1 × 1010 dm3
mole−1 s−1, kd(T1 − S0) = 2.2 × 105 s−1, and kr(T1) reaction with thymine (S0) to give dimeric product = 7 × 108 dm3
mole−1 s−1.
Further, the kinetic of photochemical reactions aids in determining the concentration of substances using the
method of chemical/kinetic method of analysis. In this method, a pseudo first-order condition is accomplished by
adding excess amount of all reactants except the one to be assayed. The rate of the reaction is measured as a func-
tion of the concentration of the substance under examination. Fluorescence is one of the most sensitive indicators
of reaction rates of photochemical reactions. One can measure the initial rate of change of fluorescence of the
reactant or product or the time taken to reach a particular fluorescence level or the fluorescence after a preset time
interval. Any one of these can be used in the assay of the reactant.
5.24.12 Fluorescence Quenching

The decrease in the intensity of fluorescence due to specific effects of constituents of the solution under study is
called quenching. Quenching occurs due to the partial absorption of the fluorescence light by some component
of the solution. When the fluorescent substance itself absorbs the fluorescent light, the phenomenon is called
self-quenching. Quenching results due to the transfer of energy from the activated molecules of the fluorescent
substance to the solvent or other solutes present in the solution by the collision process and subsequently, the
activated molecules relax to the ground state simultaneously, both by the nonradiative mechanism and emission
process. Temperature affects quenching. The higher the temperature, the larger will be the number of collisions
between the fluorescent and quenching molecules and hence larger will be quenching. The fluorescence is also
quenched when the wavelength of the exciting light exceeds a certain value. The study of the effect of quencher
concentration provides valuable information regarding the rates of photochemical processes in solution.
The intensity of fluorescence from an excited state is proportional to the population, i.e. concentration of the
molecules in that excited state.
If = kf [A*] (5.191)
where kf the specific rate constant for fluorescence decay from an excited state and A* is the concentration of the
activated molecules in that excited state.
Rate of formation of the excited state
A* = kabs (5.192)
The excited state relaxes to the ground state simultaneously by both the radiative decay and transfer of energy
to the quencher. Therefore,
Rate of loss of A* = kf [A*] + kq[A*][Q] (5.193)
where kq is the specific rate constant for the decay of A* by the quenching process. In the steady state,
kabs = kf [A*] + kq[A*][Q] (5.194)
Rearrangement of this equation yields
k abs
[ *] = (5.195)
kf k q [Q ]
The fluorescence intensity If , is then expressed as
k f k abs
If [A *] =
k f [A (5.196)
k f k q [Q ]
The inverse form of this equation is called the Stern–Volmer equation, i.e.
1 1 ⎡ kq ⎤
= +⎢ ⎥ [Q ] (5.197)
If k ⎣k k f ⎦
The inverse form is useful to determine the constants involved in the equation and is similar to Eq. (5.173).
1 kq 1
The plot of versus [Q] will yield a straight line with slope = and intercept = . From the slope and
If k abs k f k abs
intercept, we can obtain the values of (kq/kf) and kabs. In order to determine the values of kq and kf rather than their
ratio, a second set of experiments is performed. The excitation source is switched off and the fluorescence decay
to zero is measured as a function of Q. The fluorescence decay is exponential and is governed by Eq. (5.145a).
If I f0 e −t / τ
where If and I f0 are the fluorescence intensities at times t = t and t = 0, respectively and τ is the time constant of
fluorescence.
Rate of loss of A* = kf [A*] + kq [A*][ Q*]
[ *]
or = k f [ *] k q [A *][Q ]
τ
or finally,
1
= k f + k q [Q ] (5.198)
τ
Thus, the fluorescence lifetime τ of the process depends on concentration of quencher. From the slope and
intercept of the graph of 1/τ against [Q], we can determine the values of kq and kf respectively.
Note that unlike fluorescence, the general expression of an exponential form for the decay of phosphorescence
is a complicated one. However, the phosphorescence decay of ZnS phosphor is governed by the hyperbolic law in
its simplest form Ip = At −α (5.199). The value of α varies from 1.0 to 1.7 and t is the time of decay, Ip is the intensity
of phosphorescence and A is a constant. The values of α and A are determined from the plot of ln Ip against ln t.
Slope of the line will be equal to α while its intercept will be equal to ln A.
Problem 5.60: The luminescence (fluorescence) of a solution of a plant pigment activated by 330 nm has been
studied in the presence of a quencher, and the following results are obtained.
[Q], (mmole lit-1) 0.5 1.0 1.5 2.0
If
0.15 0.09 0.06 0.05
I abs
In the second series of experiments, the excitation source is switched off, and the data pertaining to the fluores-
cence decay is indicated below:
[Q] (mmole lit-1 ) 0.5 1.0 1.5 2.0

t, (n s) 38 22.5 16 12.5
Determine the quenching rate constant and the half-life of the fluorescence.
Solution Draw a graph between (1/If ) and Q. The intercept of the graph will give the value of I/kabs whereas the
slope will be equal to kq/kabskf . From the slope and intercept, the value of kq/kf can be determined. The value of kq/ kf
can also be determined from the plot of kabs/If versus [Q]. Here slope of the graph equals to kq/kf .
In order to determine kq, draw a graph between 1/τ and Q. The slope of the resultant curve equals kq and the
intercept equals kf . τ1/2 will then be equal to l n 2/kf The value of kf comes out to be equal to 9.2 × 109 lit mole−1 s−1
and that of τ1/2 comes out to be equal to 1.9 × 10−7 s.
Problem 5.61: The decay of fluorescence of a dye activated by radiation of wavelength 350 nm fits well into
the equation 1/τ = 2.1 × 106 + 8.9 × 106 [Q] provided Q is expressed in mmole lit−1 and τ in s. What might be the
value of τ and τ1/2 for Q = 3.5 mmole lit −1?
Solution Substituting the value of Q = 3.5 mmole lit−1 into the equation given in the problem, we obtain
1
( 2.1 × 106 + 8.9 × 106 × 3.5) s −1
= (2
τ
= 106 ( 2.1 + 31.1) s 1 = 33

33.2 × 106 s 1 .
1
or τ= = 30 × 10-9 s = 30 ns.
33.2 × 106 s −1
and
1 2 0.693
τ1/ 2 = = = 3 3 × 10 −7 s
2.1 10 (s ) 2.1 106 (s 1 )
6 1
Problem 5.62: Which of the answers given characterises correctly the phenomenon of fluorescence
quenching?
(a) Quenching in the presence of a foreign substance is attributed to the fact that a foreign substance emits fluo-
rescence, the fluorescence of the unknown substance becomes increasingly weak.
(b) The fluorescence is quenched when the wavelength of the exciting light exceeds a certain value.
(c) The fluorescence is always quenched as pH of the solution increases.
(d) The quenching of fluorescence by foreign substances (particularly by oxygen of air) is associated with the
interaction of these substances with a fluorescensing substance, which results in the formation of a nonfluo-
rescensing product.
(e) The fluorescence is quenched when the temperature falls abruptly.
(f ) The fluorescence is quenched by foreign substances because they change the ionic strength of the solution.
(g) The presence of impurities results in the fluorescence being quenched and therefore, they must be removed
from the solution.
(h) The fluorescence is quenched as the concentration of the fluorescent substance exceeds a certain value.
[Ans: (b), (d), (h)]
Problem 5.63: The slow reaction of oxidation of lucigenin by hydrogen peroxide is accompanied by a decrease
in the luminescence of lucigenin and is catalysed by osmium. The luminescence of standard solution amounts to
25 arbitrary units after the following time intervals.
COs mg/mL 0.5 1.0 1.5 2.0 2.5 3.0

t, min 11.2 6.25 3.55 2.7 2.17 1.83
Determine the concentration of osmium in the solution if the value of 25 arbitrary units is reached for 4.416 minutes.
Solution The equation for slow reactions is

1 ⎛k⎞
= ⎜ ⎟ Cos (5.199)
t ⎝C⎠
Calculate the value of 1/t from the above data and plot a graph of l/t versus COs. From the resultant graph, the
value of COs corresponding to t = 4.416 min, l/t = 0.226 comes out to be equal to 1.20 mg/mL.
5.24.13 Applications of Fluorescence and Phosphorescence

Fluorescence and phosphorescence spectrometry is used for the analysis of agrochemicals, biochemicals, phar-
maceuticals, petrochemicals, air pollutants, etc., in vivo and in vitro. Keeping in view the various factors affecting
fluorescence and phosphorescence, one can devise fluoremetric and phosphoremetric methods of analysis. The
shift in the spectrum is important for analytical work since one measures the intensity at one particular wavelength
corresponding to the peak of the spectrum. Depending on the solvent, this wavelength will differ. Some of the
methods are described below.
• Direct Method of Analysis This method is applied when the natural luminescent varies linearly with the
concentration of the analyte. The concentration of the analyte in the unknown sample is read off from the standard
graph between luminescent intensity and concentration of the analyte.
• Fluorescence-Increment Method This is just the extension of direct fluoremetric method. In this method,
the fluorescence of a portion of the standard is measured in the same solution with the unknown. This method
enables one to make sure that any impurities have the same effect upon the standard and the unknown. Moreover,
this method is especially useful when minute quantities of the sample are to be assayed from the crude sample. For
example, the amount of uranium in a uranium ore and riboflavin in foodstuff, etc., can be assayed by this method.
• Quenching Method The substances which quench the natural fluorescence/quench fluorescence and subse-
quently enhance phosphorescence can be analysed by measuring the intensity of fluorescence/ phosphorescence
light as a function of concentration of the quencher. Quenching may or may not vary linearly with the concentra-
tion of the quencher.
• Kinetic Method of Analysys This method has been described in detail in section (5.20.4.h). In this method,
the rate of fluorescent of reactant or of product of photochemical reaction as the case may be is measured as a
function of the concentration of the analyte under examination.
Applications of fluorescence and phosphorescence are described here under two subtitles, viz. nonbiological and
biological applications. However, there may be some ambiguity in applications under these subtitles.
(a) Nonbiological Applications Aromatic hydrocarbons, individually as well as in a mixture, and het-
erocyclics are analysed by direct method of fluoremetry. Perylene can be determined in a mixture containing
3,4-benzopyrene and tetracene as major components by using heavy atom. In methyliodide, the fluorescence of
the latter components is quenched. 3,4-benzopyrene has been analysed in smoked food, beer, cigarette smoke, etc.
Usually, chromatography, in conjunction with fluoremetry, is used in such studies.
Direct fluoremetric analysis has been applied to optical brighteners, i.e. fluorescence dyes, paints and glasses.
These absorb in the near ultraviolet at around 350 nm and fluoresce in the violet–blue region. To avoid photo-
chemical degradation, freshly prepared solutions should be kept in dark until the actual analysis is to be made.
Biological important substances showing natural fluorescence, e.g. vitamin A, riboflavin (vitamin B2), drugs
such as quinine and porphyrin and plant pigments are analysed by the direct fluoremetric method. Riboflavin in
foodstuff such as poultry feed can be estimated by dithionite method. These results are calculated according to
the following scheme:
Solution Designation
10 mL of oxidised sample + 1 mL water sample
A
(acid extract is oxidised with KMnO4)
B
(acid extract is oxidised with KMnO4 ) + solid sodium dithionite (reducing agent).
C
(acid extract is oxidised with KMnO4) + 1 mL standard
I Bf I Af ms
Thus, = (5.200) CHOH OH
I Cf I Af m s + m std
CH
where ms and mstd are the masses of riboflavin in the sample and stan- 2
N
dard in the cell. NH
Nonfluorescent biological substances, namely thiamine, can be R
assayed by converting to fluorescent product. Thiamine is commonly R
assayed by the blue fluorescence of its oxidation product, thiochrome.

Even thiamine can be determined in pork and other food products. The Epinephrine, R=CH3 (fluorescent)
sample is treated with phosphatase, an enzyme, which causes hydroly-
(I) Norepinephrine, R = H
sis of the phosphate esters of thiamine which are frequently present in
food materials.
Amino acids containing aromatic systems are fluorescent, viz., tryptophan, tyrosine and phenylanaline. Even
these absorb below 300 nm, the region which excites other compounds occurring in hydrolysates. A chemical
method for tyrosine (reaction with a-nitroso b-naphthol in the presence of nitric acid and nitrous acid) gives a
fluorescent product that can be excited in the visible region at 460 nm.
Simultaneous determination of epinephrine and norepinephrine can be carried out by converting to trihydroxy
indole.
Epinephrine, R= CH3 (fluorescent)
Norepinephrine, R = H (I)
Both amines react at pH 6.5 but only epinephrine reacts at pH 3.5. Thus, by carrying out the fluoremetric
analysis at pH 3.5, one can determine the amount of epinephrine in the mixture.
Acetyl salicyclic acid in the concentration range 0 to 50 mg lit−1 can be determined by direct fluoremeteric
method using 280 nm excitation and emission at 335 nm.
Acrolein and related compounds are nonfluorescent but they react with m-aminophenol to give fluorescent
7-hydroxy quinoline derivative. The derivative, on excitation at 400 nm, emits fluorescence at 510 nm. Concentra-
tion of acrolein of the order of 10−5 mole/mL can be determined.
We know that a number of substances quench the fluorescence of the substances. Thus, the quenchers can be
assayed by the quenching method, Some of the fluorescer-quencher systems are listed here:
Fluorescer Quencher
Eosin Iodide ion
Dyes Oxygen
3,4-benzopyrene Ozone
Rhodamine B Flavanones
Anthranilic acid Glucose
4, 4′-diamino-2,2′– disulphostilbene- Iron (of the order of 10−7 mg/mL)
N-N-N′-N′
tetra acetic acid
Cyanide can be determined by the kinetic method of analysis. The quinones are nonfluorescent, but can be
converted to highly fluorescent cyanohydroquinones upon reaction. The rate of production of fluorescence with
time is measured and is proportional to the concentration of cyanide.
As has already been stated, compounds which show phosphorescence are also likely to show fluorescence. So, a
phosphoremeter must be able to differentiate between the two. This is achieved using a rotating shutter such that a
definite time lag is introduced between the times during which the sample is excited and observed (See Fig. 5.93).
Phosphorescence is invariably weak and it has to be observed at low temperature (= 77 K). Analytical curves are
plotted between phosphorescence intensity and concentration of standard solution of analytes. The standard curves
are linear over a 104 fold concentration range. Negative deviation from straight-line behaviour results in the 10−4
molar and higher concentration range due to self-absorption, molecular association and concentration quenching.
Positive curvature near the minimum detectable concentration (∼10−9 molar) is generally a result of contaminant
luminescence. The contaminants could be detergents used for cleansing glassware, stopcock grease, filter paper,
ion-exchange resins, chromatic materials and plastic containers. Phosphorescence being a long-lived phenomenon
(>10−5 s), it can be used to analyse a mixture where components have varying decay times. By adjusting the phospho-
rescence, one can eliminate one or more components of the mixture.
(b) Biological Applications Spectrofluoremetry has made deep inroads into the fields of biochemistry,
biology and medicine. Many biological molecules such as aromatic amino acids (i.e. acids containing tyrosine
and tryptophan residues), proteins, vitamins, coenzymes, purines and oestrogens are fluorescent under normal
physiological conditions. A wide variety of biologically important components sometimes fluoresce at shorter
wavelengths, so the best methods for analyses of biological compounds are based on the natural fluorescence
which generally occurs at longer wavelengths and is not interfered by many of the other compounds present in
biological material. For example, natural methods are used for vitamin A, riboflavin, drugs such as quinine, and
porphyrin and plant pigments. Nonfluorescent biological materials can be assayed by chemical method of analysis
in which one has to choose a substance which fluoresces or yields a fluorescent product of reaction. The other
methods used in the study of biological compounds are based on fluorescent probes, fluorescent quenchers, etc.
Some of the typical biological applications of fluorescence are the following:
(i) Determination of Concentration of Nonfluorescent Enzymes Nonfluorescent enzymes also catalyse
the photochemical reactions. So, they can be assayed by the kinetic method of analysis. One has to select a sub-
stance which fluoresces or gives a fluorescent product of reaction. The time taken to reach a particular level of
fluorescence is then measured at different concentrations of the enzyme. Thus, based on the equation,
1 ⎛k⎞
= ⎜ ⎟ C enzyme (5.201)
t ⎝C⎠
One can draw a standard curve between l/t and concentration of the enzyme. Now, the time taken to reach a speci-
fied value of fluorescence in the presence of the enzyme sample to be assayed is measured and using the standard
graph, one can determine the amount of enzyme in the unknown sample. For example, lipase can be assayed by
this method, since it catalyses the hydrolysis of the esters of fluorscein.
(ii) Drug Interaction The native fluorescence of biological substances is often profoundly modified by drug
molecules. The modification and degree of modification can thus, be fixed as criterion to accomplish binding and
extent of binding respectively of drug to biological molecules. For example, the natural fluorescence of Bovine
Serum Albumin (BSA) is quenched by androstane and other steroids. Since the absorption spectra of these steroids
do not overlap the fluorescence emission spectrum of BSA, it is inferred that fluorescence quenching results from
a change in protein conformation (but not because of energy transfer from BSA to steroid) due to BSA–steroid
binding. The tryptophan fluorescence of Human Serum Albumin (HSA) excited by 290 nm (bandwidth = 12 nm)
is observed at 335 nm (bandwidth = 12 nm). The tryptophan emission band overlaps extensively the absorption
maximum (∼365 nm) of 4-butyl-l- (p-nitrophenyl) -2- phenyl-3,5-pyrazolidinedione, an analog of antiinflam-
matory drug phenylbutazone. This overlap permits the resonance transfer of energy from the excited tryptophan
residue of HSA to the bound phenylbutazone analog and results in a quenching of native protein fluorescence.
The point of inflexion in the titration curve of the intensity of fluorescence monitored at 335 nm while exciting at
290 nm against drug/HSA is observed at a molar ratio of one, indicating that energy transfer occurs at only one
protein-binding site. The association constant for this interaction has been found to be 5 × 105 mole lit−1.
The intrinsic fluorescence of certain drug molecules may be modified by binding to proteins, DNA, RNA. As
an example, the fluorescent quantum yield of N, N, -dimethyl-l-naphthalene-5-sulphonamide (DNSA) increases
17-fold on binding to bovine erythrocyte carbonic anhydrase. At the same time, the drug fluorescence emission
spectrum shows a large shift to shorter wavelengths. This suggests that DNSA is bound to a hydrophobic region
of carbonic anhydrase.
The binding of nonfluorescent drugs to macromolecules can be studied by means of fluorescent probes such as
dansylglysine (DSG). Note that DSG is weakly fluorescent in water but its fluorescence maximum shifts towards
the blue and its quantum yield increases markedly in solvents of low dielectric constants, such as dioxane. Since
similar changes have been observed when DSG binds to BSA, it is suggested that dansyl-amino acids can be used
as probes for the hydrophobic regions of proteins. The interaction of nonfluorescent anionic drugs (phenylbuta-
zone, dicumarol, flufenamic acid) to serum albumin has been established by measuring their ability to completely
displace DSG. Similarly, the displacement of fluorescent drug warfarin by nonfluorescent dicumarol has been
monitored by measuring the resultant decrease in warfarin fluorescence.
In addition, the technique of fluorescence polarisation can also be used to steady protein–drug interaction.
This technique is only applicable to drug molecules which are highly fluorescent before and after binding to the
macromolecule, e.g. the interaction of chlorpromazine with horseshoe crabmysin B.
(iii) Conformational Equilibria in Biological Macromolecules The conformational equilibria in macro-
molecules can be studied by pulsed fluorescence measurements. The information regarding the conformational
states can be deduced from the changes in the lifetime of excited states of the fluorescent probe placed at a stra-
tegic position in the structure of the carrier molecule. When the isomers T1 and T2 of the structure T affect the
lifetime of the probe differently, one can detect the presence of these two conformations by a pulsed fluorescence
relaxation spectrometer. The fluorescence emission of probe excited by a short pulse of light will decay exponen-
tially to the ground state after the light pulse has vanished; for each conformation with a different time constant.
For the equilibrium,
fast
T1 T 2 , the observed total fluorescence intensity Imf(t) will decay as
2
Imf(t) = ∑a e
i =1
i
−t / τ
(5.202)
In case T is labeled at a single site, the amplitude ai is related to the concentration of the isomer Ti. by
[Ti ]ε i k ie (i = , 2)
ai = Constant [T (5.203)
where ei; is the absorption coefficient and k1e is the emission rate in the absence of a quenching process. When
the ratio ( e ) /((ε1k1e ) is known, the equilibrium constant K = [T2]/ [T1] can be calculated from the ratio of
amplitudes a2/a1.
Phe
When ethidium (ED) – labeled tRNAPhe (tRNAPhe ≡ Phenylalanine– tRNA from yeast tRNA yeast (Fig. 5.103) the
probe is inserted between nucleotide bases) is excited with short pulse of polarised light, the intensity decay of
polarised component ( I11f ) and of the nonpolarised component I mf is monitored and analysed. tRNAPhe is labeled
at the position 16/17 or at the position 37 by replacing the corresponding nucleotides by ethidium bromide (EB).
t RNA Phe A OH
Yeast C
C
A
1 PG C
. . . . . . .
C G
G C
G U
A U
U A
U A GA C A C C U 6 5 4 3
16 NH2 7 2 H2N
h U . . . . . m1 A 8 1
2
m A
GA
h
17 U U
C U CG m 5C U G U
GT C G 9 10
. . . . C Ψ N+ Br −
U
G m7 G
G C2H5
G GA G A C G G
C GA
m2
. . . . .
2
C G
A U
G m5C Ethidium bromide (EB)
A Ψ
Cm A
U Y 37
Gm A
A
Fig. 5.103 Primary sequence of tRNAPheyeast ≡ Phenylanaline –tRNA from yeast.
From the derivative at the position 37 two fractions are separated by column chromatography. They are desig-
nated as tRNAPheEB37B and tRNAPheEB37C and most probably represent isomers with respect to the amino group of
EB which has reacted with the ribose. Note that the natural fluorephore Y base at the position 37 is replaced by
the probe EB. The advantage of EB over the Y-base as a probe is that it has a longer lifetime which facilitates the
analysis of the rotational motion of tRNA.
For tRNAPheEB37C, the decay of Imf is described by at least two exponential terms.
t t
− −
I mf (t ) a1e τ1
+ a2e τ2
(5.204)
Note that the sample is excited at 510 nm and fluorescence is measured at 590 nm. The parameters have been
estimated as τ1 = 34 ns, τ2 = 11 ns, a1 = 4.64 ×10−4, a2 = 4.18 × 10−4.
Let us now examine the effect of MgCl2 on the amplitudes a1 and a2 and on the amplitude ratio a2/a1 of
tRNAPheEB37C. For MgCl2 (in the concentration range between 0–20 m mole lit−1) at 10°C, at all concentrations,
the same two lifetimes τ1 and τ2 are found while the amplitudes corresponding to τ1 and τ2 varies. The amplitude
ratio a2/a1 increases from 0.17 to about 1 when MgCl2 concentration is increased to about 16 mmole lit−1. At
higher temperatures (25 and 35°C), a third term τ3 has to be introduced in order to describe the decay of I fm (t ).
The analysis of data at 35 °C show that τ3 = 20 ns while the values of τ1 and τ2 are the same as that at 10°C.
The variation of the amplitudes a1 and a2 with increasing concentration of MgCl2 are similar to that found at
10°C, the third amplitude a3 vanishes as the concentration of MgCl2 is increased. If the spectroscopic parameters
eikie(i = 1, 2, 3) are the same, the amplitudes a1, a2, a3 are proportional to the concentration of the three excited
states of ethidium. The ‘amplitudes ratios a2/ a1 and a3/a1 thus, correspond directly to the constants describing the
equilibrium between three conformational states reflected by the probe. Thus, it is inferred that tRNAPhe exists
in three conformational states, one of which is populated to a significant degree only at elevated temperature
(> 10°C) and low Mg2+ concentrations (<1 m mole lit−1). Further, note that the same results have been found for
tRNAPhe labeled at position 16/17.
The existence of two different configurations of tRNAPheyeast is also indicated by the analysis of the polarised
component 1f||(t). To simplify the analysis, rotating spheres (each of which is described by one rotational relax-
ation time) are used as models for tRNA.
Thus, for fitting 1f||(t) at least two rotational relaxation times are required.
t / τ1 t /τr t /τr
I f (t ) = a1e ( re ) + a2e −t / τ 2 ( re ) (5.205)
In this model, τ r1 is the rotational relaxation time of the structure T1 with lifetime τ1 and rotational amplitude
r1, and τ r2 is the rotational relaxation time of structure T2 with a lifetime τ2 and rotational amplitude r2. τr1 = 29 ns;
r1 = 0.633 and τr = 14 ns; r2 = 0.680.
2
The presence of two significantly different rotational relaxation times supports the interpretation that tRNA
exists in two conformational states at low temperature. The large difference in τr1 and τr2 suggests a substantial
change in the conFiguration of tRNA.
PROBLEMS
1. Sketch the Fortrat diagram for (a) Bv ″ > Bv ′ and a 20 percent 3. The values of we and xewe for the two electronic states, i.e.
difference between the two. Given Bv ″ ’ = 10 cm–1, Bv ′ =
2
pi. and 2∑+ (2pi ↔ 2∑+), of the CN molecule are 1814.43
8 cm−1. Read off the position of the band head from the cm−1, 12.883 cm−1; and 2068.705 cm−1, 13.144 cm−1
Fortrat diagram and compare it with that determined respectively. The energy difference between the two
Morse curves is 9241.66 cm−1. What is the energy cor-
theoretically. (b) Bv ″ > Bv ′ and a 15 per cent difference
responding to V ′ = 0 → V ′′ = 9 transition in cm−1?
between the two. Take Bv ″ = 10 cm−1 and Bv ′ = 11.5 cm−1.
Determine the value of J corresponding to the band head. [Ans: 9114.59 cm−1]
[Ans: (a) 5 (b) 7] 4. The analysis of the rotational fine structure in the elec-
m1 m2 tronic absorption spectrum of CN molecule correspond-
2. Prove that in isotopic molecules of the type X Y and
ing to the transition 2pi ↔ 2∑+ yields Be (2pi) = 1.7165 cm−1
m1∗
Y (e.g. HC1 and DC1), where m1∗ − m1 = Δm and Δm and Be (2∑+) = 1.8996 cm−1. Using the required data from
m2
X
cannot be disregarded compared to m1, the quantity Problem 3 and from the given data, calculate (a) the force
constants for the fundamental modes of vibration of CN
m2 Δm
ρ −1= in the two states, (b) the energy difference between band
(m1 + ⎡ ∗
+ m1 ⎤
2) head and null line, (c) the CN bond distances in the two
⎣ 1 1
⎦
states, and (d) zero point energies correseponding to the 32094, 31104 cm−1. The intensity of the bands decreases as
two states. the wave number increases. Deduce and assign the frequen-
[Ans: (c) 1.2328 Å, 1.1719 Å] cies corresponding to the fundamental vibration modes of
5. The values of we and xe for the two electron states, i.e. 2pi isoamylnitrite. On the basis of IR studies, the bands at 1600,
+
800 and 600 cm−1 have been assigned to N = O stretching,
and ∑ i , i ↔ ∑ of CO+ are 1562.06 cm−1, 8.661 × 10−3;
2 2 2
N⎯O stretching and ONO bending modes respectively.
and 2214.24 cm ; 6.848 × 10−3 respectively. The energy
−1
⎡ Ans : 0 0 band at 27056 cm−l ⎤
difference between the two Morse curves is 247.926 kJ ⎢ − ⎥
mole−1. What is the energy of corresponding to the 0 →0 ⎢⎣ν ′ (C − O) = 1021 ; ν ′′ (C O) = 1009cm ⎥⎦
band? [Ans 20407.5 cm−1]
6. The energy levels of the rotational fine structure of CO+ 11. Fill the value of X in the following Table:
molecule fit well into the expression Ion or Reaction λmax e Path C A
−1
substance by which (nm) lengh
νelec-rot = ν00 + 3.5666 2
5666 m 0 3878 m , cm being coloured (cm)
determined compounds
Calculate (a) the value of band head and its frequency, are formed
(b) the separation between band head and null line of the
spectrum, (c) the CO+ bond distances in the two states of Al3+ With oxyqui- 2.5 × 10–4
the molecule, and (d) using the data from Problem 5, find 390 6700X 0.836
noline AlR3
the values of dissociation energies in the two states. mole lit–1
Intrinsic
⎡ Ans : (a) 5; 20415.64 cm−1 ⎤ Quinoline 275 45002.00 3 mg lit–1 X
⎢ ⎥ colour
⎢(b) 8.14 cm , Band is red shaded⎥
−1
Intrinsic
⎢ ⎥ Azobenzene 438 11005.00 X mg lit–1 0.356
⎢(c)1115
.1150 cm−11, 12436. cm 1 ⎥ colour
⎢ ⎥
⎣ ( d)10 . 0 0 eV , 5. 58 eV ⎦ 12. The galvanometer scale reading on a single-beam pho-
toelectric colourimeter for a series of solutions having dif-
7. The molar extinction coefficient of beryllium–acetylacetone ferent concentrations are
complex in chloroform is 31,600 for a wavelength of 295 nm.
What is the minimum percentage content of beryllium that Concen-
can be measured in a 2 g sample dissolved in 100 mL, in a tration, C 0 C1 C2 C3 C4 C5 C6
cell of 5 cm path length, assuming that the minimum read- (mole lit-1)
ing on the absorbance scale of the instrument is 0.025 ? The
coloured compound contains a molecule of acetylacetone Deflection 100 80 70 60 50 40 30
per beryllium atom.
[Ans: 7.15 × 10−6 percent] Calculate the absorbances corresponding to the galvanom-
eter scale readings.
8. Calculate the dissociation constant of a reagent HR if at
pH = 7.33 the absorbance Amix equals 0.442. In an acidic • Hint: A = logg I0 = log ( )
medium at pH < 2, the absorbance AHR is 0.017, and in It
a basic medium at pH > 11, the absorbance AR is 0.705.
[Ans: 7.54 × 10−8] ⎡ Ans : For C = C1, ⎤
⎢ ⎥
9. Calculate the contents of copper, cobalt and nickel in a ⎢ A = 0.097; For C = C6 ⎥
mixture from the light absorption by their diethyldithio- ⎢ A= 0.523 ⎥
⎣ ⎦
carbaminate complexes (DDC) in chloroform at 436,367
and 328 nm. The absorbance of the chloroform solution 13. Br2 absorbs at 400 nm and it reacts with acetone. The
of mixture of the ions to be determined as measured in a overall reaction but not the actual molecular collisions in
cell of 1 cm path length are as follows: A436 = 0.45, A367 = reaction is given by the following equation:
0.36; and A328 0.86. The volume of solution analysed spec-
O
trophotometrically is 25 ml; the molar decadic coefficients
of the complexes in chloroform are CH3 C CH3 + Br2
λ (nm) eCu(DDC)2 eCo(DDC)2 eNi(DDC)2 O
436 12850 2260 1720 CH3 C CH2Br + H + + Br −
367 1260 13340 3910 Only bromine molecule absorbs visible light which means
that its concentration can be determined by measuring
328 2230 21820 35210 the intensity of Br2 absorption at 400 nm. The amount of
light absorbed decreases with time as Br2 disappears.
Derive the differential rate equation in terms of absor-
⎡ Ans : CCu 3. 1 × 1 5 mole
l lit ;⎤
1
⎢ ⎥ bance from which one can get information about the

5 1
⎢CCo =197
.97x1
1 mole l lit ; ⎥ actual molecular processes involved.
⎢ −5 ⎥
⎣CNi = 1.02x10 mole
5
l lit 1 ⎦ 14. When light is passed through a layer of solution of 2 cm
thickness, its intensity decreases by 5 per cent. What will
10. The following vibrational bands have been observed in the be the intensity of light when it is passed through a layer
near UV electronic absorption spectrum of isoamylnitrite of the same solution having a thickness of 5 cm?
in the vapour phase: 26035, 27056, 28065, 30127, 29132, [Ans: 88 per cent of the initial value]
15. The absorbance data for various concentrations of 19. From Fig. 5.64 given in the text, choose the correct graph
acetone (in mg/100 mL) after applying the correction of spectrophotometric titration for the cases listed below:
for reagent blank fits well into the equation A = 2.285 × The symbol X represents the unknown substance, T is
10−2 C. (a) Calculate the molar extinction coefficient and the titrant, and P is the resultant product.
specific absorptivity. (b) Calculate the concentration of (a) X and P are coloured and T is colourless.
acetone in the blood and urine of a normal person and a (b) T and P are coloured and X is colourless.
ketonic patient from the following absorption data: normal
(c) X and T are coloured and P is colourless.
blood; A = 0.065, ketonic blood, A = 0.192, normal urine;
(d) T and P are colourless and X is coloured.
1
A = 0.095, ketonic urine, A = 0.200 after dilution. (e) X and P are colourless and T is coloured.
40
(f) T and X are colourless and P is coloured.
The absorbance values are after the correction for the
reagent blank. [Ans: (a) 3 (b) 7 (c) 4 (d) 9 (e) 2 (f) 1]
]
20. A certain volume of solution containing a mixture of
[
Ans: (a) 1.325 × 10−2 ; 2.285 × 10−2 (mg/100)−1 cm−1
amines having different absorption bands, i.e. aniline
(b) 2.84 and 8.40 mg/100 mL; 4.15 and 350 mg/100 mL
and 2-naphthylamine, is titrated with acetic anhydride
16. To determine the traces of copper, the reaction at the absorption maximum of the latter which has a
between H2O2 and ethoxyaniline is used which pro- slower rate of acetylation. Calculate the volume of
ceeds with the formation of coloured products and is acetic anhydride consumed in reaching the each end
catalysed with copper salts. The variation of absor- point if the values of A1 and A2 after the first point of
bance of a solution with time in the presence of vari- equivalence corresponding to the volumes of added
ous amounts of copper is measured and the resultant titrant V1= 1.20 and V2 = 1.40 mL are known to be equal
data are given here. to 0.36 and 0.275 respectively. The absorbances of
the solution, A′ and A″ , before the second end point
Concentration of 0 0.0015 0.0030 0.0050 0.0075 corresponding to volumes of titrant V1 = 1.60 and V2 =
Cu (in mm g/mL) 3.0 are equal to 0.19 and 0.04 respectively. The con-
Absorbance after stant value of absorbance A0 reached at the end of
titration is 0.035 while it is 0.46 before the first end
1 min... 0 0.05 0.09 0.11 0.17 point.
2 min... 0.02 0.07 0.13 0.21 0.30 [Ans: Aniline: 0.96 mL, 2-naphthylamine: 1.95 mL]
3 min... 0.03 0.12 0.21 0.34 0.43 21. A mixture of sodium acetate and o-chloroaniline, 10
mL each in glacial acetic acid, is titrated with 0.1010
4 min... 0.05 0.18 0.28 0.43 0.58
M HCIO4 solution at 312 nm. Sodium acetate does not
5 min... 0.04 0.19 0.36 0.53 0.76 absorb in UV region but is a stronger base than o-chlo-
The values of absorbance for a solution with unknown roaniline. Calculate the original aliquots of sodium ace-
amount of copper under the same conditions are tate and of o-chloroaniline if the values of A1, A2 and A3
after the first end point corresponding to the volumes
Time (in min) 1 2 3 4 5 of added titrant V1. = 4.0, V2 = 6.0 and V3 = 7.0 mL are
known to be equal to 0.67, 0.63, and 0.56 respectively.
Absorbance 0.09 0.17 0.29 0.35 0.45
The constant value of absorbance A0 before this end
Determine the concentration of copper of the unknown point is 0.68.
solution. [Ans: 0.0043 mg/mL] The absorbances of solutions A′, A″ and A′″ before
17. Vitamins D (D2—calciferol; D3—cholecalciferol are fat-solu- the second end point corresponding to volumes of titrant
ble substances which are essential constituents of human V1 = 8.5, V2 = 9.0 and V3 = 9.5 mL are equal to 0.32, 0.20
(animal) diet. They are present in fish-liver oil, butter, milk, and 0.09 respectively. The value of A0 reached at the end
eggs (yolk) and mushrooms. They control the assimilation point of titration is 0.02.
of calcium and the growth of bones and teeth. Deficiency [Ans: Sodium acetate = 0.0681 mole lit−1; o-chloroaniline
of vitamin D in a young organism causes ‘rickets’, i.e. soft- = 0.0313 mole lit−1]
ness and distortion of bones. The absorbance of 3.296 22. Cholesterol is a sterol (C27 H46 O; molecular mass =
×10−5 mole/lit calceferol in alcohol at 264 nm is 0.6. The 386.67) found in animal tissues and brain tissue con-
molecular mass of calceferol is 396 (C28H44O) and l = 1 tains up to 7 per cent (on the dry basis) of cholesterol.
cm. (a) Calculate the molar extinction coefficient, and (b) The deposition of cholesterol in the blood vessels is
determine the per cent content of calceferol in the com- considered to be responsible for atherosclerosis. Cho-
mercial sample if 2 mg of the sample dissolved in 100 mL lesterol or any related sterol solution in chloroform when
of alcohol shows an absorbance of 0.8 at lmax = 264 nm. treated with acetic anhydride: H2SO4 (30 : 1) yields
[Ans: (a) emax = 18203 (b) 87per cent] the corresponding bluish green ester which shows an
18. Determine the pKa of (a) p-nitrophenol, and (b) papaver- absorption maximum at 625 nm. To plot a calibration
ine (cation form) from the following data. curve for the determination of cholesterol colourimetri-
cally, 12.5 mg of cholesterol was dissolved in 100 mL of
p-nitrophenol Papaverine chloroform. Four standard solutions of different concen-
emax × 10–3 emax × 10–4 trations (60–250 mg per mL) were prepared. These sam-
ples were mixed with reagents in appropriate amounts;
pH 317 nm 407 nm pH 239 nm 251 nm
the final volume was made up to 4 mL with chloroform
4.0 9.72 0.33 2.0 3.36 5.90 indicated below. The solutions were allowed to stand
7.0 5.55 9.16 6.4 4.86 3.71 for 15 minutes. After these, the following readings were
obtained on an absorption spectrophotometer with
10.0 1.39 18.33 12.0 6.44 1.56 respect to the blank, i.e. 2 mL chloroform + 2 mL acetic
[Ans: (a) pKa = 6.99; (b) pKa = 6.44] anhydride: H2SO4 (30 : l).
Volume of standard 25. By making use of the data reported in Table 5.28, deter-
cholesterol solution 0.5 1.0 1.5 2.0 mine the value of absorption maximum for
(12.5 mg/10 mL) (a) 3-carbethoxy-4-methyl-5-chloro-8-hydroxytetralone,
Concentration of stan-
62.5 125 187.5 250
dard cholesterol (in mg)
Cl
Volume of chloroform
1.5 1.0 0.5 0.0 5 4 COOC2H5
added (in mL).
Volume of acetic anhy- 3
2.0 2.0 2.0 2.0 2
dride: H2SO4 (30: 1) 1
8
Absorbance at 625 nm. 0.17 0.34 0.51 0.68 OH O
To 0.2 mL of blood serum sample was added 10 mL of alco-
hol: acetone (1 : 1) and the solution was boiled for 15 minutes.
The solution was filtered and the residues were dissolved in
• Hint:
5 mL of chloroform. The absorbance of 1 mL of unknown
solution measured under same conditions was 0.25. (a) Cl
Draw the calibration graph. (b) Determine the amount of m
ring residue
cholesterol in mg per 100 mL of the serum sample.
[Ans: 202.5 mg per 100 mL. Literature value: m-Cl, o-ring residue and o-OH,
o
150–240 mg per 100 mL]
o C
23. Identify theoretically the absorption maximum in the
following compounds. OH O
R
[Ans: 256 nm (obs: 257 nm)]

(b) 3, 4-dimethoxy-4 b, 5, 6, 7, 8, 8 (a), 9, 10-octahydro-
phenanthren-10-one
CH3 HO • Hint:
(273 nm) (234 nm) (257 nm) OMe
CH3—(CH=CH)3—CH3 3
MeO 4
2
(273 nm) 5 1
24. Calculate lmax for the following compounds in ethanol 6
(a) Cholesta −1, 4 − dien −3 − one. 7
10
O
OH 8 9
OMe
OMe
O MeO p
m
O p-OMe, m-OMe, o-ring residue, m
(a) (b) o
o
R = C8H17
C
O
[Ans: 281 nm (obs: 278 nm)]

OO (c) 4-hydroxy-3-methoxy benzaldehyde. [Ans: 282 nm,
(c) (d) (obs: 279 nm; log e = 4.01)]
• Hint: 26. Just like DNA, ribonucleic acid (RNA) is also present in
d+2 animal-and plant-cell plasma. When RNA is treated with
d+1 concentrated HC1, D-ribose, a pentose, sugar present
b OH in it gets converted into furfural which reacts with orcinol
reagent to give a blue-coloured complex. FeCl3.6H2O in
b a d
b the reagent acts as a catalyst. The blue-coloured com-
O
a O g a O plex shows an absorption maximum at 665 nm. To plot a
(a) (b) (c) calibration curve for the determination of RNA, 10 mg of
RNA was dissolved in 100 mL of 98 per cent ethanol. Four
d=2
standard solutions of varying concentrations (10–25 mg)
d+1 were prepared. Appropriate amounts of reagents, i.e.
b d orcinol reagent (3.5 mL of orcinol (6 per cent) in alcohol
O
a g
and 100 ml of 0.l per cent ferric chloride in concentrated
HCl) were added to these solutions, and the final volume
(d) [Ans: (a) 244 nm (obs: 245 nm), was made up to 4 mL with ethanol indicated below. The
(b) 274 nm (obs; 270 nm) solutions were heated in a water bath for 15–20 minutes.
(c) 349 nm (obs: 348 nm), (d) 385 nm] After these, the following readings were obtained on an
absorption spectrophotometer with respect to blank, i.e. The absorbance of 1.0 mL of the extract when measured
1mL ethanol + 3 mL orcinol reagent. under the same conditions was found to be 0.28. Deter-
mine the amount of amino acids in 100 g of flour.
Volume of standard RNA
0.2 0.3 0.4 0.5 [Ans: 250 mg]
solution (10 mg/100 mL)
Concentration of standard 28. Peptide bonds of proteins react with copper sulphate
10 15 20 25 in an akaline medium to yield a blue-coloured complex.
RNA (in mg)
The intensity of the colour enhances when in addition,
Volume of alcohol added the tyrosine and tryptophan residues of proteins produce
0.8 0.7 0.6 0.5
(in mL) blue-coloured products with Folin’s reagent. The resultant
Volume of orcinol reagent intensity of blue colour depends on the amount of these
3.0 3.0 3.0 3.0 aromatic amino acids in proteins. The blue- coloured
added (in mL)
complexes show an absorption maximum at 570 nm. To
Heated the solutions for
plot a calibration curve for the determination of proteins,
15–20 minutes
10 mg of bovine serum albumin (BSA) was dissolved
Absorbance at lmax = 665 in 100 mL of water. Five standard solutions of varying
0.07 0.11 0.15 0.19
nm concentrations (50–150 mg) were prepared. Appropriate
amounts of reagents, i.e. C (A : B: :50 : 1). (A) 2 per cent
Total yeast RNA was obtained by extracting the whole sodium carbonate in 0.1 N NaOH and (B) 0.5 per cent
cell with 90 per cent phenol. The turbid suspension was CuSO4.5 H2O in 1 per cent sodium potasium tartarate in
centrifuged and supernatant was taken in a separating the ratio 50 : 1, and Folin’s reagent (D) were added to
funnel. The upper aqueous layer containing RNA was these solutions, and the final volume was made up to
used for analysis while the lower layer containing DNA 7.5 mL with 0.1 N NaOH, indicated below. The solutions
was disregarded. The absorbance of 0.5 mL of RNA were allowed to stand for 10–15 minutes with rigorous
extract measured under the same conditions was 0.37. shaking. After these, the absorbances of the five solu-
What will be the amount of RNA in 100 mL of the extract? tions were measured with respect to blank, i.e. 2 mL 0.1
[Ans: 10.6 mg/100 mL] N NaOH + 5 mL C + 0.5 mL D.
27. L-amino acid, when boiled with ninhydrin reagent solu-

tion, yields a bluish purple product which has an absorp-
tion maximum at 570 nm. In case of proline, a different Volume of standard BSA 0.50 0.75 1.00 1.25 1.50
product having yellow colour is formed. Asparagine, solution
which has a free amide group, reacts with ninhydrin (10 mg/100 mL)
to give a brown-coloured product. The intensity of the Concentration of standard
bluish purple colour depends upon the amount of proline 50 75 100 125 150
RNA soultions (in mg)
and asparagine in the system. To plot a calibration curve
for the determination of amino acids, 10 mg of glycine in Volume of 0.1 N NaOH add-
1.50 1.25 1.00 0.75 0.50
the present case (or leucine) was dissolved in 100 mL ed (in mL)
of H2O. Four standard solutions of varying concentra-
Volume of solution C add-
tions (20–160 mg) were prepared. Appropriate amounts 5.0 5.0 5.0 5.0 5.0
ed (in mL)
of reagents were added to these solutions, and the final
volume was made up to 5 mL with 0.5 M sodium citrate Volume of Folin’s reagent
buffer (14.5 mL citric acid + 35.5 mL 0.5 M sodium cit- i.e. solution D added (in 0.5 0.5 0.5 0.5 0.5
rate), indicated below. The solutions were heated in a mL)
boiling water bath for 10–15 minutes and allowed to cool.
After these, the following readings were obtained on an Shake the solutions for
absorption spectrophotometer with respect to blank, i.e. 10–15 minutes at room
3.8 mL ninhydrin +1.2 mL buffer. temperature
Absorbance at lmax = 520
Volume of standard glycine 0.068 0.10 0.132 0.166 0.20
nm
solution 0.1 0.2 0.4 0.6 0.8
(10 mg/100 mL)
Concentration of glycine (in mg) 20 40 80 120 160 500 mg of grains were ground with 10 mL of sodium ace-
Volume of ninhydrin (1 per cent tate buffer and supernatant was collected for analysis of
ninhydrin in 0.5 M solution of proteins. If 0.5 mL of the supernatant showed an absor-
3.8 3.8 3.8 3.8 3.8 bance of 0.02, what will be the percentage of proteins in
sodium citrate buffer added
in mL) the given sample. (6.0 percent) [Ans: 6.0 percent]
29. Reducing sugars, i.e. glucose, lactose; galactose
Volume of sodium citrate buffer
1.1 1.0 0.8 0.6 0.4 and maltose, etc., when treated with alkaline solution
added (mL) of copper tartarate reagent and arseno molybdate
Heated the solutions in a boiling reagent, yields molybdenum blue which has an
water bath for 10–15 minutes absorption maximum at 620 nm. To plot a calibration
and allowed to cool curve for determination of reducing sugars, colouri-
metrically 10 mg of glucose was dissolved in 100 mL
of H2O. Four standard solutions of varying concen-
trations (20–80 mg) were prepared. The appropriate
25 mL extract containing free amino acids was obtained amounts of reagents were added to these standard
by treating 0.5 g of gram flour with 80 per cent alcohol. samples, the final volume was made up to 10 mL with
water indicated below. After these, the absorbances

of all the blue-coloured solutions were measured with Volume of HCl (30 per
3.0 3.0 3.0 3.0 3.0
respect to blank, i.e. 1 mL H2O + 1 mL alkaline copper cent) added (in mL)
tartarate reagent + 1 mL arsenomolybdate reagent
+ 7 mL water (for dilution). Heated the solutions
in a boiling water
thermostat at 80˚C
Volume of standard for 10 minutes and al-
glucose solution 0.2 0.4 0.6 0.8 lowed to cool.
(10 mg/100 mL)
Absorbance at lmax =
0.15 0.22 0.30 0.38 0.45
Concentration of glucose 520 nm
20 40 60 80
(in mg)
Water (in mL)
0.8 0.6 0.4 0.2 Using the data from the problem on the unknown sample,
Volume of alkaline copper calculate the amount of fructose if the absorbance of
tartarate reagent added 1. 0 1.0 1.0 1.0 0.2 mL of the extracted sugars in water under the present
(in mL) set of conditions is 0.30. [Ans: 2 per cent]
Volume of arseno molyb- Note that insulin is a polymer consisting of fructose
date added (in mL) 1.0 1.0 1.0 1.0 units with b-2-1 linkage. It is present in onion, garlic
and in many other plant parts. The amount of insulin is
Volume of water added 7.0 7.0 7.0 7.0 expressed in terms of fructose concentration. The extract
for dilution (in mL) for its estimation colourimetrically is prepared as follows.
Heated the solutions in The ground sample is treated with 80 per cent ethanol for
boiling watwer bath for 20 six hours to remove free sugars and the sample is dried.
minutes and then allowed Take 500 mg of dried sample in a conical flask and add
to cool; to it 20 mL of water. Heat the contents in the flask in a
water bath at 90°C for 10 minutes. Collect the extract and
Absorbance at 620 nm 0.11 0.22 0.33 0.44 add 70 mL of water to the remaining residues in the flask.
Repeat the extraction process for 30 minutes and collect
Using the data from the problem on the unknown sample, the extract again. Pool the extracts and filter the solution
calculate the amount of reducing sugars if the absorbance for clarification and make up to 100 mL.
of 0.2 mL of extracted sugars in water under the present set 31. Absorption spectra of Ni [(CH3 )2 SO ]2+
6 and
of conditions is 0.15. Ni [CH 3 C(O ))N 2+
N (CH3 )2 ]6 ions show absorption maxima
[Ans: 1.4 per cent] at 7728, 12970, 24038 cm−1 and 7576, 12738 (14285),
30. Fructose, a keto-hexose is usually accompanied by 23809 cm−1 respectively. The value of Racah parameter
sucrose in fruits like apple. Honey is a rich source of B for the Ni2+ gaseous ion is 1056 cm−1. Using the given
fructose. When treated with concentrated HCl, it gives spectral data and energy level diagram for d 7-configura-
hydroxymethyl furfural which then reacts with 0.1 per cent tion, calculate the values of Dq, B′, b and b ° and the posi-
resorcinol in 95 per cent ethanol to give a red-coloured tions of the absorption maxima corresponding to the
product which has an absorption maximum at 520 nm.
3 3
To plot a calibration curve for determination of fructose A 2g T 1g (P ) and
colourimetrically, 10 mg of fructose was dissolved in 100 3
A → T1 (F ) transitions in both the complexes.
3
2g
mL of H2O. Five standard solutions of varying concentra-
tions (20–60/mg) were prepared. Appropriate amounts
of reagents were added to the standard samples, the [Ans: Ni[(CH3 )2 SO]2+ 6 , 772.8 cm−1, 921.5 cm−1,
final volume was made up to 5 mL with water, indicated 0.872, 12.73 per cent, 24028 cm−1 and 12970 cm−1;
below. The solutions were then heated in a thermostat N (CH3 )2 ]62+ , 757.6 cm−1, 921 cm−1, 0.87,
[CH3C(O ))N
at 80°C for exactly ten minutes. After these, the absor- 12.78 per cent, 23808 cm−1 and 12734 cm−1]
bances of all the solutions cooled to room temperature
32. The electronic absorption spectrum of [Cr (F)6]3− exhibits
were measured with respect to blank, i.e. 1 mL H2O + 1
three bands at 34400, 22700 and 14900 cm−1. Calculate
mL 0.1 percent resorcinol + 3 mL 30 per cent HCl at lmax
the values of Racah parameters and that of the bending
= 440 nm.
constant. The value of B for bare Cr3+ ion is 918 cm−1.
[Ans: B′ = 826.6 cm−1, b = 0.90; b° = 9.9 per cent and
x = 4120 cm−1
Volume of standard 0.2 0.3 0.4 0.5 0.6
33. The electronic absorption spectrum of [Co(F)6]4− shows
fructose solution
three bands at 7150, 15200 and 19200 cm−1. Calculate
(10 mg/100 mL) the values of Racah parameters as well as that of Dq and
Concentration of frcu- bending constant. B(Co2+ ) = 971 cm−1.
20 30 40 50 60 [Ans: Dq = 805 cm−1, x = 710 cm−1, B′=863 cm−1,
tose soultions (in mg)
b = 0.89; and b° = 11.09 per cent ]
Volume of water
0.80 0.70 0.60 0.50 0.40 34. The electronic absorption spectrum of tetrahedral
added (in mL)
[Co(TMG)4]2+ where TMG is tetramethyl guanidine exhibits
Volume of 0.1 per
a doublet with maxima at 530 nm (18867 cm−1), emax = 204
cent resorcinol 1.0 1.0 1.0 1.0 1.0 and 590 nm (16949 cm−1), emax = 269 and is assigned to
added (in mL) 4
A2→ 4T1 (P) transition. The near infrared spectrum shows
a triplet: 1204 nm (8306 cm−1), emax=91.5; 1320 nm (7576 39. The following data are reported on the CT band observed
cm−1) emax =85.0 and 1540 nm (6494 cm−1), emax = 23.5 and around 300 nm in the 2, 4-dinitrophenol (as an electron
is assigned to 4A2→ 4T1(F) transition. The splitting of T1 (F) acceptor) −p systems.
states is because of spin orbit coupling and the weightage
factors of the respective triplet peaks are (–9/4) l′, (+6/4) l′ Donor emax (dm3 mole–1 cm-1) Δn–1/2(cm–1)
and (+15/4) l′, calculate the values of the Racah param-
eters as well as the energy in cm−1 for the 4A2→ 4T2 (F) chlorobenzene 2042.0 590
transition.
[Ans: B ′ = 811 cm−1, b = 0.83, b° = 16.5 per cent, E [ 4A2 p-xylene 2491.0 827
→ 4T2 (F) = 17908 cm−1]
Hexamethylbenzene 2777.0 967
Hint: The average energy corresponding to the 4A2 → 4T2
(F) transition
Calculate the values of oscillator strengths and transition
( + )cm−1 −1 and to the dipole moments for the CT band observed in the donor−
= = 17908 cm
2 acceptor systems.
4
A2 → 4 T1(F ) transition ⎡ Ans : 5.2 × 1 3 , 0.57D; 8.9 × 10 3 ,⎤
⎢ ⎥
4 ⎡9 −1 ⎤ 6 −1 15 −1 ⎣0.75D; and 11.6 .6 10−3 , 0.86D ⎦
= (6494 )⎥ + (7576
5 6 )+ (8306 )
30 ⎢⎣ 4 ⎦ 4 4
40. Calculate the concentrations of tyrosine and tryptophan
4 −1 −1 −1 in a sample obtained by hydrolysis of a protein from the
= (14612 + 11612 + 31148 )
30 absorption data indicated below.
4 −1
= (57124 m−1
) = 7616.5 cm
30 lmax
emax (dm3 mole–1 cm–1)
35. Calculate the value of lmax for CH3, ⎯ (CH = CH)3 COOH. (nm) Absorbance (Hy-
[Ans: 294 nm] drolysed protein
36. The specific rotation of the polarisation plane for ephed- Tryptophan Tyrosine sample)
rine and pseudo-ephedrine in a 68 per cent water–ethanol
solution at various wavelengths has the following values. 240 2.00 × 103 1.12 × 104 0.660
280 5.40 × 103 1.50 × 103 0.221
Wavelength, nm 313 334 365 405 43 6 546 589
Specific rotation, ⎡ Ans : 2.58 10−55 mole

l lit 1;⎤
degrees ⎢ −5
5 ⎥
⎣5 43 10 molel lit 1 ⎦
Ephedrine 44.5 20.0 12.6 6.28 2.96 0 - 0.24
Pseudo-ephedrine 108 173 138 107 86.7 54.2 46.1 41. Calculate the Doppler half-intensity width of sodium line
at 5893 Å in pm if the light source is at 500 K and R = 8.3
× 107 ergs mole cm−1 [Ans: 1.9 pm]
Find the conditions for a polarimetric determination of
ephedrine and pseudo-ephedrine in their mixture and set 42. The fluorescence decay of a dye excited by
up the necessary equations for calculation. light of 350 nm wavelength follows the equation
[Ans: Cpe = 1.64 [a]546/l, Ce = [[a]334− 3.18 [a]546]/0.05l 1
15 107 7 7 × 107 [Q ] provided τ is in second
= 1.15
37. The specific rotation of the polarisation plane for glucose τ
at various wavelengths has the following values: and Q is expressed in units of m mole lit−1. Determine the
value of τ and τ1/2 if [Q] = 6.0 m mole lit−1.
Wave- 447 479 508 535 656 ⎡ Ans : τ 1/ 2 0 06 μ s;⎤
length, ⎢ ⎥
⎣τ = 2.1 ns. ⎦
nm
43. The relation between the luminescence intensity of a
Specific
solution of a mixture of luminol, copper ammoniate and
rotation, + 96.62 + 83.88 + 73.61 + 65.35 + 41.89
hydrogen peroxide and the concentration of fluoroglu-
degrees
cine used as inhibitor is characterised by the following
data.
Determine the specific rotation of the polarisation plane
for the characteristic hydrogen lines: yellow (589 nm),
blue (486 nm), and violet (434 nm). Conc of
[Ans: a589 = 52° 1′ a486 = 80°, a434 = 104°] inhibitor, 0 0.013 0.023 0.071 0.141 0.269 0.446 0.708
mole ×103
38. (+) Malic acid rotates the plane of polarisation to the
right, and (−) malic acid rotates it to the left. The specific Lumines-
rotation of the polarisation plane for both acids is 2.3°. cence 180 143 90.4 53.3 38.5 27.3 20.2 14.3
Determine the content of (+)− malic acid in a solution for intensity,
which the angle of rotation of the polarisation plane is
arbitrary
+ 0.8° if the total concentration of the acids is taken to be
units.
equal to unity. [Ans: 0.67 g /100 mL]
Determine the concentration of fluoroglucine (in mg/mL) if its (c) The transfer ‘e’ occurs in two stages. First, the heat
luminescence intensity equals 45 arbitrary units. energy is evolved and then the energy in the form
[Ans: 1.26 x 10−2 mg/mL] of luminescent radiation.
44. The degree of quenching of the luminescence of rhod- (d) The transitions ‘c’, ‘e’, and ‘f’ take place in several
amine by zinc is characterised by the following data. stages, each of which involves the evolution of radi-
ant energy in the form of luminescent radiation.
(e) The transitions ‘c’ and ‘f’ involve the evolution of
Conc, of zinc,
1 2 3 4 5 6 7 heat energy when electrons are transferred within
mg/5mL
the confines of one level, and radiant energy in the
Luminescence form of luminescence when the electrons are trans-
63 50 45 39 32 29 25 ferred from one level to another.
intensity,
Per cent of initial (f) Of all the electron transfers given, only the transition
‘e’ involves luminescent emission, the magnitude of
the quantum of absorbed energy being higher than
A 0.5 g sample of indium sulphide containing traces that of the quantum of radiant energy.
of zinc is treated to prepare 100 mL of a solution. (g) In ‘c’, ‘e’, and ‘f’, the transfer of electrons from
The addition of 5 mL of this solution to rhodamine one sublevel to another involves a weak lumines-
under the conditions in which the calibration curve cent emission, and in transitions from one level to
is plotted reduced its intensity to 42 per cent. Deter- another, heat energy is evolved.
mine the percentage content of zinc in the sample. (h) The transitions ‘d’ and ‘g’ are accompanied by the
[Ans: 0.013 per cent] emission of a strong luminescence.
45. To determine iron in waste water by luminescence inten-
sity of its compound with stilbexone, a standard solution 2
is prepared by dissolving 0.586 g of chemically pure E2 1
0
Fe2O3 containing 2.5 per cent of hygroscopic water, in
250 mL of water. Stilbexone is added to the following vol-
umes of the standard solution, which are then made up
to 100 mL. The following data are obtained.
2
E1 1
Vol. of stan- 0
dard solution; 0.5 1.0 1.5 2.0 2.5 3.0
mL
Luminescence 2
18 31 42 51 58 64 E0
intensity, 1
×10+2, arbitrary 0
(a) (b) (c) (d) (e) (f) (g) (h)
units
To 5 mL of sample of waste water under analysis, Fig. X

stilbexone is added and the resultant solution is diluted [Ans: (b),(c),(f)]
with water to 25 mL. The luminescence intensity of 48. (a) The quantum yield of ICN into I(g) and CN∗(g), X 2S +)
the solution so obtained equals 0.25 arbitrary units. by a 306 nm pump pulse is 1.00. If the energy of the pump
Determine the amount of iron in the waste water (in pulse is 1.5 × 10−7 kJ, determine the number of CN∗ (g)
g/m 3). (B2 S+) radicals produced per pulse if only 0.2 per cent of
the incident light is absorbed by the ICN (g) sample.
[Ans: 30 g/m3]
46. A 0.15 g sample of uranium ore is dissolved and after • Hint: Energy of a 306 nm photon = 6.49 ×10−22 kJ.
appropriate treatment, the solution is diluted with water 02
Number of photons absorbed = × number of
to 100 mL. The luminescence intensity of the solution 100
has been found to be 60 arbitrary units. When 5 mg of photons in a 1.55 × 10 kJ pulse.
−7
uranium is added to 20 mL of this solution, the lumines-

[Ans: 4.78×1011]
cence intensity is raised to 110 arbitrary units. The lumi-
nescence intensity of a blank sample is equivalent to that (b) Time resolved laser studies for the reaction in part
of 1 mg of uranium. (a) lpump = 306 nm and lprobe = 388 nm yield the following
What might be the content of uranium in the ore in kilo- data.
gram per ton, if the luminescence intensity is proportional
to the amount of uranium present? Intensity (If in 1 0.979 0.946 0.857
47. Figure X shows schemes for transfer of electrons from arbitray units) 0.632 0.386 0.0476 –0.637
various levels and sublevels of molecules. Which of the
following answers explain correctly the energy processes Time of delay (in f s) 1200 800 600 400
taking place in these transitions? 205 100 10 –400
(a) The transfer ‘a’ involves the evolution of energy as
heat. Plot a graph between If and delay time and fit the resul-
(b) The transition ‘h’ involves the liberation of an tant curve into some suiTable function. Determine also
absorbed energy quantum in the form of a radiation half-life of the reaction.
of the same frequency. Such transitions are hardly 49. The upper level of the H2 (g) laser is the lowest excited
probable. state of the molecule, the B 1Su+ state and the lower level
is the X 1Sg+ ground state. The lasing occurs between 51. (a) The force constants for the diatomic molecules B2
the V″ = 5 level of the excited and V′ = 12 level of the through F2, are given below: (i) Is the order what you
ground state. Determine the wavelength of the laser light expect? Explain. (ii) Compute the vibrational frequencies
from the H2 (g) laser by using the spectroscopic data of the diatomics.
indicated below.
Molecule B2 C2 N2 O2 F2
State νelec(cm-1) ωe(cm-1) ωexe (cm-1)
k(millidynes/Å) 3.50 9.30 22.60 11.40 4.50
B 1Su+ 91689.9 1356.9 19.93
X 1Sg+ 0 4401.2 121.34 (b) Compare the bond orders of O2+ O2 , O2− and O22− ,
A 1.0 ns pulse can be produced with a pulse radiant C2 and C2− ; NO+ and NO
power of 100 kW. Calculate the radiant energy of such a
• Hint: Write the electronic configuration and compute
laser pulse and the number of photons in this pulse.
the bond orders.
[Ans: 160 nm, 1.0 × 10−4 J, 810 × 1011 photons]
1
50. The term symbols for some of the homonuclear diatomics Bond order = [(number of electrons in the bonding orbit-
2
are given below:
als) − (number of electrons in the antibonding orbitals)]
The larger the bond order, the smaller the bond length.
Molecule Term symbol
⎡ 5 3 5 5⎤
H2, C2, Li2, F2
1
Σ g+ ⎢ Ans 2 , 2, 2 , 1; 2, 2 ; 3, 2 ⎥
⎣ ⎦
He2+
2
Σ u+
B2, O2
3
Σ g−
O2+ 2
pg
Write the meanings of the notations involved in the term

symbols.
CHAPTER 6 PHOTOELECTRON
SPECTROSCOPY
I have little patience with scientists, who take a board of wood, look for its thinnest part, and drill a great number
of holes where drilling is easy.
—Albert Einstein
6.1 INTRODUCTION
Photoelectron Spectroscopy (PES) involves the measurement of parameters, namely kinetic energies, abundance’s
and angular distributions of photoelectrons emitted from the occupied atomic orbitals or molecular orbitals,
i.e. bonding, nonbonding and antibonding of a sample irradiated by high-energy photons. The photoelectron
parameters are a function of the element of the sample irradiated and its electronic environment. They are
characteristic of the individual energy of the individual atomic/molecular orbitals from which they originate and
binding energy of the individual electron in the element. It is presently the best available technique for directly
measuring the ionisation energies (IE) or potentials (IP) or binding energy (BE) of electronic states of positive
ions which are closely related to the energies of different atomic/molecular orbitals in neutral atoms/molecules.
PES has a symbiotic relationship with atomic/molecular orbitals. The language of AO/MO proved itself to be
the most suitable vehicle for understanding the photoelectron spectrum. According to Koopman’s theorem, the
verticle ionisation energies, IEj (or ionisation potentials, IPj ) of an atom or a molecule is equal in magnitude to
the orbital energies Ej, determined by Self-Consistent Field (SCF) molecular orbital calculations;
IPj / IEj = −Ej (6.1)
where IEj is taken as the difference in energy between an electron at an infinite distance from the atomic/ molecu-
lar ion and the same electron in the atom/molecule.
According to this theorem, one band is expected per orbital.
There are two types of ionisation energies encouatered in PES, viz., vertical and adiabatic. The vertical ionisation
energy (Ivert) refers to the most probable transition predicted by the Franck Condon principle and corresponds to
the experimentally measured value (Fig. 6.1), while the adiabatic ionisation energy (Iad,) refers to V ′ → V ″ =
0 → 0 transition, and corresponds to true ionisation potential of atom/molecule (Fig. 6.1). Ivert is always greater
than Iad.
6.2 TYPES OF PHOTOELECTRON SPECTROSCOPY

Traditionally, it has been subdivided according to the source of exciting Molecular Ion
radiation for ionisation into the following:
(i) Vacuum ultraviolet photoelectron spectroscopy (UPS or UV-PES)
which uses vaccum UV (10 – 45 eV) radiation to examine the valence
Potential Energy
V⬙ = 0
levels. In UPS, the outer-shell electrons are removed from an atom/
(1)
molecule using He (I), 21.22 eV as UV radiation source and hence is
(2) Neutral Molecule
also called PhotoElectron Spectroscopy for Outer Shells (PESOS). It
was discovered independently by two groups, one led by F I Vilesove in
Leningrad (Russia) and the other by D W Turner in London (England) in
V⬘ = 0
early 1960. (ii) x-ray Photoelectron Spectroscopy (XPS) uses soft x-ray
(200 – 2000 eV) radiation to examine the core levels. XPS when used for
r
analysis is termed Electron Spectroscopy for Chemical Analysis (ESCA). Fig. 6.1 Schematic diagram showing the
In XPS, the inner-shell electrons are removed from an atom/molecule difference between vertical and adiabatic
using MgKa’ 1253.6 eV or AlKa 1486.6 eV as x-ray radiation source and ionisation energies.
hence is also called PhotoElectron Spectroscopy for Inner Shell (PESIS) emission. This technique was developed
by K M Siegbahn and his group at Uppsala (Switzerland) in 1946. The studies were concentrated more on solids
than on free molecules. The most important and interesting aspect of XPS is that the core electron-binding energy
of an atom in a molecule is a significant factor of the chemical environment of the atom. In 1981, the Nobel prize
was awared to him for his contribution to the development of high-resolution electron spectroscopy, especially
ESCA. Both the techniques are used to determine the structure of atoms/molecules and for analysis; the detection
limit being of the order 10−5 mg.
6.3 PHOTOELECTRIC EFFECT

It is the very basis of photoelectron spectroscopy. Therefore, it is imperative to discuss in brief this phenomenon.
When light of suitable wavelength strikes a metal surface, it acquries a positive charge with the emission of elec-
trons. It was accidently discovered by Hertz in 1887 as he was investigating the electromagnetic waves predicted
by Maxwell’s theory of the electromagnetic field. The interpretation of the photoelectric effect was proposed by
Albert Einstein on the basis of quantum theory. According to him, when a photon of energy hv falls on a surface,
an electron is emitted with some velocity and the kinetic energy of the electron given by
1
me v 2 = hν − ϕ e (6.2)
2
where me is the mass of the electron and je is a constant characteristic of the metal. This is also called the
electronic work function of the metal which is defined as the minimum energy required to liberate an electron
from the attractive forces of positive ions in a metal or a semiconductor at absolute zero temperature. The lower
the work function, the greater the number of electrons that can break loose from the specimen. This relationship
is called the photoelectric law. Further, it is to be noted that (i) the number of photoelectrons liberated is propor-
tional to the intensity of the impinging radiations, and (ii) the minimum kinetic energy of photoelectrons depends
on the energy, but not on the intensity of incident radiation. In 1921, Einstein was awarded the Nobel prize for
theoretical physics and especially for correctly developing the theory of photoelectric effect.
The first direct experimental evidence to Einstein’s law was produced by Robert Andrews Millikan in 1914
and in 1923, the Nobel prize was awarded to him for his work on elementary charge of electricity and on the
photoelectric effect.
6.4 THEORY OF PHOTOELECTRON SPECTROSCOPY

Before describing the theory of photoelectron spectroscopy, it Vacuum level
is useful to briefly review the electronic structure of isolated M 3s/3p
atoms and of solids, and the associated nomenclature with them.
According to conventional chemical nomenclature, the electronic L 2s/2p
structure of isolated atoms and of solids along with their associ- Binding energy
ated nomenclature has schematically been illustrated in Fig. 6.2 K 1s
(a), (b) and Fig. 6.3. respectively. The core levels or shells K, L,
(a)
M… of atoms are given on the left-hand side while the orbitals
associated with the respective shells are given on the right-hand
side. However, scientists working with x-rays use the ‘alternative
nomenclature’ on the left. The shells or core levels designated Vacuum level
3d/4s/4p
as K, L, M… correspond to the principle quantum numbers
1, 2, 3... respectively. In Fig. 6.2. (b), the numerical component
MII, III 3p
of K, L, M… style of nomenclature is usually written as a sub- MI 3s
script immediately following the main-shell designation. Levels Binding energy
with nonzero value of the orbital angular momentum quantum
2p
number, i.e. azimuthal quantum number (l > 0), or p, d, f, levels, LII, III
LI 2s
show spin-orbit splitting. When the magnitude of this splitting is
(b)
too small, the double subscript for these levels represents both
Fig. 6.2 (a) Schematic diagram showing the electronic
the levels, i.e. LII III represents both the L2 and L3 levels.
structure of isolated atoms and the nomenclature
Figure 6.3 shows that in the solid state, the core levels associated with it (b) An expanded view of the part of the
of atoms are little perturbed but remain as discrete, localised energy scale of Fig. 6.2 (a) close to the vacuum level.
Photoelectron Spectroscopy 459
(i.e. atomic-like) levels. The valence orbitals, however, Vacuum Level

V Valence Band
overlap significantly with those of neighbouring atoms
giving rise to bands of spatially delocalised energy
levels. The energy-level diagram for the solid closely LII III 2p
resembles that of the corresponding isolated atom, LI 2s
except for the levels closest to the vacuum level.
It is to be borne in mind that for gases, the refer-
ence level of zero binding energy is the zero energy
K 1s
of an electron at rest in a vacuum and the BE in gases
Fig. 6.3 Schematic diagram showing the electronic structure of
is designated as BEV while for solids, the Fermi level, atoms in solids and the nomenclature associated with it.
which is the energy associated with the highest filled
orbital/energy level at absolute zero, is taken as the reference level
Vacuum Level
of zero binding energy and the BE is designated as BE f, Both the
scales have been shown in Fig. 6.4. Valence Band
The Fermi level in the conduction band is named after the famous
Italian physicist Enrico Fermi who got the 1938 Nobel prize for
demonstration of the existence of new radioactive elements produced Fermi Level
by neutron irradiation and for related discovery of nuclear reactions Conduction Band
brought about by slow neutrons.
Figure 6.4 is drawn assuming the sample to be metallic. The BEv (K) BEf (K)
results also hold true if the sample is an insulator or semiconduc- L
tor. Further, it is to be noted that the energy difference between the
Fermi level and the vaccum level just outside the surface in the case
of both semiconductors and metals is a measure of the quantity je. K Metal
Thus, if je of a reference solid is known, the je of another solid Fig. 6.4 Schematic diagram showing the reference
surface can be determined from the contact potential difference. For levels of zero-point energy in gases and solids.
metals, the photoelectric threshold is equal to je. This is not true in
general for semiconductors. For semiconductors, the photoelectric threshold is equal to the difference in energy
between the vacuum level and the top of the valence band (i.e. BE / ionisation potential) when the bands are flat to
the surface over the escape depth of emitted electrons and when the surface-state emission is negligible.
The maximum kinetic energy, the Fermi energy of any electron is
2/3
h2 ⎛ 3N ⎞
Ef = ⎜ ⎟ (6.3)
8me ⎝ πV ⎠
where N is the total number of electrons, me (9.11 × 10−31 kg) is the electron rest mass and V is the volume. The
volume of one kilomole of metal
Molecular mass of metal(kg/kmole)

Vm = (6.4)
Densityy of metal(kg/m3 )
Thus, the volume of each atom of metal (in m3/atom),
Vm (m3 /k mole) V
Vatom = = m m3 /atom (6.5)
N A (atoms/k mole) N A
1
On the assumption that there is one free electron per atom, the number of electrons per unit volume N L =
(electrons/m3). This number is called the Loschmidt number. Equation, (6.3) becomes Vatom
2/3 2/3 2/3

h2 ⎛ 3 N A ⎞ h2 ⎛ 3 1 ⎞ h2 ⎛ 3N L ⎞
Ef = = = ⎜ ⎟ (6.6)
8me ⎜⎝ π Vm ⎟⎠ 8me ⎜⎝ π Vatom ⎟⎠ 8me ⎝ π ⎠
2
⎡6.625 × 10 −34 (J s) ⎤ ⎡ 3 × 7 × 6.023 × 1026 (atoms)
2/3
s /kmole ⎤
or Ef = ⎣ −31
⎦
⎢ ⎥
8 × 9.11 × 10 (kg) ⎢⎣ 22 × Vm (m3 ) / kmole ⎥⎦
2/3
−38
⎡ 5.749 × 1026 ⎤
E f = 6.022 × 10 ⎢ ⎥ J
⎢⎣ Vm ⎥⎦
2/3
−20 ⎡ 1 ⎤
E f = 4.163 × 10 ⎢ ⎥ J (6.7)
⎣ Vm ⎦
Here, Vm is the volume in m3/k mole, i.e. the volume of one kilomole of metal.
23 kg/k mole
For sodium metal, Vm = 3
= 0.0237 m3 /kmole
970 kg/m
2/3
Ef (Na metal) = 4.163 × 10 −20 ⎡⎢
1 ⎤
Hence, J
⎣ 0.0237 ⎥⎦
= 4.163 × 10 −20 × 1.211 × 101 J

= 5.04 × 10 −19 J = 3 1 eV
The basic principle of UV-PES and XPS is the same except that in the former, the outer-shell electrons
are involved whereas in the latter, the inner-shell electrons are involved in photo ionisation of neutral atom/
molecule.
When a monochromatic beam of photons of short wavelength strikes a neutral atom/molecule, say A, it pro-
motes an electron from the ground-energy state, i.e. low-energy orbital (say Ψ0 with energy Eo ) to a vacant lesser
potential-energy state, i.e. high-energy orbital (say Ψj with energy Ej). This process of ‘jump’ of an electron from
lower to higher energy state under the influence of radiation of short wavelength is identical with an electron
transition and is represented as
A (Ψ0) + hv → A (Ψj) [Electron transition] (6.8)
When the incident photon imparts sufficient energy to the electron, the latter escapes from atom/molecule leaving
behind a cation. The process of eviction of an electron from the chemical species under the influence of a photon
of high energy occurs in a single step and is termed as direct photoionisation. It is represented as
A (ψ 0 ) + hv → A+* (ψ +j ) + e − (ESCA electron). (6.9)
Here, A+* is a cation in a highly excited state. Furthermore, all of the absorbed energy which is in excess of the
minimum needed to expel the electron from the neutral atom/molecule manifests itself as the discrete kinetic
energy (Ekin) of the photoelectron.
The law of conservation of energy requires
= ( E +j E ) + Ekin (6.10)
This relationship was first proposed by Albert Einstein in 1905 and is known as photelectric law. Here, E +j
is the energy of the highly excited quantum state ψ +j of A+*. The difference ( +j 0 ) is termed as the electron-
binding energy BE or the ionisation energy IEj (or ionisation potential IPj ). It is defined as the energy difference
between the ground state of the neutral atom/molecule and the final state of the cation. In general,
j = (E N
j
1
E0N ) (6.11)
−1
where E0N and E N j are the total energies of the N electron initial state and the N −l, hole final state of the system.
In a one-electron picture, each BEj can be associated with an ion having a hole in the ith orbital, including the
−1
possibility of multiplet splittings. The change in the energy of the final state E N j due to the relaxation of N −l
possible electrons towards the j hole is called relaxation energy. The relaxation energy or shift of BE with respect
th
to the orbital energy is not a constant or a fixed fraction of the BE. For a given photon energy, various final states
of the cation (A+*) will be reached corresponding to the different binding energies.
The photoelectrons with discrete kinetic energies liberated from the neutral atom/molecule by the direct
photoionisation process, also called the primary process, using x-ray photons are called ESCA electrons.
The mechanism of removal of 1s ESCA electrons from neutral atom/molecule is shown in Fig. 6.5 (a). The kinetic
energy of an ESCA electron is expressed as
Ekin = hv − BE(ls) = hv − BE (K) (6.12)
In addition to the direct photoionisation process, the other primary process of ionisation is electron ionisation.
Electron ionisation of an atom/molecule produces both an excited ion and a second electron.
However, because of electron−electron interactions, discrete electron energies are not observed. The process of
electron ionisation is represented as
−
A+ 0 → A+* + 2e− [Electron ionisation, not discrete energy] (6.13)
The electron ionisation can produce only Auger electrons but not ESCA electrons.
Further, the ion A+* in the highly excited quantum state ψ +j may relax to some stable excited state of A by the
following two secondary processes.
Firstly, the excited state A+* cation may relax to an excited state of cation A+ by x-ray fluorescence with the
emission of x-ray photon with energy h ν ′ such that h ν ′ < hv. This process is represented as
A+* → A+ + hv′ (x-ray) [x-ray fluorescence] (6.14)
The mechanism of x-ray fluorescence is shown in Fig. 6.5 (b). An electron drops from 2p electron to fill the 1s
hole, and an x-ray photon is emitted. This result is Ka x-ray emission. The energy of the emitted x-ray photon is
expressed as
hν ′ − BE (ls) − BE (2p) = BE(K) − BE (LII) (6.15)
The second secondary process competing simultaneously with the x-ray fluorescence is the Auger electron emis-
sion. In this process, the highly excited cation A+* relaxes to the excited state of doubly ionisied atom A++ with the
emission of a second electron having discrete energy. The electron emitted is called the Auger electron, and the
process of electron emission is represented as
A+*→ A++ + e− [Auger-electron emission] (6.16)
The latter process is more probable than the former for elements of atomic numbers less than about 32. Unlike
photoelectrons, the energy of the Auger electrons is independent of the energy of the incident photons. The pro-
cess of Auger-electron emission is shown in Fig. 6.5 (c). A 2s electron drops to fill the ls hole, simultaneously
liberating a 2p electron. This is known as KLILIII Auger process. The energy of the Auger electrons is independent
of the initial core-hole formation. The kinetic energy of an Auger electron is given by
Ekin = BE (ls) − BE (2s) − BE (2p)
= BE(K) − BE(LI) − BE(LIII ) (6.17)
An Auger transition is thus characterised primarily by (i) the location of the initial hole, (ii) the location of the
final two holes although the existence of different electronic states (terms) of the final doubly ionised atom may
lead to a fine structure in the high-resolution spectrum. Note that when describing the transition, the initial hole
e−
hn ′ e−
n
oto
ph
ay
LIII(2p 3/2)
x-r
2p
LII(2p1/2)
LI(2s1/2)
2s
hn ″
x-ray photon n
tio e− e−
p
hn or
hn abs
y
ra K(2s1/2)
x-
(a) (b) (c)

Fig. 6.5 Schematic diagram illustrating the mechanism of (a) liberation of an ESCA electron and the x-ray
absorption; Ekin = hν − BE (K); hν ″= hν − EKin (b) x-ray fluorescence; hv’ = BE(K) − BE(LII) (c) production of an
Auger electron; Ekin = BE (K) − BE (LI) − BE(LIII).
location is given first, followed by locations of the final two holes in order of decreasing binding energy, i.e. the
transition illustrated is KLI LII III when the LII and LIII shells are degenerate. The final state is a doubly ionised atom
with core holes in the LI and LII III, shells. The inner shell vacancies created in either of the two secondary processes
are filled within 10−17 to 10−12 second by electrons from outer shells.
For low-energy processes (1000 eV), the Auger effect will dominate while for higher energy processes
(10,000 eV) x-ray emission will dominate. x-ray ionising sources are used for ESCA while electron guns are ued
for Auger Electron Spectroscopy (AES). The basis of AES is an effect discovered in 1932 by Auger in which
electrons are liberated in an atom with energies characteristic for the emitting atom.
The general expression for the kinetic energy of an Auger electron in an atom is given by
kin
Eijk B i ( Z ) − BE j ( Z ) BE
BE B k (Z + Z ) (6.18)
where BEi, BEj and BEk are the binding energies of the electron in the shells (K, L, M…) involved. Since the
Auger electron is emitted from a singly charged ion, its binding energy is larger than that of the same electron in
the neutral atom, i.e. Z (0.5 ≤ ΔZ ≤ 1). The number of possible lines grows rapidly as the number of electrons
increases, i.e. with Z. The number of possible Auger lines for a given combination of shellsdepends upon the type
of coupling between angular momentum and spin of electrons (LS, jj or intermediate coupling). For instance, the
KLL spectra of very light atoms (LS coupling) consists of five lines, those of very heavy atoms ( jj coupling),
six lines while those of intermediate atoms should have ten lines. However, even in light atoms, the number of
observed lines is more than the number of lines expected on the basis of RS coupling. The weaker additional lines
are due to Auger emission from a doubly ionised atom which produces a triply ionised atom, i.e. A++* → A+++ +
e−. Such transitions involving initial double ionisation is quite common when the primary radiations are highly
energetic. When the Auger electron excitation is carried out by ions, even higher degree of ionisation occurs, i.e
A+++* → A++++ + e−, e.g. Ne KLL spectra excited by He+, H+ or O+5 beams.
Transitons in which an intitial vacancy is filled by an electron from the same shell, e.g. an LI vacancy when
filled by an electron from the LII shell are called Coster–Kronig transitions. These transitions contribute to the
production of Auger electrons and thus to the complexity of their spectrum. BEs’ of levels j and k being small
compared to BEi may be neglected. Consequently,
Eikin ∝ BEi
Thus, the dependence of the Auger energies upon the atomic number can be obtained from the plot of energies of
the initial vacancy against Z. It is found that inner shells can be used if the energy analyser has an upper energy
limit of 1500 eV: the K shell up to Al or the L shell up to beyond Ni. Higher Auger-electron energies are not prac-
tical because of intensity reasons.
The intensity of an Auger line depends on the probability for creation of an inner shell vacancy, i.e. the ioni-
sation cross-section Qi and the probability aijk, that is the filling of this vacancy is accompanied by an Auger or
Coster–Kronig transition.
kin
I ( Eijk ) ≈ Qi aijk (6.19)
In case of ionisation by electrons,
a ln U ⎡ 2 ⎤
Qi = Å (6.20)
BEi2 U ⎣ ⎦
E (energy
y of incident electron )
where BEi is in eV
V, U = , a = 960
BEi
The relation holds good for many ionisation processes with ‘a’ ranging from 600–1100. The function has a
maximum at Emax ≈ 2.72 BEi.
Qimax Ei2 ⎡ Å 2 ⎤
a / BE (6.21)
⎣ ⎦
This is lower than the observed value. This leads to a maximum attenuation coefficients
μimax = Qimax ≤ 1 × 10 −2 Å −1 for all materials and ionisation energies BEi ≥ 50 eV. This reveals that the attenua-
tion due inner-shell ionisation is small compared to attenuation by valence electrons. Equations (6.20) and (6.21)
lead to the following values for AES with electrons:
(i) The energy of the incident beam should be at least 2.5 BEi for efficient ionisation.
(ii) BEi should be smaller than about 1500 eV if an ionisation cross section exceeding 10−20 cm2 is to be obtained.
Further, it has been proved for transitions involving K and L shells that nearly all vacancies are filled by Auger
transitions (ai = ∑aijk ≈ 1) provided BEi >1500 eV. At higher BEi values, x-ray emission becomes increasingly
stronger. Thus, AES is efficient below about 1500 eV.
The Auger-electron line width of free atoms is determined from the lifespan of the vacancy i. It ranges from frac-
tions of an eV for small BEi to about 10 eV for high BEi values ( ≈1500 eV). This also leads to infer that shells with low
ionisation energies are suitable for AES. When atoms are in condensed matter, i.e. atoms are not free, the Auger elec-
tron bandwidth increases considerably. This is because of broadening of the valence levels into bands. The removal
of the Auger electron causes a disturbance in the atom from which it is removed. The main disturbances are dynamic
relaxation and static relaxation which is further divided into two types, namely atomic and extra-atomic relaxations. In
addition, the disturbance is also caused by the multiplet coupling in the final state, which causes the splitting. Dynamic
relaxation describes the rearrangement of an atom accompanying ionisation. The dynamic relaxation energies of elec-
trons are automatically included in Eq. (6.18) for Δ Z = 0 if experimental BEs’ are used. Static relaxation describes the
rearrangement of the atoms connected with the transition of the j electron into the i hole. Static atomic relaxation is
the process which occurs in the free atom, static extra-atomic relaxation represents the response of the atomic envi-
ronment in the solid to the formation of the j hole. The contribution of the various processes to the Auger energy have
been determined for the Cu LII III MIVV MIVV− Auger spectrum to be 25 eV, 10 eV and 20 eV for the multiplet, the static
atomic and the static extra-atomic relaxation contributions respectively. The Auger energies calculated with Eq. (6.18)
for ΔZ = 0 but with the inclusion of these terms agree within 2–3 eV with experiment.
In summary, AES is based upon the measurement of the kinetic energies of the emitted electrons. Each element
in a sample being studied will give rise to a characteristic spectrum of peaks at various kineitic energies. It is a
surface-sensitive spectroscopic technique used for elemental analysis. It offers high sensitivity (typically, Ca of
1 per cent mono layer) for all elements except H and He, a means of monitoring surface cleanliness of samples, and
quantitative compositional analysis of the surface region of specimens, by comparison with standard samples of
known composition. Moreover, Auger depth profiling provides quantitative compositional information as a func-
tion of depth below the surface. Scanning Auger Microscopy (SAM) provides spatially resolved compositional
information of heterogenous samples.
6.5 PHOTOELECTRON SPECTRUM

The kinetic-energy distribution of the emitted electrons, i.e, the number of emitted electrons as a function of their
kinetic energies, is measured with the aid of an instrument called a photoelectron spectrometer and the plot of
intensity of electrons, i.e. the number of electrons per unit time at each energy level, against their kinetic energy
is termed the photoelectron spectrum. A peak is found in the spectrum at each energy hv − IEj’ corresponding to
the binding energy IEj of an electron in the atom; the BEs of energy levels in solids are conventionally measured
with respect to the Fermi level rather than the vacuum level.
When vibrational and rotational motions are also excited on ionisation, the energies of the photoelectrons are
defined by
Ekin hv IE
I j Ev*ib,rot (6.22)
The spectrum may contain many vibrational lines for each type of electron ionised and the system of lines cor-
responding to ionisation from a single molecular orbital constitutes a band. In addition to Koopman’s theorem,
the two simple rules that correlates the PES spectrum with molecular electronic structure are (i) each band in
the spectrum corresponds to ionisation from a single molecular orbital, and (ii) each occupied molecular orbital
of BE less than hv gives rise to a single band in the spectrum. However, the number of bands in the PES is more
than expected from valence orbitals in the molecule. The additional bands in the PES may arise due to (i) the
ionisation of one electron with the simultaneous excitation of a second electron to an unoccupied excited orbital.
The bands generated in such a two-electron process are much weaker compared to simple ionisation bands,
(ii) ionisation from a degenerate occupied molecular orbital since the degenerate orbitals in the molecule may
not be degenerate in the molecular ion because of spin-orbit coupling and Jahn–Teller effect, and (iii) ionisation
from molecules with unpaired electrons such as O2, NO. In such cases, many more bands appear than expected
on the basis of occupied orbitals in the molecule and neither the Koopman theorem nor the simple two rules are
valid.
The relationship between ionisation energies and orbital energies (i.e. Koopman’s theorem) has been illustrated
in the schematic diagram shown in Fig. 6.6.
Ionisation Energy (eV)

1
X2ex (M+) 2
3
IE3
1
X1ex (M+) 2
Intersity
3
IE2 1 (counts/sec)
Xg (M+) 2
3
IE1 IE1

1
Xg (M) 2 E1
Intensity 3
IE2
(counts/sec) E2
IE3
E3
Fig. 6.6 Schematic diagram illustrating the relationship between ionisation energies and the orbital energies (i.e. Koopman’s theorem). Xg —
I
the ground state, X ex —the excited state, M— the neutral molecule and M+— the molecular ion.The dark point (.) represents an electron.
The spectra in PES are recorded and displayed in two ways:
(a) Differential Spectrum When electrons of one energy only at time are able to reach the detector, the spec-
trum obtained is called the differential spectrum.
(b) Integral Spectrum When all the electrons of more than a certain energy can reach the detector simultane-
ously, the spectrum is called an integral spectrum.
Note that AES spectra are different from ESCA spectra. AES spectra are plotted as derivatives, i.e. dI
dEkin
against kinetic energy (Ekin) while ESCA spectra are plotted as integrals. This is so because Auger spectra caused
by ionisation lie on top of a very high, scattered electron back-
IPII
ground and differentiation removes the spectra background.
Differentiation of ESCA spectra is not necessary because the
primary photobeam produces a lower electron background.
Intensity (counts/sec)
The intensity ratio in differential spectra is given by the areas

whereas in integral spectrum by the step heights. The hypo-
thetical differential and integral spectra of some systems are
IPIII
shown in Fig. 6.7 (a) and (b) respectively. Differential and
derivative AES spectra of Pd metal generated using 2.5 keV IPI
electron beam to produce the initial core vacancies and hence
to stimulate the Auger emission process have been shown in
Fig. 6.8 (a) and (b) respectively. It is to be noted that in PES, Ionization energy (eV)
differential and derivative spectra are not same but different. (a)
So be careful while using these terms.
The other phenomenon related to the interaction of x-ray
radiation with matter is the absorption of x-ray photons, i.e.
hv″ = hv − Ekin [Fig. 6.5 (a)]. The intensity of the x-ray beam
decreases as it passes through the materials and it depends IPIII

upon the amount of the material. By measuring the absorp-
IPII
tion as a function of wavelength of x-rays, the absorption
IPI
spectrum is obtained. The wavelength in the spectrum where
a sudden change of absorption occurs is called the absorption
edge. Absorption edges occur at those wavelengths where the
x-ray photon has sufficient energy to excite an electron from Ionization energy (eV)
a particular energy level to infinity.
(b)
For a metal, infinity is taken to be at above the Fermi level Fig. 6.7 Hypothetical (a) differential, and (b) integral
and at the bottom of the conduction band for an insulator. photoelectron spectra of some system. Ionisation potentials
The positions of the absorption edges give us information (IP): IPIII > IPII > IPI .
2E + 05
Intensity, I(counts/sec)
K LI LII LIII
(a)
Binding Energy
LIII
LII
LI (b)
Kedge
0.0
65.0 Kinetic Energy (eV) 565.0
(a)
Intensity
Wavelength
(c)
Kα1 Kα2
dI/dEkin
Wavelength
(d)
KLM KLL
Emission
65.0 Kinetic Energy (eV) 565.0
(b) Kinetic Energy

Fig. 6.8 Auger electron spectra of Pd metal pro- Fig. 6.9 Schematic diagram showing structural features of (a) photo-
duced by using 2.5 keV electron beam: (a) differential electron spectrum, (b) x-ray absorption spectrum, (c) x-ray fluorescence
spectrum (b) derivative spectrum. spectrum, and (d) Auger electron spectrum.
regarding the electronic energy structure of an atom. Each element has its own characteristics of K, L, M, N spec-
tra. The structural features of x-ray absorption, x-ray fluorescence−, Auger electron−, and x-ray photoelectron
spectroscopy are shown in Fig. 6.9. The positions of the photoelectron peaks coincide with the x-ray absorption
edges. The peak positions in the other spectra are only relatively calculated with respect to one another.
6.6 BANDWIDTH AND THE FACTORS AFFECTING IT

The photoionised excited state will have lifespan, say τ . According to the Heisenberg uncertainty principle, i.e.
h
ΔE Δt ≈ , the uncertainty in the energy of the ionised excited state will manifest itself in the width, say Γ,
2π
of the photoelectron signal.
h 41.346 × 10 −16 (eVs )
Thus, Γτ ≈ = = 6.577 × 10 −10 (eV s)
2π 2π
h 6.577 × 10 −16 (eV s)

or Γ= = (6.23)
2πτ τ
For t = 10−13 s
6.577 × 10 −16 (eV s)
Γ= −13
= 6.577 × 10 −3 eV = 0.0065 eV = 6.5 meV
10 s
i.e. the natural bandwidth of the photoelectron signal is 0.0065 eV = 6.5 meV.
The other factors which contribute to natural bandwidth are Doppler effect, width of the irradiating line, spec-
trometer broadening, etc.
The Doppler effect aries due to the random thermal motions of the target molecules before ionisation and
the motion of the ejected electrons. The target molecules in random thermal motions have components of their
velocities in the direction of the ejected electrons and add to their velocities. The electrons that move in a single
direction only, the one-dimensional law of thermal distribution of velocity of molecules holds good. In such a
situation, the observable line width, ΔV, in millielectron volts is then given by
1
⎛ VT ⎞ 2
ΔV = 0 75 ⎜ ⎟ (6.24)
⎝M⎠
where V is the electron energy in electron volts, T is the absolute temperature and M is the mass in atomic mass
units.
For M = 100 amu; V = 10 eV and T =298 K
1
⎛ 10 × 298 ⎞ 2
ΔV = 0.75 ⎜ = 4.09 meV; which is well below the normal resolution of the differential analysers.
⎝ 100 ⎟⎠
The Doppler effect becomes important for light molecules and high electron energies or gas temperatures.
The extreme case is hydrogen ionisation of H2, with He (I) 21.22 eV and He (II) 40.8 eV radiations sources
respectively. The observable line width in the former is 22 meV and 46 meV in the latter. The Doppler widths can
easily be reduced by reducing the effective transverse kinetic temperatures used in Eq. (6.24) to below 10 K. This
temperature can be achieved if the target gas is used in the form of a jet or a true molecular beam.
The broadening of width by the irradiating line as well as spectrometer broadening may be judiciously con-
trolled to a certain extent.
6.7 RESOLUTION OF PHOTOELECTRON SPECTRAL PEAK

The ability of a technique to determine the location of two closely spaced peaks/bands in the presence of each
other is called resolution. It is governed by the width of the spectral band relative to the spacing between the
peaks which is influenced not only by Doppler effect but also by the line width of the exciting radiation. There-
fore, for high-resolution work, the Doppler widths must be reduced. In PES, this is achieved using target gas in
the form of a jet or a true molecular beam in which the transverse kinetic temperature is reduced to below 10 K
(see Section 6.6). As stated above, the width of the PES peak is limited by the line width of the excitation radia-
tion which increases as the third power of the frequency. According to this rule, the ratio UPS line width to that
3
⎡ 21.22(eV) ⎤ −6
of the XPS should be of the order of ⎢ ⎥ ≈ 10 . This means, that the UPS line width is very small
⎣ 1486 .5( eV ) ⎦
as compared to that of the XPS. Consequently UPS is capable of much better peak resolution than XPS (0.01 eV
against 0.5 eV in XPS). Due to better resolution of UPS, the fine structure resulting from the different vibrational
and rotational transitions are also observed in the UPS; but not all of these peaks are resolved. However from the
fine structure in the high-resolution UPS, some deductions regarding the nature of molecular orbital involved in
photoionisation have been made and are so summarised in the tabular form in Section 6.14.4.
The additional complexity in UPS arises due to spin-orbit coupling for heavy atoms and by Jahn–Teller
distortion effects in highly symmetric molecules. Even analysis of UPS spectra of atoms and simple molecules
is very hard.
6.8 INTENSITIES OF THE PHOTOELECTRON SPECTRAL PEAKS

The intensities of the ionisation bands help to analyse the PES of molecules whose ionic states have not been
identified. The areas of the photoelectron bands are approximately proportional to the relative probability of
ionisations, also called relative partial ionisation cross sections or branching ratios in ionisation, to the different
ionic states. For resolved bands, the total intensity of a particular band is equal to the sum of the intensities of all
the vibrational lines in that band.
6.8.1 Rules to Predict the Intensities of Ionisation Peaks

The relative intensities of ionisation bands in the photoelectron spectrum not only depend upon ionisation prob-
ability but also upon the nature of the orbital, and their size, number of nodes and their localisation in the mol-
ecule, etc. Experimental parameters also contribute to the measured intensity. However, the three approximate
rules framed to predict the intensities of the peaks in the PES are based purely on the probability of transition and
the parameters governing it.
(a) Rule 1 The relative probability of ionisation from a given orbital is proportional to the number of equiva-
lent ionisable electrons. Thus, the intensity of the band because of two equivalent ionisable electrons will be one
half of the intensity of the band due to four equivalent ionisable electrons. All orbitals having the same number
of ionisable electrons will give rise to bands of equal intensities. This rule is valid when ionisation occurs from
molecular orbitals consisting of the same set of atomic orbitals; however even otherwise also, this is the only rule
useful for analysing the spectra of closed-shell molecules, e.g. the partial photoelectron p6
spectrum of biphenyl in the p-electron region, i.e. 8–12 eV exhibits first three bands at 8.4,
9.4 and 9.8 eV in the approximate intensity ratio of 1 : 2 : 1. The p-orbital pattern in biphe- p4 p5
nyl predicted from the Huckel Molecular Orbital (HMO) calculations is indicated below:
The degenerate p4 and p5 orbitals yield a degenerate state which has four electrons. p3
Thus, the experimental intensity ratio of the three p-electron ionisation bands conforms
to the p-orbital pattern of biphenyl. The PES of CCl4 in the 8–18 eV region exhibits four p2
bands at ~11.7, ~12.7; ~13.5 and ~16.7 eV. The intensity ratios of the first three bands are
1.9:2.6:1.0, which indicate that the third ionisation band is because of e–l ionisation. MO p1
calculations reveal that the nonbonding chlorine p-atomic orbitals generate the outermost
occupied orbitals in CC14. In tetrahedral symmetry, these orbitals combine to give the three molecular orbitals t1,
t2 and e; and t1 and t2 being triply degenerate while e is doubly degenerate. Thus, the predicted pattern of inten-
sity ratios of the ionised bands will be 3:3:2. The predicted ratios, though do not conform to the experimental
ratios of the first three bands, but on comparison show that the first three ionisation bands correspond to the
t1−1 , t2−1 and e2−1 ionisations respectively.
(b) Rule 2 The probability of ionisation is proportional to the statistical weights of the ionic states produced.
This rule is equivalent to Rule 1 when ionisation occurs from closed-shell molecules but is different when ioni-
sation takes place from atoms and open-shell molecules, for example the 2P3/2 and 2P1/2 states of atomic ions of
rare gases though originating from the same orbital, have different statistical weights (i.e. 2J + 1) of 4 and 2
respectively. Further, oxygen on ionisation, produces O +2 ion in quartet (4∑) and doublet (2P) states. Thus, the
intensity ratios of the bands corresponding to the quartet and doublet states will be 2:1. Note that both rules 1 and
2 hold good for ionisations from closed shells of an open-shell molecule such as oxygen. Furthermore, when
ionisation occurs from a filled orbital, the total intensity of the ionised band equals the sum of the intensities of
such bands arising from the electronic configuration of the ionised molecule, i.e. ion and the relative intensities
of the bands corresponding to single-electron configuration of the ion is proportional to the statistical weights
(i.e. 2J + 1) of the different ionic states, e.g. the electronic configuration of nitrogen oxide 2p...4s2 5s2 1p4 2p on
1p −1 ionisation yields a configuration of 4s2 5s2 1p3 2p for NO+ molecule. This configuration gives rise to six
+ 3
ionic states: ∑ 1
3
5 2
3
∑ 1− , 1 ∑ −0 1 1 2 1∑ +0 Accordingly, 1p −1 ionisation of NO molecule should
give six subbands of intensity ratios 3:6:3:1:2:1 in the PES spectrum of NO. Similarly, the ionisation from the
5s orbital gives 1∑0 and 3∑1 ionic states. Consequently, two subbands in the intensity ratios of 1:3 are expected in
the PES spectrum. Accordingly, the total intensity of the 1p −1 ionisation bands due to ∑ states = 8 and those of
5s −1 bands = 4. Thus according to Rule 1, the intensity ratio of the 1p −1 and 5s −1 ionisations should be 2:1 but
experimentally it has not been observed in the spectrum. The discrepancy may be attributed to the fact that
while framing the rules which are purely on probabilistic considerations, the character of the orbitals have not
been taken into account, e.g. the 6s−1 and 5d−1 ionisations bands from atomic mercury have intensity ratios of
1 : 21 whereas orbital occupancies predict a ratio of 1:5. In addition, as mentioned earlier, experimental param-
eters also contribute to the total intensity of the band measured by the instrument and will be discussed here in
Section 6.8.2.
(c) Rule 3 This rule was given by Cox and Orchard who were also the first to enunciate the rules 1 and 2 as they
apply to open-shell molecules. The third rule is for ionisation from the open shell of an open-shell molecule. The
relative band intensities, are in general, proportion to the coefficients of the fractional parentage. When molecular
configuration generates only one term, Rule 3 is equivalent to Rule 2, and this is so if the open shell contains either
a single electron or one electron less than the number required for a complete shell. When the molecules have
more than one open shell, a more complex rule which is beyond the scope of this book is available in the literature;
Cox and Orchard; Chem. Phys. Lett., 7 (1970) 273; Cox, Evans and Orchard, Chem Phys. Lett., 13 (1972) 386;
Brink and Satcher, Angular Momentum, Oxford University Press, London (1968).
6.8.2 Measured Intensities of Photoelectron Spectral Peaks

The presence of peaks at particular energies indicates the presence of a specific element in the sample under
study—the measured intensity of a peak is related to the concentration of the element within the sampled region.
Thus, photoelectron spectroscopy provides a quantitative analysis of surface composition, etc., the accuracy of
which will be governed by the accuracy of measured intensities.
The intensity of the PES signal is measured in terms of the number of ejected electrons per second at a particular
energy level. In differential PES spectra, the intensities of the peaks depend upon the resolution of the analyser.
As resolution increases, the height of the sharp peak apparently increases while the height of the broad peak
decreases. Therefore, areas under the peaks, rather than heights, are taken as the true measure of the intensities of
the PES signals. On the other hand, this problem is not encountered in integral spectra. The measured step heights
in the integral spectra correspond exactly to the areas in differential spectra.The angular distribution of electrons
being not even will vary from one band to another in the PES but will affect the intensity. When measurements
are made at an angle of 54° 44′/cos−1 ( 1/ 3 ), called magic angle from the photon beams (for unpolarised light),
the variations in the angular distribution have no effect on the intensities. When electrons normal to the photon
beam are examined, the intensity due to angular effects alone varies by a factor of two.
The photoelectron angular distributions of a mixture of s, p, d, f or higher partial electron angular wave functions
have been given by
σ ⎡ β ⎤
I( ) = 1 + (3cos 2θ 1) ⎥ (6.25)
4π ⎣ 2 ⎦
Here, q is the observation angle away from the direction of the electric vector of the plane polarised radiation, s
is the total cross section integrated over all angles and b is an anisotropy parameter which determines the angular
distribution of the photoelectrons in the ionisation process. For a pure p-wave, b = + 2.
For unpolarised radiation (generally used in PES), the distribution of intensity is defined as
σ ⎡ β⎛3 2 ⎞⎤
I ( ′) =
4π ⎢1 + 2 ⎜⎝ 2 sin θ ′ − 1⎟⎠ ⎥ (6.26)
⎣ ⎦
Here, θ ′ is the observation angle away from the direction of the radiation beam. The range of b values is from −1
and +2 and has the usual meanings. The value of b determines the angular distributions in the ionisation of both
the atoms and molecules.
Equations (6.25) and (6.26) are valid for all atoms, for diatomic molecules, and for all molecules. The value of
b can be determined experimentally using the relationship
3β
I ( ′)) = 1 + sin
si n 2θ ′ (6.27)
4 2β
The value of b can be determined by measuring the intensity at two angles or by making a least-square fit to the
theoretical distribution given by Eq. (6.27).
Relationships between b and theoretical parameters that govern the ionisation process, i.e. angular momentum
of the electron in the atom (R − S coupling holds good), partial cross section for the production of l − 1 and l + 1
waves and the phase difference (dl + 1 − dl − 1) between the two waves exist for atoms and diatomics. The formula
for atomic ionisation is expressed as
l (l − )σ 2l − + (ll + l+ ) 2
l +1 − 6l (l + )σ l+
l + σ l −1 cos(δ l + − (δ l − 1)
= (6.28)
(2 + 1)[ 2
l− + (l + 1
1)) 2
+1 ]
l +1
b depends not only on the strengths of the two partial waves but on their phases also, which govern the inter-
ference between them. When the interference term is large, b values become negative since both the partial cross
sections and the phases depend upon the energy of the electron, so b values of atoms and molecules also depend
upon the electron energy and are usefu1 in the analysis of photoelectron spectra. Direct calculation of b values
have been carried out on some atoms and molecules but, in general, are hard due to mathematical difficulties.
Experimental and theoretical studies reveal that the b values depend almost entirely on the nature of the orbital
from which the electrons originate and not on the details of the ionic states produced. b is constant within the vibra-
tional structure of a single band. If two or more bands arise from ionisation out of a single orbital (e.g. because
of spin-orbit coupling or Jahn–Teller effects), the b values for such bands are the same. This relationship also
probably applies to several bands per orbital produced by ionisation of open-shell molecules provided the electron
energies are not very different.
To summarise, it is concluded that the measurements on the angular distribution of electrons photoemitted
from atoms or molecules provide a tool to determine which molecular orbital is associated with a given photo-
ionisation peak in the spectrum of a molecule or an atom. The variation in the angular dependence of one orbital
compared to another is never very dramatic in the gas phase because the atoms or molecules in the gas phase have
random orientation. On the other hand, a surface can orient all the adsorbed atoms or molecules so that the angular
distribution of the emitted electrons has much more structure than the corresponding gas-phase spectra. A proper
analysis of the angular dependence can be used to determine the symmetry of the bonding site and potentially the
symmetry of each orbital.
Magic Angle in PES As already described, the angle q = 54° 44′ is called the magic angle since at this angle,
the intensity is completely independent of b whether the radiation is polarised or not. When the photoelectron
spectrum is recorded at this angle, the relative intensities of the peaks in the spectrum are not disturbed by the
angular distributions of the emitted electrons.
6.9 SENSITIVITY AND DETECTION LIMIT OF PHOTOELECTRON

SPECTROSCOPY
Sensitivity and detection limits are important for the quantitative application of any spectroscopic technique.
The intensity of the signal can be increased using a higher source intensity provided increased source intensity
does not cause decomposition of the sample. However, an increase in signal by using a higher source intensity is
accompanied by an increase in background called noise. Actually, the quantity signal-to-noise ratio determines
how much the quantitative information can be obtained from the spectrum.
A typical ESCA spectrum with superimposed background noise is shown in Fig. 6.10.
IP (Peak intensity)
Noise ∼ lB
IB (Background intensity)
Kinetic energy (eV)

Fig. 6.10 Schematic diagram of a typical ESCA spectrum.
The difference between the peak and background intensities is defined as the signal intensity of the peak.
Thus,
Signal intensity = IP − IB (6.29)
IP − IB
and signal-to-background ratio = (6.30)
IB
The signal-to-noise ratio is defined as (IP − IB,) / RMS noise and is a more important parameter than signal-
to-background ratio. For most spectra, as IP → IB, RMS noise ≈ I B ; consequently,
Signal-to-noise ratio = IP − IB/ I B (6.31)
The other important factor which is important in any spectroscopic technique is the behaviour of observed
intensity with concentration. For most surface techniques, a linear relationship exists between net intensity and
concentration, i.e. Ip− IB = kC. Thus, the sensitivity of a spectroscopic technique is defined as the slope of a plot
of intensity against concentrations as shown in Fig. 6.11.
Mathmatically,
dI Δ ( I P I B )
Sensitivity = = (6.32)
dC ΔC
Further, detection limit is defined as the minimum amount of a (i)

material that can be detected using a given spectral line. This rep- (ii)
resents the concentration for which

IP − IB = 2 × RMS noise or (IP − IB) = 2 I B (6.33)
The most sensitive spectral line need not have the lowest
detection limit. In Fig. 6.11, line (i) has higher sensitiv-
Intensity, l
ity (greater slope) and occurs at a higher background, i.e. IB lB (i)
(i) while line (ii) has lower sensitivity (smaller slope) and
occurs at lower background, i.e. IB. (ii). Thus, line (ii) will show
a lower detection limit although line (i) shows a higher sensitiv- lB (ii)
ity. Detection limits are important and generally the matter of
concern is what fraction of a monolayer can give a signal equal
C°(ii) C°(i)
to 2 (RMS noise)1/2. C
Fig. 6.11 Calibration curve for two spectral lines:
(i) greater slope and higher background (ii) lower
slope and lower background.
6.10 XPS CHEMICAL SHIFTS OR CORE
ELECTRON SHIFTS
The exact BE of an electron not only depends upon the level from which the electron originates but also upon
(i) the local chemical and physical environments of the atom which is being influenced by the molecular charge
distribution on the shielding of core electrons, and (ii) the formal oxidation state of the atom. Changes in either
chemical and physical environmenmts or oxidation state of the atom give rise to small shifts in the peak positions
in the spectrum which are termed chemical shifts or core electron-shifts. Such shifts are readily observable and
interpretable in XP spectra (unlike in Auger spectra) because the XP technique, being a one-electron process is
simple to interpret and is of high intrinsic resolution as core levels are discrete and generally of a well defined
energy. The chemical shifts need be standardised with respect
to the BE of some reference compound. The difference between
electron binding energy for a particular atom A within a given
CO
molecule and binding energy of the same electron in a reference
molecule is called the standard XPS chemical shift. The standard CH4
chemical shift is then defined as CO2
Δ IP (A) = IP (A) − IP (OA) (6.34)
where, IP (A) is the BE of the electron (say 1s) in atom A in

the molecule, and IP (OA) is the BE of the electron (say 1s) in
the atom A in the reference molecule, e.g. for carbon atom the
reference is CH4. The C, 1s PES spectra of CH4, CO and CO2 is
shown in Fig. 6.12. Thus ΔIP (C, 1s) = 5.30 eV for CO and ΔIP
(C, 1s) = 7.7 eV for CO2, the IP (C, 1s) of CH4 being 290.7 eV.
Further, XPS chemical shifts of same atoms in a molecule will
be different provided they are in a different chemical environ- 290 295
ment, e.g. the chemical shift of 1s carbon electrons for various Ionization energy (eV)
carbon atoms of ethylchloroformate (I) and trifluroacetate (II) Fig. 6.12 Carbon 1s photoelectron spoctrum from
follows the order. gas mixture of CH4, CO and CO2 .
O O
H H
Cl C C C H ΔIP(Cl C O) > ΔIP(O − CH2) > ΔIP(− CH3)
H H
(I) and in
F O
O
H H
F C C O C C H ΔIP(CF3) > ΔIP( C O) > ΔIP(O − CH2) > ΔIP(− CH2)
F H H
(II)
This reveals that XPS chemical shift is governed by the electron-withdrawing power of the atoms attached
to the atom whose chemical shift is under consideration. The more the withdrawing power of the neigh-
bouring atoms or substituents, the greater the chemical shift. The BEs (in eV) of 1s electron of some atoms
in compounds are listed here for ready reference. C1s (288), CH4 (290.7), C2H5OH (290.9) CH3COCH3
(291.2), CH3CHO (291.3), CH3COOH (291.4), CH3OH (292.3), Nls (403.0), C6H5NH2 (405.5), NH3 (405.6),
N2 (409.9), NO (410.7), NO2 (412.9), Ols (538.0), CH3CHO (537.6), C2H5OH (538.6), CH3OH (538.9),
CH3COCH3 (539), SO2 (539.6), H2O (539.7), CO2 (540.8), CO (542.1), O2 (543.1), Fls(694.0), SOF2 (693.6),
SF6(694.6), CF4(695.0).
As has already been stated, changes in oxidation state also cause XPS chemical shifts (i.e. change in BE),
e.g. chemical shifts in the K and L1 shells of chlorine in a series of inorganic compounds reveal that the chemical
shift increases with the increase in the oxidation number, since atoms of higher positive oxidation state exhibit a
higher BE due to the extra coulombic interaction between the photoemitted electron and the ion core, i.e.
NaCl (−1) < NaClO2 (+3) < NaC1O3 (+5) < NaClO4 (+7)
The same argument holds good for S, N, I, Be, Cu, Fe, B, Cr, Eu and C compounds, e.g. for S (2s): oxidation
state / chemical shift :: 0/0 , +4 /4.5; for N2 (ls): −1/0, +1/+4.5,+3 /+5.1; for Cl2(2p): −1 / 0, +3 /+3.8, +5 /+7.1, for
I2(4s): −1/0, +5 /+5.3.
Accordingly, the chemical shifts permit insight into the nature of bonding and the valency of the atom in the
molecule.
6.11 ATOMIC PHOTOELECTRON SPECTRA

Atoms have the simplest photoelectron spectra. The light atoms, i.e. up to alkaline earth metals, obey Russells–
Saunders (i.e. L–S) coupling where the orbital (l) and the spin (s) angular momenta of electrons are coupled
separately.
n
L l1 = l2 … ∑ li
i
n
S s1 = s2 + … ∑ si
i
For two electrons, S = s1 ± s2 = 0 or 1. Thus, the total angular momentrum is given by J = L + S and the atomic
term symbol is 2 S 1 LJ . Here, J can take on values L + S, L + S —1 ..., L S . L S coupling is opposite to j–j
coupling which is operative in heavy atoms. The strong coupling between the individual orbital and spin angular
momenta is deduced as follows:
j1 l1 + s1
j2 l2 + s2
n
j1 j2 + … = ∑ ji J
i
State L S Relative order

Ψ1,Ψ2 Same S1 > S2 E2 > E1
Ψ1,Ψ2 L1 > L2 Same E 2 > E1
The relative order of energies of the orbitals is predicted from the Hund’s rule of multiplicity, i.e.
For less than half-filled orbitals,
Ψ1,Ψ2 J1 > J 2 E1 >E2
For more than half-filled orbitals,
Ψ1,Ψ2 J1 > J 2 E2 >E1
For carbon atom 2p2: 3P0 < 3P1 < 3P2< 1D2 < 1S0
and for mercury atom: (6s6p) 1P1 >3P2 >3P1 >3P0 >(6s6s) 1S0
(a) 2p PES of Ne (Atomic Number 10) The electronic configuration of Ne is 1s2 2s2 2p6, which is a closed
shell configuration. Photoionisation of 2p electrons will yield Ne+ ion with a hole in the 2p orbital according to
the scheme
−1
Ne (Is2 2s1 2p6) lS0 ⎯2 p → Ne+ (1s2 2 s2 2p5) 2P3/2 and 2P1/2 + e −
The atomic states involved in this process are worked out from the general term symbol 2s+1LJ as follows.
For Ne(ls22s22 p6): l = 0, s = 0; L = 0, S = 0. ∴ J = 0 and 2S+1LJ = 1S0
1 1 3
For Ne+ (Is22s22 p5): l = 1, s = ; L = 1, S = 1/2 ∴J = 1+ = ;
2 2 2
1 1 1 1
l = 1, s ; L = 1, S j =1− =
2 2 2 2
Thus, the possible terms are 2 S 1 LJ 2

P3 / 2 and 2 P1/
1 2
+
Thus, the 2p PES of Ne will consist of two peaks corresponding to the 2P1/2 and 2P3/2 higher energy states
respectively. According to Hund’s multiplicity rule, the BE of 2p−1 electron in 2P3/2 state should be small relative
to that of in 2P1/2 i.e. BE (2P1/2) > BE (2P3/2). It can be noted that the relative intensities of these peaks are directly
proportional to the multiplicity of the final states, i.e. the number of 2J + 1 components, J, J − l, J −2, …−J, thus
indicating the type of electron ionised. In other words, we can say that the relative intensity of the peaks reflects
the degeneracy of final states, i.e. gJ = 2J + 1, which in turn determines the probability of transition to such a
state during photoionisation. Accordingly, the ratio of intensities of 2P1/2 and 2P3/2 peaks will be 2:4. The predicted
2p PES of Ne is consistent with the observed spectrum of Ne using He (II) 40.81 eV radiation source. The two
states are separated by 0.1 eV.
According to Koopman’s theorem, one peak is expected per orbital. This rule fails in case of p- and d-orbitals
in light as well as in heavy atoms where the number of peaks of an electron in these orbitals corresponds to
the number of states corresponding to the orbitals (i.e. characteristic spin-orbit doublets are observed in case
of p-, d-, f- orbitals since l = 2, 3 and 4 for the p-, d- and f- orbitals respectively. l = 0 for s-orbital so there is no
spin-orbit splitting). In such cases, a weighted mean of all energies for ionisation from a single orbital must be
taken. The splitting of the energy states corresponding to the p- and d-orbitals may be due to either of the two
factors: (i) due to spin-orbital, i.e. j−j coupling, e.g. in case of heavy atoms, and (ii) ionic state has both orbital
and spin degeneracy, e.g. in case of light atoms.
(b) 3 s- and 2p PES of Na (Atomic Number = 11) The electronic configuration of Na is 1s22s22 p63s1, which
is an open-shell configuration. The photoionsation of 3s electron is represented as
−1
Na (…2p6 3s1) 2S1/2 ⎯3s → Na+ ( 2p6) 1S 0 + −
Thus, 3s PES of Na will exhibit a single peak corresponding to 1S 0 state Na+. It conforms to the observed 3s
spectrum where a single peak has been observed at 5.14 eV using He (II) 40.8 eV radiation source.
Now, when 2p electrons are involved in photoionisation, the process is represented as
−1
Na (…2p6 3s1) 2S1/2 ⎯2 p → Na+ ( 2p5 3s1) + e −
1 1 1
For Na (…2p6 3 s1): l = 0, s = , L = 0, S = ∴ J = and
2 2 2
2S+1
LJ = 2S1/2
1 1
and for Na+(…2p53S1): l1 l2 = 0; L l1 + l2 0 ; s1 = , s2 , S = s1 ± s2 = 0,1
2 2
∴ J = L + S = 1 + 1 = 2 and 1 + 0 = 1.The possible terms are 2S 1

LJ 3
P2 and 1P1. The 3P2 term further splits
into 3P1 and 3P0 terms. Thus, the four states of Na+ are 3P2, 3 P1 ,3 P0 and 1P1 . Consequently, the 2p spectrum of Na
will exhibit four peaks corresponding to the 3P2, 3P1, 3P0 and 1P1 states respectively. According to Hund’s rule of
multiplicity, the binding energy of the states should follow the order 1P1 (J = 1) > 3P0 (J = 0) > 3P1 (J = 1) > 3P2 (J = 2).
According to the (2J + 1) multiplicity rule, the intensity ratio of the peaks is 3:1:3:5. The predicted spectrum is con-
sistent with the experimentally observed spectrum obtained by using He (II) 40.8 eV radiation source. Note that the
energy of the states follow the order 1P1>3P2>3P1>3P0.
(c) 4d PES of Ag and Cd The atomic number of Ag is 47 and that of Cd is 48. These two atoms are selected to
see how the multiplet structure in the PES is influenced if the open shell is closed by the addition of an electron.
The 4d photoionisation of Ag proceeds according to the scheme
(
Ag ...4 d10 5s1 ⎯ ) 4 d −1
(
→ Ag + ...4 )
5 1 + e−
4 d 9 5s
1 1 1 1
For Ag+(..4d95s1): s1= , s2 = , S = ± =1, 0, l1 = 2, l2 =0, L = l1 + l2 = 2 + 0 = 2
2 2 2 2
∴J = 1 + 2 = 3 and 2 + 0 = 2 and 2S+1LJ = 3DJ and 1D2. The term 3DJ further splits into 3D3, 3D2 and 3D1 terms.
Thus, the 4d PES of Ag will consist of four peaks. The BE corresponding to these peaks follows the order: 3D3 <
3
D2 < 3D1 < 1D2. The intensity ratio of these peaks according to the (2J + 1) rule is 7:5:3:5. The predicted spectrum
is consistent with experimental spectrum obtained by using He (II) 40.8 eV radiation source.
In case of Cd, the electronic configuration of Cd+ cation is formed because of 4d photoionisation and is 4d 95s2.
The photoionsation process in this case is represented as
(
Cd ...4d 10 5s 2 ⎯ ) 4 d −1
(
→ Cd + ...4d 9 5s 2 + e )
In order to predict the 4d PES of Cd, we are to work out the spectroscopic terms for Cd+(…4d95s2).
1 1 1
Here s ,S= ;l , L = 2 ∴J 2 / 2 d 2 S +1LJ = 2 D5 / 2 .2 D5 / 2 term further splits into 2D3/2
2 2 2
⎛ −1 −1 1 ⎞
state ⎜ i.e. s ,S= ;l , L = 2 ∴ J = 2 − = 3 / 2⎟ . The BE of the 2D5/2 state will be small as compard
⎝ 2 2 2 ⎠
to that of the 2D3/2 state. The intensity ratio of the two peaks is 6:3. Again, the predicted spectrum matches with the
experimentally observed spectrum obtained by using He (II) 40.8 eV radiation source. The spectrum reveals that the
four peaked structure of the … 4d 95s1 configuration decreases to double peaked structure by closing one of the open
shells (5s1) by the addition of an electron.
(d) PES of Ar (Atomic Number = 18) Uptil now, we have discussed the partial PES arising because of photo-
ionisation of an electron from a particular level of an atom. Let us now apply the rules to work out the complete
PES of the Ar atom. The electronic configuration of Ar is 1S2 2S2.2p63s2 3p6. The complete PES will result from the
photoionisation of 1s, 2s, 2p, 3s and 3p electrons of the Ar atom. The spectroscopic terms corresponding to these
five different configurations of photoionised Ar+ atom are the following:
l 2S1/2, 2 2S1/2, (2 2P1/2, 2 2P3/2), 3 2S1/2 and (3 2 P1/2, 3 2P3/2) respectively. Accordingly the complete spectrum will
consist of seven peaks. BE of the photoelectrons from different orbitals will follow the order 3p(3 2P3/2 <3 2P1/2)
<3s (3 2S1/2) <2p (2 2P3/2 <2 2P1/2) <2s (2 2S1/2) <1s(1 2S1/2). The intensity ratio of the peaks is (4 :2) : 2 :(4 : 2): 2 : 2.
The completely resolved experimentally observed spectrum of Ar obtained by combining both ESCA (Cr Ka =
5414.7 eV) and PES [He (I) = 21.21 eV] is consistent with the predicted spectrum. The spectrum reveals that only
the 3p electrons of Ar are within the range of 6–21 eV of the He (I) radiation source. The BEs corresponding to
3 2P1/2 and 3 2P3/2 states are 15.933 and 15.755 eV respectively.
(e) 6s and 5d PES of Hg (Atomic Number = 80) Hg is a candidate from heavy atoms with closed-shell struc-
ture. The photoionisation of 6s and 5d electrons is represented as
( ) 6 s −1
( )
1
Hg ...5d 1 6s 2 S0 g + ... d 10 6 s1 + e −
Hg
Hg (...5d 6s ) S g (... d 6s ) + e
5d −1
1
1 2 + 9 2 −
0 Hg
For Hg + ( ) : s = 12 , S = 12 , l = 0,, L = 0,∴ J = L + S = 12 and the spectroscopic term 2s +1

LJ = 2S1/2
Thus, the 6s PES of Hg should consist of a single line. Indeed the same has been observed experimentally
[He (I) 21.22 eV]. The BE corresponding to 2S1/2 state is 10.44 eV.
Now in order to predict the 5d PES of Hg, we are to write the states for the configuration Hg+ (… 5d9 6s2). In
1 1 1 5
this case, l = 2, L = 2, s = , S J = L S = 2+ = .
2 2 2 2
Thus, the spectroscopic term corresponding to this configuration is 2S+1LJ = 2D5/2. This term splits further into
⎛ −1 −1 3⎞
2
D3/2 terms ⎜ i.e. l 2, L = 2, s , S = , ∴ J L + S = 2 −1 1/ 2 . Consequeuently, the 5d PES of Hg
⎝ 2 2 2⎠
will show two peaks corresponding to the 2D5/2 and 2D3/2 states. According to Hund’s rule of multiplicity, the energy
of the 2D5/2 state should be on the lower side of the 2D3/2 state. The experimentally observed data on BE follows
the expected trend, i.e., 2D3/2 (16.71 eV) > (2D5/2 )14.84 eV. The intensity ratio of the two peaks, as deduced from
the (2J + 1) rule, has been found to be 4:6. However, the experimentally observed intensity ratio is 2.4:6 (from
integral spectrum). The intensity ratio for the complete 6s and 5d photoionisation from the differential spectrum
is ≈ 6 (2D3/2): 21 (2D5/2) :1 (2S1/2). The discrepancy between the observed and theoretical intensity ratios reveals that
the relative intensities of the bands are not solely governed by the spin-orbital coupling but by other factors also.
To cite a few, the relative probability of ionisation to different ionic states depends on the nature of the orbitals
and their size, number of nodes, etc., and angular distribution of electrons.
The 5p3/2 electron spectral lines of Th metal, V metal, and UO3 also split whereas the 4f 5/2 and 4f 7/2 spectral lines
of the same sample do not split. On the other hand, 5p3/2 line for gold in some gold compounds varies linearly with
MB quadrupole splittings of the compounds. Since quadrupole splitting depends on the symmetry of the electric
field around the nucleus of the atom in question, the x-ray photoeleetron spectroscopy may also give information
about the symmetry of the electronic environment around the atom.
Splitting has also been observed in paramagnetic molecules and transition metal ions. The splitting in para-
magnetic species is due to the different behaviour of unpaired electrons towards core electrons. The exchange
interaction affects core electrons with a (spin up) and b (spin down) spins differently.
6.12 RULES FOR THE PREDICTION OF ATOMIC PHOTOELECTRON

SPECTRUM
The atomic PES can be deduced by making use of the following steps:
1. Write the electronic configuration of the photoionised cation.
2. Using the general term 2S+1LJ, write the terms for the photoionised states for the cation.
3. The number of peaks in the PES corresponds to the number of photoionised states.
4. According to Hund’s rule of multiplicity, the BE’s of the states will follow the reverse trend of J’s.
5. The intensity ratio of the peaks can be predicted from the gJ =(2J + 1) multiplicity rule.
6. Draw the stick spectrum and compare with the experimentally observed values if available.
6.13 PHOTOELECTRON SPECTROMETER

The schematic diagram of a photoeleetron spectrometer is shown in the self-explanatory Fig. 6.13. The basic ele-
ments of the photoeleetron spectrometer are radiation source, target, ionisation chamber, electron energy analyser,
detector, recorder and ultra high vacuum chamber.
(a) Radiation Sources They form the backbone of the photoeleetron spectrometer. In UV-PES, the radiation
source is normally a noble gas discharge lamp and sometimes helium mixed with any other noble gas. The energy
of the HeIa line at 584.3 Å (21.22 eV) from the He discharge lamp is sufficient to cause ionisation of the majority
of the valence shell electrons. Moreover, this line is not accompanied by any lines of lower energy down to about
4 eV. The HeIIa line at 303.8 Å (40.81 eV) from ionised He (He+) produced in the He discharge lamp is also used
since its energy is sufficient to cause ionisation of most of the valence-shell electrons. Discharge lamps of other
inert gases alone or mixed with helium are used to produce low-energy radiations. Discharge in neon gas gives
NeIa lines at 16.67 and 16.85 eV while in argon gas gives ArIa lines at 11.62 and 11.83 eV and Ar II (Ar+) lines at
13.30 and 13.48 eV. Lyman −a, −b and −g lines from hydrogen discharge lamps have energies of 10.198, 12.087
and 12.748 eV respectively.
Molecular beam
−
Electrons
Ekin <E⬘kin
Target
Chamber
Detector
Computer
+ Chart
Recorder
Radiation Source Electron Energy Amplifier
analyser Oscilloscope
Fig. 6.13 Block diagram of a photoelectron spectrometer.
In XPS, soft x-ray sources which have energies of 1000 eV and higher are most commonly employed, e.g. Mg Ka
line at 1253.6 eV and Al Ka line at 1486.5 eV. The emitted photoelectrons from Mg Ka source will have KE's in the
range of Ca 0–1250 eV while Ca 0–1480 eV from Al Ka. The very soft x-ray sources in the energy range of 130–300
eV can bridge the gap between the UV and x-ray regions. These very soft x-rays were discovered by M A Krause
in 1970. 132.3 eV Yttrium line is most useful but its intrinsic line width is 0.6 eV and intensity is also very low.
Presently, the synchrotron radiation is the most promissing emission source in the intervening wavelength region
for photoelectron studies. The sychrotron radiations of variable wavelengths are generated by electron/positron
accelerators. The energy range of these sources is Ca 5–5000 eV plus and the intensity is also sufficient for PES.
Moreover, the radiation is fully polarised in the plane of electron paths (80 per cent or more) and has a very small
angular divergence out of plane. Though such radiation sources have enabled high-resolution studies in photoionisa-
tion and adsorption studies of gases and of solids spanning much wider and more complete energy range, such work
is, and will remain, limited due to expense, complexity and limited availability of such sources.
(b) Sample and Ionisation Chambers Samples can be in solid, liquid, or vapour form, depending on the
pumping techniques. For surface studies, solid samples freed from moisture and other impurities are placed in a
suitable sample holder which is mounted in a fixed position close to the monochromatic photoelectron source in
the ionisation chamber (not shown in Fig. 6.13); otherwise a stream of volatilised sample or a beam of molecules
in cases of gases is introduced from the top into the target chamber where it is irradiated with photons from an
appropriate radiation source. The electrons emitted from the sample are made to enter the electron-energy analyser.
Liquids at room temperature have been studied by vapourising or freezing. Solutions of K3Fe(CN)6’ K4Fe(CN)6
and NaCl are examined by this method. XPS of frozen solutions of K4Fe(CN)6 and K3Fe (CN)6 show that the BEs
of the Fe 3p electrons differ by about 0.2 eV from the corresponding solid samples.
(c) Ultra High Vacuum Chamber All the electron optics is placed in a chamber fitted with high vacuum pumps
to produce high vacuum (10−5 mmHg or less) (not shown in Fig 6.13). The ionisation is carried out in a high vacuum
environment to enable the emitted electrons to be analysed without interference from gas-phase collisions.
(d) Electron-energy Analysers They are the heart of any photoelectron spectrometer and the function of an
analyser is to sort out the emitted electrons according to their kinetic energy. There are many different designs of
electron energy analysers but the most commonly preferred and widely used for photoemission experiments is
a concentric cylindrical or hemispherical analyser which uses an electrostatic field between the two cylindrical/
hemispherical surfaces to deflect or disperse the electrons according to their kinetic energy. Accordingly, they are
known as deflection or dispersion analysers.
For cylindrical plates, the kinetic energy Ekin of electron is given by
V2 V1
Ekin = (6.35)
2r log(r1 / r2 )
Here, V1 and V2 are the plate voltages, r1 and r2 the radii of the plates and r is the average radius.
The path of the electrons in the analyser is affected by an external magnetic field. So for better resolution,
external fields are reduced to about 0.1 mG ferromagnetic shielding. Magnetic-deflection analysers are also used
but they are not very suitable for low-energy electrons because weak magnetic fields are required and it is a
problem to nullify the effect of external fields.
The other class of analysers is the retarding-field analysers. In these types of analysers, use is made of cylin-
drical/spherical/slotted geometry and the electrons with energies higher than a retarding potential are permitted
to reach the detector. The detector in this case measures the photoelectron current as the retarding potential is
varied. The recorded spectrum, thus, should contain a step in the current for each group of photoelectrons of
discrete energy. The steps of low-energy electrons are superimposed on the electron current of all electrons of
higher energy, and hence it becomes difficult to examine the low-energy electrons. However, they are capable
of accepting electrons in a large solid angle and from extended sources. They are equally sensitive to electrons
of all energies, particularly low-energy electrons where the differential analysers are usually least sensitive as
compared to deflection analysers.
The theoretical resolution of such an analyser is given by
2
ΔV ⎛ r1 ⎞ (6.36)
=⎜ ⎟
V ⎝ r2 ⎠
where DV is the smallest resolvable energy difference at the electron energy, r1 is the radius of the ionisation region
and r2 is the radius of the retarding electrode. The maximum value of resolution achieved is 0.5 meV, but in these
type of analysers it is difficult to maintain high resolution in the presence of target high reactive gases since their
adsorption on the surface of the retarding electrode will alter the contact potential and hence the resolving power.
(e) Detector The electrons from the electrostatic analyser with varying voltage are introduced into the detec-
tor for further analysis. The detector counts the number of elecltrons per second as a function of their kinetic
energy, i.e. all electrons of the same kinetic energy are counted. Note that the electron energy is scanned by the
variation of the analyser field or preaccelerating potential, i.e. the potential difference between the target cham-
ber and the analyser entry slit. The latter method is preferred because the analyser is set to transmit electrons of
fixed energy for which its resolution and transmission can be optimised and that the resolution of the resulting
spectrum is independent of electron energy. However, the contact potential varies from compound to compound
and shifts the energy scales as a function of pressure. Hence, it is essential to record the spectrum of a mixture of
sample and calibrant. A calibrant should give sharp peaks precisely at known ionisation potentials. The detector
is calibrated with known lines of CH3I, argon, krypton or xenon, (which are useful for ionisation energy range
below 15 eV), the recorded half width of which gives a guide to the resolution. Resolution of 10–20 mV (80−160
cm−1) can easily be obtained. Some other gases and different ionising radiations can be used in order to extend the
calibration range to higher and lower ionisation energies. Some of the useful calibrant lines due to D R Lloyd are
mentioned here in Table 6.1.
Table 6.1 Some calibrant lines due to D R Lloyd.
Substance Ionic state Ionising line Electron energy (eV) Ionisation potential (eV)
2
P3/2 He IIa 19.2494, 1.9681,
Ne 2
P1/2 He IIa 19.3463 2.0650
He 2
S He IIa 16.2268 4.9907
2
E1/2
He Ib, He Ia 12.922,11.680, 8.296, 9.538
MeI 2
E3/2
He Ia 11.053 10.165
2
E1/2
2
P3/2 He Ia 9.088, 12.130,
Xe 2
P1/2 He Ia 7.782 13.436
Kr
2
P3/2 He Ia 7.219 13.999
2
P1/2 He Ia 6.533 14.665
Hg 2
D5/2 He Ia 6.378 14.840
2
P3/2 He Ia 5.459, 15.759,
Ar 2
P1/2 He Ia 5.281 15.937
Hg 2
D3/2 He Ia 4.514 16.704
+
N2 ∑u He Ia 2.467 18.751
2
P3/2 He Ib 1.5223, 19.6952,
Ne 2
P1/2 He Ib 1.4253 19.7922
CH3I 2
E3/2 H Lymana 0.661 20.557
(f) Recorders The rate at which electrons of each energy arrive at the detector can then be stored in the com-
puter or displayed on an oscilloscope screen or recorded on an X-Y chart recorder for qualitative analysis.
The XPS spectrum obtained from a Pd metal sample using MGKa radiation on a deflection-type photo-
electron spectrometer is shown in Fig. 6.14 (a). The main peaks occur at kinetic energies of Ca 330, 690,
720, 910 and 920 eV. If the energy of the radiation is known, it is a simple matter to transform the spectrum
so that it is plotted against BE as opposed to KE. The resultant spectrum in case of Pd spectrum is shown in
Fig. 6.14 (b).
3E + 05
0
270 1270
Kinetic Energy (eV)
(a)
3E + 05
3d
MNN
3ρ
3s
0
1000 800 600 400 200 0
Binding Energy (eV)
(b)
Fig. 6.14 X-ray photoelectron spectra of Pd generated using MgKa 1253.6 eV radiation source:
(a) intensity vs kinetic energy (b) intensity vs binding energy.
Problem 6.1: Calculate the wavelength of resonance line of the helium atom if excitation energy of the
resonance level is 21.22 eV.
Solution The wavelength λ (in Å) is equal to
hc 12404( )
λ= = = 584.5 Å
ΔE 21.22(eV)
where ΔE is the difference in the potentials of the ground and excited states in eV; h is Planck’s constant and c is
the velocity of light in vacuum.
Problem 6.2: In the C (ls) photoelectron spectrum of CO2 and CH4 mixture, there is one more line between the
CH4 and CO2 lines at BE (CH4) = 290.7 eV and BE (CO2) = 298.4 eV. Calculate the energy of the unknown line if
on the screen of the oscilloscope, it is 1.8 mm away from the line of CH4 and 2.2 mm from the line of CO2.
Solution In order to determine the energy of the unknown line that lies between the lines of CH4 and CO2, we
need to convert the energy corresponding to the CH4 and CO2 lines into wavelength units which are additive in
nature.
12404( Å )
Thus, λ( 4) = = 42.414 Å
290. (eV)
12404(eVÅ )
and λ (CO 2 ) = = 41.320 Å
298. (eV)
we find λx as
dx
λx λ (C
(CH 4 ) + [λ ( 2) λ (CH 4 ) ×
d
Here, dx is distance of the unknown line from CH4 line and d is the distance between the two known lines. Substi-
tuting the values of the respective parameters into the above expression, we obtain
1 8 mm
λ x = 42.414 Å + Å− Å) ×
4 0 mm
= 42.414 Å − 1.094 Å × 0.450 = 42.414 Å − 0.492 Å = 41.922 Å
12404( Å )
The energy of the unknown line = = 295.8 eV
41.922 Å
Problem 6.3: The x-ray PES of Pd metal produced using MgKa 1253.6 eV radiation source is shown in
Fig. 6.14 (b). It exhibits peaks at BEs (measured with respect to Fermi level) of: 0−8 eV, 54 eV (very weak), 88 eV
(very weak), 335 eV (very intense), 534/561 eV (medium), 673 eV (weak), and 924 (strong).
(a) Assign the peaks to the various orbitals involved in photoionisation in the Pd metal, (b) The high-resolution
spectrum of Pd metal shows a doublet instead of a single peak at 335 eV as shown below in Fig 6.15. Explain.
2E + 05
334.9 (2D5/2/3d5/2)
340.2 (2D3/2/3d3/2)
Intensity (conts/sec)
0.00
330.0 Binding Energy (eV) 345.0
Fig. 6.15 High resolution spectrum of Pd.
Solution The electronic structure of Pd with an atomic number of 46 is

1s 2 2 s 2 2 p 6 3s 2 3 p 6 3d 10 4 s 2 4 p 6 5 1 4d 9
K L M N
In order to assign the various peaks in the problem, let us proceed from the valence electrons towards the
nucleus. The BE of electrons will increase as we proceed from the valence electrons towards the nueleus. Thus,
(i) The peaks in the BE range of 0–8 eV are attributed to the valence electrons from 4d and 5s orbitals respectively,
(ii) The very weak peaks at 54 and 88 eV originates because of photoionisation from 4p and 4s levels/orbitals
respectively. (iii) The very strong peak at 335 eV is due to emission of electron from 3d level. (iv) The medium
and weak intensity peaks at 534/561 and 673 eV are attributed to electrons emitted from 3p and 3s levels respec-
tively. (v) The strong peak at about 924 eV, is due to x-ray induced Auger emission (MNN).
(b) The splitting of the peak at 335 eV into a doublet (BE ~335 eV, ~340 eV) with intensity ratio of 3:2 is
because of spin-orbit coupling effects in the palladium ion. The inner-core electronic configuration of the initial
state of the Pd is 1s2 2s2 2p6 3s2 3p6 3d10 … with all subshells completely full. The removal of an electron from
the 3d subshell by photoionisation will yield a configuration in which the 3d subshell will not be completely
full, i.e. Ar 3d9 … Accordingly, the of d-orbitals (l = 2) have nonzero orbital momentum; consequently, there will
be an interaction between the unpaired spin and the orbital angular momenta. The spin-orbit coupling is treated
by the R−S coupling model and the j−j coupling model. Let us deduce the splitting of 3d−1 PES spectrum from
R−S model.
Photoionisation of 3d electrons will yield Pd+ ion with a hole in the 3d orbital according to the scheme
−1
Pd (Ar 3d10…) ls0 ⎯3d → Pd+ (Ar 3d9…) 2D5/2 and 2D3/2+ e
The atomic states involved in this process are worked out from the general term 2S+1LJ as follows:
For Pd (Ar 3d10…) : l = 0 s = 0 ∴ J = 0 and 2S+1LJ = 1S0.
1 1
For Pd+ (Ar3d9…) : l=2 s= , ∴ J = 2+ = 5/2
2 2
1 1
and LJ = 2D5/2, l = 2, s = − ,
2S+1
∴J=2− = 3/2
2 2
and 2S+1LJ = 2D3/2, i.e. the 2D5/2 term further splits into 2D3/2 terms. (Here, the interaction of 3d9 electrons with
the valence electrons is ignored). Thus, 3d−1 PES of Pd will consist of two peaks corresponding to the 2D3/2 and
2
D5/2 higher energy states respectively. According to Hund’s multiplicity rule, the BE of 3d−1 electron in 2D5/2 state
should be small relative to that of in 2D3/2 state, i.e. BE (2D3/2 ) > BE (2D5/2) and the same has been observed indeed.
The intensity ratio of these peaks according to the gJ = 2J+ I rule is 2: 3 as expected.
We know that the R−S coupling model approximation is best applied only to light atoms. However, this splitting
can alternatively be described using individual electron l-s coupling. In this case, the resultant angular momenta
arise from the single hole in the d-shell. A single d shell electron (or hole) has l = 2 and s = 1 which gives rise to
2
permitted values of j = 3/2 and 5/2, the former being higher in energy than the latter as expected.
6.14 APPLICATIONS OF PHOTOELECTRON SPECTROSCOPY

The applications of photoelectron spectroscopy are centered around the three parameters namely the position of
the photoelectron signal expressed in terms of BE or KE, area under the photoelectron signal, width of the signal
and the chemical shift. Based on these parameters, a number of qualitative and quantitative methods have been
developed to study the surface properties of metals and alloys, electronic and molecular structure of diatomic
and polyatomic molecules, analysis of gases, etc. Some of the typical applications of photoelectron spectroscopy
based on the above-said parameters are being described here.
6.14.1 Ionisation Energies

PES provides a direct method for the determination of ionisation energies of the molecules. The first and second
ionisation energies can simultaneously be determined. The energy required to evict a single electron from the
first highest filled orbital is called first ionisation energy and from the second highest filled orbital is called the
second ionisation energy. The highest bonding molecular orbitals p2 and p3 in benzene are degenerate, and so a
single ionisation band is expected corresponding to these orbitals. In monosubstituted benzenes, the degeneracy
is removed, and consequently two bands corresponding to p2 and p3 ionisations are expected and the same has
been observed indeed, e.g.
C6H6(9.40); RF (9.5, 9.86), RCl (9.31, 9.71), RBr (9.25, 9.78). RI (8.78, 9.75);
RNH2(8.04, 9.11) RNH CH3(7.73, 9.03), RN(CH3)2 (7.51, 9.03); ROH (8.75, 9.45),
ROCH3(8.54, 9.37), ROC (CH3)3(8.75, 9.33). R = C6H5 and the energies are in eV.
6.14.2 Surface or Interelectronic Structure Studies

XPS, being a surface-sensitive technique, is more sensitive to those atoms which
are located near the surface than it is to atoms in the bulk which are well away
Metal
from the surface, i.e. the main part of the signal comes from the surface region. Evaporation
However, XPS is not completely surface specific, (i.e. it does not give signals source
due to atoms only in the surface region) in that whilst most of the signal comes
Flux of atoms
from within a few atomic layers of the surface, a small part of the signal comes
from much deeper into the solid. Thus, XPS is best described as being only
surface sensitive but is not also a surface-specific technique. This can be dem- Substrate
onstrated by depositing some metal by evaporation onto a substrate of another
material as shown in Fig. 6.16. It is experimentally observed that in such a situ- Fig. 6.16 Deposition of one material
ation, the XPS signals due to the substrate are rapidly attenuated, whilst those onto a substrate of another material.
due to the condensed evaporant simultaneously increase to a limiting value. It is possible to measure deposition
rates using equipment such as quartz crystal oscillators and thus obtain independent verification of the deposited
film thickness. A detailed analysis of the variation of signals due to depositing material and substrate can also
give information about the mechanism of film growth. We know that the soft x-rays employed in XPS penetrate
a substantial distance into the sample (~μm). Thus, this method of excitation imparts no surface sensitivity at the
required atomic scale. The surface sensitivity arises from the emission and detection of photoemitted electrons.
Thus, if the source atom is closer to the surface then a greater fraction of the emitted electron will be detected
since; (i) the detector would subtend a greater solid angle at the emitting atom, and (ii) for an electron emitted
towards the surface, there would be less chance of inelastic scattering before it escaped from the solid. Note that
the former factor is insignificant in real experiments. Thus, the probability of detection depends upon the distance
of the emitting atom below the surface. The probability of escape from a given depth, P(d), is determined by the
inelastic mean free path, λ (IMFP) also called the penetratien (escape) depth of electrons, which is a measure of
the average distance travelled by an electron through a solid before it is inelastically scattered.
The IMFP is actually related to the probability of an electron travelling a distance d through a solid without
undergoing scattering by the expression.
P (d) = exp (−d/λ) (6.37)
Further, IMFP depends upon (i) the initial kinetic energy of the electron and λ ≅ (KE)l/2 for 100−10,000 eV
kinetic energy range, and (ii) the nature of the solid. Generally, ESCA electrons have KE’s of approximately
100−1500 eV with a nominal escape depth of 20 Å while AEs’ electrons have KEs of 50−2500 eV with an escape
depth of 20 Å. The variation of the IMFP with initial electron energy (eV) is shown in Fig. 6.17.
This behaviour of IMFP with energy is reasonably accurate for most metals, although experimental measure-
ments on different metals show a substantial scatter about the general curve. Other classes of solids also show
qualitatively similar universal curves, although the absolute values of the IMFP of electrons in materials such as
inorganic solids and polymers differ quite significantly from those in metals. The IMFP data is usually displayed
on a log-log plot as shown in Fig. 6.18.
The graph clearly shows how the IMFP varies at low KEs. The IMFP shows a minimum for electrons
with a KE of around 50–100 eV; at lower energies the probability of inelastic scattering decreases since
the electron has insufficient energy to cause plasmon excitation (the main scattering mechanism in metals),
and consequently, the distance between inelastic collisions and the IMFP increases. The IMFP in metals is
typically less than
2.5
10000
Escape Depth, λ(nn)
Escape Depth, λ (nm)
1000
10
0.10
0.0 1 10 100 1000 10000
0.00 Energy (eV) 3000 Kinetic Energy (eV)
Fig. 6.17 Universal curve for the variation of the escape Fig. 6.18 A log-log plot of electron escape depth against
depth with initial electron energy. kinetic energy of the electron.
10 Å (1nm) for electron energies in the range 15 < El eV < 350 A

20 Å (2 nm) for electron energies in the range 10 < El eV < l l0
t
1400
i.e. the IMFP of low-energy electrons corresponds to only a few
atomic layers.
The IMFP can be used to calculate the thickness of surface films. B
Let us consider that a substrate or thick film of one material, say A,
is covered by a thin film of thickness ‘t’ of another material, say B
(Fig. 6.19). The XPS signal from the underlying substrate will be Fig. 6.19 Schematic diagram showing a thick film
attenuated, i.e. reduced in intensity due to inelastic scattering of of material ’A’ covered by a thin film of thickness‘t’
some of the photoelectrons as they pass through the thin layer of the of another material B.
covering material.
The probability of such a scattering process for any single photoelectron passing through this layer is given
by
P = exp (−t/λ) (6.38)
Here, t is the thickness of layer of the covering material.
Accordingly, the overall intensity of an XPS signal due to the substrate will be reduced by this factor. Thus,
if the intensity of this signal in the absence of any covering layer is I0 then the intensity I in the presence of the
overlayer of thickness ‘t’ is given by
I = I0 exp (−t/λ) (6.39)
It follows that it is possible to determine the thickness of a deposited layer from Eq. (6.39) provided the reduc-
tion in the substrate signal due to deposition of the covering layer is known.
Note that the degree of surface sensitivity can be increased by collecting the photoelectrons at a more grazing
emission angle. Let the atoms be located within a smaller distance, x, from the surface (Fig. 6.20) such that
x = d cos q (6.40)
Here, the photoelectrons have to travel a distance ‘d’ which is less than
q
1 λ through the solid in a direction making an angle q with the normal to x
d
the surface. So as the emission angle (q) is increased, cos q decreases, the
analysed region becomes more surface localised and the surface sensitivity Fig. 6.20 Diagram depicting the effect of
is increased. grazing emission angle of photoelectrons
This effect can be used in the study and diagnosis of sur- on the surface sensitivity.
face segregation as is evident from Fig. 6.21 which shows 20
simulated data for the variation of the XPS signal intensi-
ties with emission angle for: a random alloy of 10 percent 1.6
of element A in element B, with no segregation and the
same random alloy, but with surface segregation of A to 1.2
give a surface monolayer, i.e. IA/IB is independent of the
lA/lB
emission angle. 0.8

The average size of particles on a surface is sometimes (a)
of interest. The PE signal intensity I is related to the size of 0.4
(b)
the particles by the expression
0.0
⎛ λ2 ⎞
0 10 20 30 40 50 60 70 80 90
I 0 ⎜1 − (6.41) Emission angle, θ
I ⎟
⎝ 12r 2 ⎠ Fig. 6.21 Angular variation of XPS signal ratio for a random
alloy of 10 per cent element A in element B (a) with no segre-
Here, r is the radius of the particles on a surface and I0 gation, and (b) with surface segregation of A to give a surface
is the signal intensity in the limit r << λ. monolayer of pure element A.
Further, we know that in XPS, the photon is absorbed by an atom in a molecule or solid leading to ionisation
and the emission of core, i.e. inner-shell electron. For each and every element, there is a characteristic binding
energy associated with each core atomic orbital, i.e. each element will give rise to a characteristic set of signals
in the photoclectron spectrum at KEs determined by the photon energy and the respective BEs. The presence of
peaks at particular energies, therefore, indicates the presence of a specific element in the sample under study.
Since the intensity of the peaks is related to the concentration of the element within the sample region, the
technique provides a quantitative analysis of the surface composition and is known as ESCA. It being surface
sensitive, has been proved to be the most successful and useful tool alone and in conjunction with AES to study
the analytical chemistry of surfaces as in evident from the following typical examples. However, before describ-
ing the examples let us define a term, called surface shift, which is frequently used in XPS. If Eion (i) is the bond
energy of the ion to the surface, after an electron has been removed from the ith orbital, Eneut is the bond energy
S
of the atom or molecule to the surface in the ground state then the BE of the ith orbital on surface EBE (i ) can
g
be related to the BE in the gas phase, EBE (i ) by
S
EBE (i ) = EBgE (i ) − Eion (i ) + Eneut (6.42)
g s
The surface shift which is EBE (i ) − EBE (i ) is given by the difference in the bond strength of neutral and the
ionised molecule. This energy difference will depend upon not only the nature of the ith orbital, but the degree
of electronic and spatial relaxation of the ion. Let us now compare the penetration depth of various particles and
effect of probes on surfaces. The penetration (escape) depth of photons, electrons and ions in the energy range
used for surface analysis are given in Table 6.2, while the effect of probes on surfaces is reported in Table 6.3.
Table 6.2 Penetration depths of particles. Table 6.3 Effect of probes on surfaces.
Particle Energy(eV) Depths (Å) Probe Effects

Photon 1000 10,000 Least destructive
Electron 1000 20 Photons IR <UV < VUV < x-rays
95 per cent OK
Ions 1000 10
More destructive
Insulators >Semiconductors
Electrons
>Conductors
Charging bad for organic materials
Cause sputtering;
Ion beams
All materials sputtered
(a) Presence or Absence of Element on the Surface The ESCA spectrum of Ni/W/Al2O3 catalyst using
AlKa x-rays for ionising radiation shows the following peaks not only corresponding to the elements present in the
catalyst but an additional peak at 285 eV Table (6.4). The peak at 285 eV (Cls) is attributed to the carbonaceous
material present in the atmosphere.
Table 6.4 Peaks in the ESCA spectrum of Ni/W/Al2O3 catalyst.
Element Orbital ( ≈BE)

O 2p (9 eV)
O 2s (23 eV)
W 4f (38 eV)
Al 2p (73 eV)
Ni 3s (102 eV)
Al 2s (109 eV)
W 4d (244 eV)
C 1s (285 eV)
O 1s (544 eV)
O KLL Auger line
Ni LMM Auger lines amplified signals
Ni 2p (866 eV amplified signals
(b) Determination of Oxidation States on Catalyst Mo 3d ESCA spectrum of Mo/Al2O3 (15 percent MoO3
on Al2O3) catalyst shows a low-intensity signal at BE = 236.1 eV and a high intensity signal at BE = 233 eV. The 3d
Mo spectral data of the catalyst as function of reduction time (catalyst reduced by H2 at 500°C) shown in Fig. 6.22
indicates that initially the Mo is in the +6 state. As the reduction progresses, the rapid fall of the +6 state takes
place with the simultaneous increase of +5 and +4 states and after about 40 minutes, all of Mo +6 disappears while
the +5 state passes through a maximum. After about 10 hours, the mixture of oxidation states reaches a steady
state (60 per cent + 4 and 40 per cent +5) and remains constant.
(c) Differential Rates of Reactions of Two species 100

(a)
on a Surface The ESCA can be used to follow the 80
Percent Mo
differential rates of reactions of two species on a catalyst (b)
60
surface. The ESCA spectrum of a pure and fresh catalyst
(c)
(Ni/W/Al2O3) shows two peaks at 37.5 eV and 857.5 eV 40
which are characteristic of tungsten (4f) and nickel 20
(2p) respectively. When Ni/W/Al2O3 catalyst surface is 0
exposed to H2/H2S mixture, the ESCA signal correspond- 0 40 80 120 160 600 1200
Reduction time (min)
ing to the sulphided Ni (854 eV) is fully developed after
Fig. 6.22 Mo oxidation states as a function of reduction time.
some time while the sulphided tungsten peak (31.5 eV) is (a) Mo (+6), (b)Mo (+5), (c) Mo (+4).
not fully developed at that particular time and even after
longer periods. Some unsulphided tungsten (37.5 eV)
remains after very long sulphiding times whereas none of the unsulphided Ni (857.5 eV) is seen.
(d) Adsorption of CO on W (100) in Relation to Effects of Different Orbitals on the Surface The BEs
of O1s (1s) and Cls (2s) levels of a1 state of CO in the gas phase are 542.6 and 296.2 eV respectively, while that of
CO adsorbed on (100) tungsten are 538.0 and 291.2 eV respectively. Thus, the spectral data on the measurements
for the a1 state of CO adsorbed on (100) tungsten show that the surface shift for the C1s level is 5.0 eV which is
larger than the surface shift of 4.60 eV for the Ols level. This discrepancy between the magnitudes of the surface
shifts is presumably because CO is bonded to the surface standing up with the carbon closer to the surface. Both
the Cls and Ols levels have a larger shift than the 1p level (16.54 − 13.7 = 2.84 eV), since they are more localised.
Thus, the measurements for the a1 state of CO adsorbed on W (100) illustrate the effects of different orbitals on
the surface of the catalyst. In case the value of the surface shift for both the orbitals is constant, it is presumed that
both the orbitals have a similar charge distribution relative to the surface, e.g. Ip and 4s levels of a2 state of CO
on W (100) have a constant value of surface shift, i.e. 3.2 eV.
(e) Change in the Nature of Surface Species as a Function of Composition of the Catalyst When the
ESCA intensity ratio of Ni/Al peaks in a Ni/Al2O3 catalyst with different compositions of Ni are plotted against
the nickel content in the catalyst, it is found that as the amount of Ni content increases, the intensity also varies
linearly and at about 17 per cent of Ni, the slope of the line increases abruptly. This indicates that upto 17 per cent
the nickel binds to alumina lattice in a highly dispersive fashion and above 17 per cent crystallites of NiO begin to
form on the catalyst surface. Thus depending upon the nature of species in the catalyst surface, the ESCA shows
two linear regions which form the basis of detection of the change in the nature of nickel species on the catalyst.
(f) Separation of Polymer from a Metal Surface The ESCA spectrum of clean aluminium surface shows
three signals at BEs’:1348 eV, P (2p); 1362 eV, A1 (2s) and 1409 eV, Al (2p). The P(2p) signal appears due to
contamination of the surface during the cleaning process. The ESCA spectrum of pure bulk polymer shows two
signals corresponding to Si (2s) and Si (2p) levels at BEs’ of 1329 and 1382 eV respectively. After striping of the
polymer, the aluminium surface still exhibits five signals. Presence of five signals in the spectrum reveals that
the tearing occured within the polymer layer. The polymer layer removed from the aluminium surface shows no
signals corresponding to aluminium thereby substantiating the findings that tearning occured within the polymer
layer but not within metal or metal and polymer layers.
(g) Study of Polymer Surface Reactions The ESCA spectrum can be used to study the polymer surface
reactions. The ESCA spectrum of Teflon shows Cls (294 eV) and Fls (694 eV) lines from the surface of Teflon.
On itching with Na/NH3, the fluorine line disappears, carbon shifts to a lower value of 286 eV and an oxygen
signal at 535 eV appears. These are the characteristics of carboxylic acid. When the itched surface is treated with
chlorox, some of the nonfluorinated polymer formed by Na/NH3 treatment is dissolved and a high BE carbon line
at 294 eV reappears along with fluorine line at BE of 694 eV. The shifted Cls line at 286 eV and the Ols line at
535 eV also persist.
(h) Study of Annealing of Cu–Ni Alloy Use can also be made of Auger lines in the ESCA spectrum for
analysis of surfaces. Let us see the effects on copper spectra caused by annealing a Cu–Ni alloy at 250°C. CuLLM
Auger lines in the ESCA spectrum of a Cu–Ni alloy as a function of annealing time indicate that initially Cu is
present as copper oxide (Cu2O) having a peak at 917 eV on the KE scale. During the annealing process, a peak
characteristic of elemental Cu appears at 920 eV on the KE scale. During the process, the intensity of the Cu2O
peak decreases with the increase of annealing time while that of NiO increases. These results conform to the
following reaction that occurs in the annealing process:
Cu2O + Ni→2Cu + NiO
In addition to ESCA spectra, Auger spectra alone and in conjunction with ESCA can also be used both qualitatively
and quantitatively to study the elemental surface composition. All elements (Z < 10), except H2 and He, can be
estimated quantitatively. The small escape depth of Auger electrons make them an ideal tool for the study of atomic
distribution normal to the surface by successive removal of surface layers. This can be done in several ways but
the most convenient method is ion bombardment sputtering which is possible while the Auger spectrum is being
measured. Typical sputtering conditions are Ar gas pressure (1 – 5). 10−5 Torr, Ar+ beam energy 0.5–2 keV, Ar+
beam current density 0.2–20 μA/cm2 and beam diameters on the specimen of several mm to 1cm. If the ultimate
depth resolution as determined by the escape depth of Auger electron (≈10 Å) is to be achieved, then the ion-beam
current density must be constant over an area larger than the diameter of the primary electron beam. Otherwise,
the area from which the Auger electrons are emitted is not flat and the Auger signal represents an average over a
range of depths. At sufficiently low sputtering rates (1–10 Å/min depending upon concentration gradients), several
Auger lines can be monitored sequentially. The in depth analysis of silicon wafer oxidised after phosphrous pre-
deposition is shown in Fig. 6.23. Four peaks, the oxygen 511 eV, phosphorous 120 eV, silicon 92 eV and the silicon
75 eV line characteristic of Si in SiO2 are measured while the oxide layer is sputtered away. The limitations of this
technique are that the sputtering cross sections vary from atom to atom and depend upon its chemical environment,
for example removal of equal amounts of Mo deposited on various surfaces. Mo forms on W a uniforn film with
a normal sputtering profile while on Cu, Au and Al, it agglomerates and can be observed over a large depth range.
Roughening of surface especially in impure material occurs during sputtering. The ion bombardment induced
movement of ions in insulating films results in accumulation of such ions on boundaries of the film. Further, it is
to be noted that in contrast to low-energy electron diffraction (Leed), AES neither requires a periodic nor a flat
surface. Thus it can be applied to unusual samples as fracture surfaces or rocks from the moon or other such plan-
ets. The use of ion sputtering in conjunction with AES not only allows depth analysis but also studies in relatively
dirty vacuum because the contamination can be removed continuously. AES is at present, by far, the most universal
1
surface analysis method with sensitivities down to less than of a monolayer, except for hydrogen and helium
which are not detectable. 100
Auger Peak Heights (Peak to Peak)
O (511 eV)
×1
P (120 eV)
× 10 Si-SiO2 (75 eV)
×1
Si (92 eV)
× 0.5 P (120 eV)
× 10
0 200 400 600 800 1000 1200 1400 1600 1800 2000 2200 2400 2600
Material Removed (in Angstroms)
Fig. 6.23 The indepth analysis of silicon wafer oxidised after phosphorous
pre-depositlon.
The Auger spectrum of Ni–Fe–Cr–B alloy shows a signal to each component (Table 6.5).
The presence of S, Si and C peaks at 156, 88, and 275 eV respectively on the KE scale indicates the presence
of impurities in the alloy while a strong peak at 513 eV, characteristic of oxygen (O), reveals that the surface is
heavily oxidised. The percentage composition of the constituents can be estimated from the relative intensities of
the respective signals after calibration.
Sometimes it becomes necessary to combine ESCA and AES surface techniques for a particular study. The
chemical shift obtained from ESCA coupled with the depth profiling capability of AES is the most common
Table 6.5 Peaks in the Auger spectrum of Ni-F-Cr-B alloy.
Element KE(eV)
Si 88
S 156
B 175
C 275
O 513 (Strong)
Cr 538
Fe 600,650,706
Ni 716,788,855
combination for the study of surfaces. Note that AES spectra are plotted as derivatives while ESCA spectra are
integrals. Let us now see how such a combination is effective to study the chemical effects on a surface when it
is corroded by reactive gases, e.g. ESCA spectrum of solder before melting shows peaks characteristics of SnO2
(Sn3d5/2) at BE = 486 and PbO2 (Pb 4f) at BE = 143 and 138 eV. After melting, in addition to the peak of SnO2 at
486 eV an additional peak at 485 eV characteristic of tin (Sn) metal appears in the spectrum whereas in case of Pb,
the entire spectrum of PbO2 except a very weak peak at 138 eV disappears with the apperance of two new peaks
at 141 eV and 137 eV which are characteristic of lead metal. Further, the ESCA studies of the effect of oxygen
exposure on the Pb/Sn and O/Pb + Sn ratios show that Pb/Sn ratio drops as the log (Langmuir of oxygen exposed)
O
increases while ratio increases indicating that alloy is oxidised, the reactive metal tin tends to concentrate
Pb+Sn
on the surface. AES (2 keV) depth profile of oxidised solder, i.e. the plot of Pb/Sn ratio against Ar sputtering time
O O
and ratio against Ar sputtering time, indicate that the ratio falls off rapidly on sputtering in, say,
Pb+Sn Pb+Sn
Pb
time tsp (second) and then tends to be constant, while the ratio first increases and passes through a maximum at
Sn
about time t = tsp (second) and then decreases, approaching the bulk value. Further, when the changes in oxidation
states of solder during oxidation by NO2 are followed by ESCA (i.e. from the plot of composition of Pb in alloy
as a function of oxidation time), it is found that for Pb, PbO2(+4) and PbO (+2) are the initial products, the former
being dominant species. As the oxidation proceeds, PbO2(+4) and PbO (+2) increase, and after about 60 minutes
when the former passes through a maximum, and then drops as PbO (+2) becomes the predominant species and
even at longer periods remain constant, the latter, i.e. PbO (+2), grows and then becomes constant. The sputtering
suggests that as the PbO (+2) moiety drops with sputtering time, the Pb (0) moiety grows. On the other hand, for
Sn, the initial products are Sn (0) and SnO (+2). As the oxidation proceeds, the oxidation process is simple: Sn/
SnO → SnO2, during the entire oxidation time. The sputtering shows that as the SnO2, moiety decreases, the Sn
(0) and SnO (+2) species grow. The data show that initially a thin film of SnO2 and PbO2 is formed on the surface.
As the tin diffuses outward, the film becomes oxygen-poor forming some SnO (+2) which combines with PbO2
(+4) to give SnO2 and PbO according to the reaction
SnO (+2) + PbO2 (+4) → SnO2 (+4) + PbO (+2).
Thus, in the oxidation process, a steady-state film is formed consisting mainly of SnO2 and PbO.
6.14.3 Analysis of Gas Mixtures

The most powerful aspects of PES is that the core electron binding energy of an atom is often a significant func-
tion of the chemical environment of the atom in a molecule and the band intensities are proportional to the partial
pressure of the constituent gases of the mixture. Both these aspects make the tool of PES applicable to the analysis
of gas mixtures qualitatively and quantitatively. Only those gases which exhibit at least one sharp and characteris-
tic peak in the PES are most suitable for quantitative analysis. Moreover, such a peak should dominate the peaks
overlapping it. The following gases qualify the above characteristics: N2, O2, CO, CO2, N2O , CS2, COS, NO, H2S,
C2H2 and all atomic species.The sample of COS has been quantitatively analysed for CS2 and CS2 impurities. The
analysis of a mixture of gases by PES has the advantage over low-resolution mass spectrometry in cases where
the PES spectra are different while the mass spectra are almost the same, e.g. N2–CO, N2O–CO2, CO–CO2 pairs.
PES is also preferred for compounds where mass spectra contain no molecular ion peak but only fragment ions.
Moreover, for analyses of gases in the upper atmosphere and in space, the photoelectron spectrometer is preferred
over the mass spectrometer because of its volume and lighter weight.
The electronic structure of free radicals and other transient moieties is useful to understand the mechanism of
kinetically controlled reactions. The ionisation energies and ionic excited states of radicals are directly relevent to
mass spectrometry and to the study of ionic decomposition, for which knowledge of the energy levels of fragment
ions are requried. Photoelectron spectroscopy can impart the necessary information provided the transients can
be produced in sufficient amounts to yield a signal when ionised in the spectrometer source. Transients produced
by microwave discharge on high-temperature pyrolysis in a flowing gas stream exhibiting PES are H, N, O, F,
Cl, Br (monoatomics); SO, CS, P2 (diatomics); NF2, ClO2, S2O (triatomics) and larger molecules. PES of stable
molecules in excited states can also be observed, e.g. oxygen in its first electronically excited state lΔg.
Problem 6.4: Calculate the thickness of a deposited layer on some virgin surface if the intensity of the photo-
electron signal due to the virgin surface is reduced by the covering layer by 5 per cent λ = 20 Å.
Solution In order to determine the thickness of the covering layer on the virgin surface, use is made of
Eq. (6.39) i.e.
−t / λ
0e (Eq. 6.39)
Substituting the values of the given parameters into Eq. (6.39), we get.
−t
95
= e 20(Å)
100
Taking log on both sides and solving for ‘t’, we get
−t
ln 0.95 = lne (=1)
20(Å )
or t = +5.129 × 10-2 × 20 (Å) = 1.025 Å
Problem 6.5: The photoelectron spectrum of CO and CO2 produced from a gas mixture of CO and CO2 is
characterised by the presence of the peaks of the Cls electrons of CO and CO2 with ionisation potentials of 296
eV and 298.4 eV respectively. Determine the percentage composition of the mixture if the intensity ratio of the
bands corresponding to CO and CO2 is 2:3.
Solution The content of CO in the binary gas mixture is determined by the formula:
I CO × 100
x=
I CO I CO2
where, ICO and I CO2 are the band intensities of the CO and CO2 being determined in the PES.
2
Thus, x=× 100 = 40 per cent
5
Similarly, the content of CO2 in the mixture is given by the formula
100 3
y= = × 100 = 60 per cent
RCO2 + 1 5
6.14.4 Molecular Orbital Models Z

PES provides a tool to develop qualitative molecular orbital models σyz
and vice versa which helps not only to understand the electronic struc-
ture of the molecules but also the other molecular properties of the
N i N X
molecules. This is demonstrated by taking the example of the nitrogen
C2(x)
molecule. The possible combination of s and p orbitals in N2 can be
characterised according to their behaviour with respect to the three
symmetry elements of the molecule: an axis of rotation (C2), a reflec-
Y
tion plane (syz) and centre of inversion (i) (Fig. 6.24 Table 6.6). The Fig. 6.24 Symmetry elements of N2 molecule
interaction or mixing of two orbitals is governed by their relative ener- in relation to the characterisation of the possible
gies. In general, when there is a large energy difference between the combination of s and p orbitals in the molecule.
two orbitals, the interaction between them decreases, for example in the argon atom, for a given primary quantum
number n, E (ns) > E(np).
Accordingly, in the first approximation for N2, only mixing between two similar orbitals is considered, i.e. 2s
with 2s and 2p with 2p (Fig. 6.25). The ls electrons of nitrogen are tightly bound to their respective nuclei, and
hence their contracted orbitals do not overlap effectively and they remain almost localised. The resulting first
orbital scheme, filled with electrons according to Aufbau and Pauli exclusion principles is compared with the PES
spectrum of N2 in Fig. 6.26.
π*g σ*u σ*u
(a)
πu σg σg
(b)
Fig. 6.25 Ist order mixing between two similar orbitais: (a) antibondlng and (b) bonding configurations.
Nitrogen
Molecule
Nitrogen 2nd
Atoms 1st Approx
Approx
σ*u Photoelectron Spectrum
π*g
cps
2p
15
2091 cm−7
σg
16
πu
1850 cm−1
πu
17
σg
18
2397 cm−1
σ*u
19
σ*u
20
2s
σg
36
37
σg
38
409
= =
410
1s 1s
411 IE(eV)
Fig. 6.26 Schematic diagram showing the observed photoelectron spectrum of N2 molecule in
relation to the 1st and IInd order molecular orbitals.
The photoelectron spectrum of N2 from various energy sources exhinbit five bands expected for 14 electrons
of the nitrogen molecule, assuming degenerate 1s and pu orbitals. Besides ionisation potentials, other information
can be gathered from the spectrum. The shape of the band imparts information regarding the bonding character
of molecular orbital:
Table 6.6 Symmetry elements of N2 in relation to characterization of bonding character of
orbitals.
C2 (x) syz i
Symmetric s Percent (bonding) g (gerade)

Antisymmetric p *(antibonding) u (ungerade)
Strong, sharp spikes followed by a few other weak bands characterise the nonbonding electrons whose orbitals
are localised, e.g. the low-energy bands in the 15–20 eV reigon, the broad bands whose maxima occur towards the
centre of the band characterise both the very strongly bonding and antibonding electrons which are dclocaliscd.
Moreover, the bands in the PES spectrum may exhibit a resolvable fine structure, the spacing in which corresponds
to particular vibrational modes of the molecular cation. Ionisation from a nonbonding orbital does not affect the
vibrational frequency very much but from a bonding orbital, decrease the frequency and from an antibonding
orbital, increase the frequency. A comparison of the vibrational frequencies manifested in the ionisation spectral
bands with that of the fundamental vibrational mode at 2345 cm−1 of the neutral ground state of N2 molecule sug-
gests that the first, second and third ionisation energy bands of +2 molecular ion correspond to very slightly
bonding, strongly bonding and slightly antibonding orbitals respectively. However, this deduction does not con-
form to the first-order molecular orbital scheme according to which out of the three highest occupied orbitals, two
have significantly bonding character. Accordingly, the first-order molecular orbital scheme needs to be modified
in order to explain the experimental facts. This is done by taking into consideration the idea that orbitals of the
same symmetry interact so as to increase the difference in their energies. Based on the idea, in the second-order
molecular orbital scheme, interaction between the two occupied sg orbitals, i.e. lsg and 2sg is taken into account
and is shown in Fig. 6.27. The good agreement between the second-order molecular orbital and observed experi-
mental results confirms the rule: two molecular orbitals of like symmetry admix in such a way that the low-energy
orbital of the first (less stable), i.e. 1sg interacts with the high-energy orbital of the second (more stable) i.e. 2sg
in an antibonding destabilising fashion, i.e. (2sg − 1sg) while the admixture of type 2sg + 1sg results in a stable
configuration as shown in Fig. 6.27.
(2σg−1σg)
2σg
(a)
1σg
(1σg+2σg)
(b)
Fig. 6.27 Second-order interaction between two occupied sg orbitals of
like symmetry (a) antibonding, and (b) bonding configurations.
To summarise, we arrive at the following deductions. The photoelectron spectrum provides data on ionisation energies,
band shapes and vibrational frequencies. The experimentally observed data can satisfactorily be explained with a quali-
tative molecular orbital scheme. The experimental data in conjunction with molecular orbital scheme enables to develop
a model which satisfactorily interprets the photoelectron spectrum. This model can further be applied to explain the
other molecular properties of nitrogen, e.g. the only small increase in bond length on ionisation (bond length in N2 is
1.09 Å while in N +2 is 1.11 Å). The band at 410 eV in the ESCA region consists of only one peak. This suggests that
the inner shell 1s orbitals fail to interact and only the valence electrons need be considered. Vibrational fine structure
imparts information regarding the nature of the molecular orbital and the relationship between nature of peak and the
orbital involved in ionisation is given here in Table 6.7.
Such studies have also been carried out successfully on various types of polyatomic molecules such as H2O,
H2S, C6H6, etc.
Table 6.7 Deductions based on vibrational fine structure of the nature of molecular orbital involved in ionisation.
Nature of peak Orbital involved in ionisation Remarks

Sharp peak without any vibrational Nonbonding or very weakly bonding re is not effected Iadiabatic = Ivertical i.e no vibrational
fine structure changes
Sharp peak with number of vibrational Bonding or antibonding Decrease in frequency (bonding) or Increase in
structures frequency (antibonding)
Broad band Very strongly bonding or antibonding Dissociatioin or predissociation
6.14.5 Anharmonicity Constant and Changes in Molecular

Geometry on Ionisation
The position of the photoelectron band indicates the energy of an ionic state whereas the vibrational structure of
the band gives information about the structure and bonding of the ions in that state. When the coarse vibrational
structure manifested in an ionisation band is resolvable, the identities and frequencies of the modes excited on
ionisation and the intensities of the vibrational lines can be examined. All of these can be used to make direct
deductions about the nature of the ionic state produced, about the differences between the ionic state and molecu-
lar ground state from which ionisation occurs and the changes in molecular geometry.
(a) Intensity of the Vibrational Iines in an Ionisation Band To the relative intensity of a vibrational transi-
tion in an ionisation band is given by Franck–Condon Factor (FCF) which is equal to the square of the overlap
integral between the vibrational wave functions ΨV′ and ΨV ′′
FCF = |< ΨV ′′ .| ΨV′ >|2 (6.43)
Here, V′denotes the vibrational level in the neutral meleeule, i.e. the target molecule and V″ denotes the vibra-
tional level in the molecular ion produced on ionisation of the neutral molecule. The deductions regarding the
intensities of vibrational lines are made from Eq. (6.43) and are reconded in Table 6.8.
Table 6.8 Deductions regarding the intensities of vibrational lines based on Franch-condon Factor.
Change in bond lengths or Values of vibrational Ortho-gonality of wave Overlap integral Intensity
force constants on ionization quantum numbers in functions
i.e. change in PE surface both the levels
No change Different (V′≠ V″) Orthogonal Zero Zero
No change Same V′= V″ Orthogonal Zero Zero
Little change Same (V′= V″) Orthogonal Large value Strong
Different (V′≠ V″) Orthogonal Small value Weak
In PES, the target molecules are in their vibrational ground states so the adiabatic transitions are most intense.
However, if there is a change in bond length on ionisation so that the equilibrium position of the nuclei are dif-
ferent then the vibrational mode or modes that corresponds most closely to the change in nuclear position are
most strongly excited. This is the basis of relationship between the localisation of bonding character of particular
electrons and the structure of the ionisation bands.
When two or more vibrational modes are strongly excited in a transition, the band structure consists of pro-
gressions of progressions, i.e. each line corresponding to the excitation of a number of quanta of one mode is the
starting point for a progression in the other band. Progressions of progressions in the PES of carbonyl fluoride
(F2CO) are shown in Fig. 6.28.
(b) Selection Rules The selection rules for vibrational modes which can be excited on ionisation are
summarised in Table 6.9. The selection rules for degenerate vibrations are the same as those for antisymmetric
vibrations but only weak transitions in units of two quanta are allowed. For certain point groups, the odd quantal
states of degenerate vibrations are also antisymmetric with respect to at least one element of symmetry and rule
ΔV = 0, ±2, ±4, ±6 is strictly obeyed. In other instances, transitions with ΔV = ±3, ±5, ±7........... may be weakly
allowed, but ΔV = ±1 is strictly forbidden. Double quantal excitation of such a degenerate vibration is observed
in the PES of ethylene.
13 14 15
(a)
(b)
Fig. 6.28 (a) Photoelectron spectrum of carbonyl fluoride (13−16 eV) showing progressions
of progressions (b) Vibrational analysis.
Table 6.9 Selection rules for vibrational modes.
Symmetr of vibrational Symmetry of vibrational Vibrational quantum Selection rule Intensity

modes wave functions number or quantal
state
Symmetric Symmetric Even ΔV = 0, ±1, —

or odd ±2, ±3…
Antisymmetric Antisymmetric Odd ΔV = 0, I0>>I±2,

(CO2’ CS2) Symmetric Even ±2, ±3 I±4,I±6…
since maxima of
vibrational wave
functions lie in a
symmetrical posi-
tion directly above
the vibrationless
ground state on a
F-C diagram
(c) Determination of Anharmonicity Constant When a long vibrational progression is observed in an

ionisation band, the spacing between two adjacent lines, i.e. DG is measured on the vibrational levels V = 1, 2,
3, 4 … in order to determine the anhannonicity constant xii (See Chapter 3 on infrared spectroscopy).
The energy of a vibrational level above the vibrationless ground state (V = 0) for excitation of a single
vibrator ‘i’can be written as
E (V ) = hc viVi + hc xii viVi2 (6.44)
where, vi is the wave number in cm−1 and Vi is the vibrational quantum number.
The spacing between two adjacent lines can be worked out as follows:
Vi = hcv
EV h viVi + hcx
c ii viVi2
EV
Vi 1 = hc
h vi (Vi + 1) + hc
h xii vi (Vi + 1) 2
For the spacing between two adjacent lines, we write
EV
Vi − EV
Vi
ΔG = 1
= vi + xii vi (2Vi − 1)
hc
or ΔG vi ( − xii ) + 2 xii viVi (6.45)
The plot between ΔG and Vi will be a straight line with slope of 2xii vi = m and intercept of vi (1 -xii) = Iint. From
the slope and intercept, one can determine the anharmonicity constant,
m
i.e. xii = (6.46)
2 I int m
When the spacing, i.e. the vibrational interval decreases with the increasing Vi, the anharmonicity is said to
+
be positive, e.g. progression in X 2 ∑ g band of H +2 . On the other hand, when the spacing increases with the
increasing Vi, the anharmonicity is said to be negative, although the anharmonicity constant xii is positive. Actu-
ally the nomenclature originates from Franck–Condon principle. The potential energy curves for most diatomic
molecules lead to negative values of xii, indicating thereby that the vibrational levels converge as they approach
the dissociation limit which is normally expected, and thus xii is positive. Negative anharmonicity is observed
most commonly in progressions due to bending vibrations since in such vibrations the restoring force instead of
decreasing with increasing amplitude, rather increases by steric interactions when the vibrating groups approach
each other, e.g. second band in the spectra of H2O, H2S, H2Se, H2Te, etc., and the the first bands in the spectra of
NH3 and related molecules show the same effect.
Note that the rotational fine structure can also appear in the ionisation band if the resolving power of the cur-
rent instruments could be increased 10–100 fold, which of course seems to be hard. However, pure rotational fine
structure has been observed in the PES of H2 recorded on an instrument having a resolution of 4 meV. Note that
hydrogen is a special case not because of its large rotational intervals but also because both H2 and H +2 exist in
ortho and para states in which the spins of the two nuclei are parallel and antiparallel respectively. The selection
rule of electronic transitions (spins are not affected in electronic transitions) Δ J = 0, ±2, ±4 and a related rule
are also applicable to photoionisation of any molecule with identical nuclei in equivalent positions, e.g. O2, CO2,
C2H2 and NH3. Photoionisation of He (I) line showed that transition with ΔJ = 0 is much more intense than those
with ΔJ = 2.
Problem 6.6: When the hydrogen molecule is excited by the helium resonance line (21.22 eV), it undergoes
the following processes:
H + + e − (H +2 ion in ground vibration state)

hv
H2
H + vib + e − (H +2 ion in excited vibration state)

hv ib
H2
+ +
The PES spectrum corresponding to the H2 1 ∑ g ( ′= ) → H +2 2 ∑ g (V≤ = 0, 1, 2, 3 …) transition exhibits a
long vibrational progression given below:
V≤ 0 1 2 3 4 5 6 7 8 9
IP (eV) 15.45 15.72 15.98 16.21 16.43 16.63 16.83 17.01 17.17 17.32
Calculate (a) the anhormonicity constant for H +2 ion, (b) compare the force constant of H +2 ion with that of the
H2 molecule, and (c) the lifetime of the H +2 ion if the width of the lines is about 0.1 eV.
Solution Proceeding as in Problem 3.29 of Chapter 3 on infrared spectroscopy, the average value of −Δ2G
comes out to be 0.015 eV (= 0.015 × 8065.7 cm−1 = 120.98 cm−1).
120.98 −1
Thus, we xe = w0 x0 = cm = 60.49 cm −1
2
From Eq. (3.53) for V =1, we obtain
−1
ΔG
G⎛ 1⎞ w0 − w0 x0 − 2 w0 x0 × 1
V+
⎝ 2⎠
2177.74 cm−1 = w0 − 60.49 cm−1 − 2 × 60.49 cm−1

Rearranging, we get
w0 = (2177.74 + 60.49 + 120.98) cm−1 = 2359.21 cm−1
From Eq. (3.36), i.e.
w0 = we − wexe’
we write 2359.21= we – 60.49

Thus, we = 2455.70 cm−1
Since wexe = w0x0
w0 x0 60.49 cm −1
Therefore, xe = = − 1
= 24.63 × 10 −3
we 2455.70 cm
we 2455.70 cm −1
and De = = = 24926 cm −1 = 298 kJ/mole
4 xe 4 × 24 63 × 10 −3
For H2 molecule; we 4400 cm −1 xe = 27.4 × 10 −3 , and De = 435.9 kJ/mole

(b) In order to compare the force constant of the hydrogen molecule ion with that of the hydrogen molecule,
we make use of Eq. (3.19).
+
+ k(( 2)
Thus, we( 2 ) = 1303.16 +
μm ( 2)
k(( 2)
we( 2 ) = 1303.16
μm ( 2)
+
Since μm ( ) μm ( 2)
+ +
we( 2) k(( 2)
Therefore, =
we( 2) k(H 2 )
2 2
⎡ we (H +2 ) ⎤ ⎡ 2455.70 cm-1 ⎤
k(H +2 )
or =⎣ ⎦ = ⎣ ⎦ = 0 31
k(H 2 ) ⎡ w (H ) ⎤ 2 ⎡ 4400 cm-1
- ⎤ 2
⎣ e 2 ⎦ ⎣ ⎦
or k(H +2 ) 0.3 k(H 2 )
k(H +2 )
or k(H 2 ) = = 3.. k (H +2 )
0.31
i.e. the force constant of the H2 molecule is 3.22 times large as compare to that of the hydrogen molecule ion, H +2
(c) We know that
h 6.577 × 10 −16 (eV s)
Γ= =
2πτ τ
Since Γ = 0.1 eV
6.577 × 10 −16 (eV s)

Therefore, the lifetime of H +2 ion τ = = 6.5 × 10−15 s, i.e. of the order of 10−14 s.
0.1eV
Problem 6.7: The first ionisation band (i.e. 2Al) in the PES of NH3 exhibits a long progression as shown Fig. 6.29.
The approximate values of ionisation potentials (i.e. vibrational frequencies) on quantum number V″ = 0, 1, 2, 3 ... 13
as measured from Fig. 6.29 are listed in Table 6.10.
Solution In order to determine the anharmonicity constant, the vibrational intervals ΔG on different vibra-
tional quantum numbers V″= 1, 2, 3… 13 need to be determined first. Now by making use of Eq. (3.53), ΔG for
different values of V″ have been calculated and are listed here in Table 6.11.
2A
1
13 12 11 10

Fig. 6.29 Photoelectron spectrum of NH3, showing negative anharmonicity in the 2A1 band.
Table 6.10 Ionisation potentials from the

photoelectron specturm of NH3 .
Table 6.11 ΔG for different values of V≤.
V″ IP (eV)
V″ IP(eV) DG(meV)
0 10.110
0 10.110
1 10.220
1 10.220 110
2 10.338
2 10.338 118
3 10.456
3 10.456 118
4 10.575
4 10.575 119
5 10.695 5 10.695 120
6 10.816 6 10.816 121
7 10.940 7 10.940 124
8 11.066 8 11.066 126
9 11.194 9 11.194 128
10 11.323 10 11.323 129
11 11.452 11 11.452 129
12 11.584 12 11.584 132
13 11.718 13 11.718 134
Solution The increase in the vibrational interval ΔG at high vibrational quantum number indicates negative
anharmonicity in the vibrational progression of the 2Al ionisation band in the photoelectron spectrum of ammonia.
The anharmonicity constant xii is then determined from the plot of ΔG against V ″ shown in Fig. 6.30.
Here, Intercept Iint = 113 meV and slope m = 1.65 meV
Using Eq. (6.46), we get
= 7 25 × 10 −3
m 1.65( meV) 1 65
xii = = =
2 I int m 2 × 113( ) + 1 65
6 meV 227.65
150
140
ΔG (MeV)
130
120
110
100
0 2 4 6 8 10 12 14
V″
Fig. 6.30 Plot of vibrational intervals in the 2A1 ionisation band of NH3 against vibrational quantum number.
(d) Determination of Geometrical Changes in Molecules on Ionisation From the measured vibrational
frequencies in a particular ionisation band in different states of the molecule or ion, one can estimate the force
constants while from the intensity distributions in the vibrational progression’, the changes in molecular geom-
etry, i.e. the changes in the equilibrium bond length or bond angle can be determined. It is simple to carry out such
studies in diatomics since the normal modes of vibration in them are not localised. But it is not so in polyatomics
where most of the normal modes of vibrations are not localised. However, the stretching modes of diatomics,
totally stretching vibration of linear BAB, tetrahedral AB4, octahedral AB6, square planar AB4 or planar AB3 mol-
ecules qualify for the above studies. Bending vibrations of nonlinear BAB and the umbrella bending vibration of
pyramidal AB3 molecules are also totally symmetric.
Accodring to the Franck–Condon principle, the vibrational wave functions for high vibrational quantum num-
bers (V″) have their greatest amplitude near the PE curve itself, which gives the position of the turning points
of the classical vibrational motion. The largest Franck–Condon factor for a transition from the ground state and
therefore the maximum intensity in a vibrational progression, occurs when the turning point in the ionic state V″
comes at the same bond distance or the angle as the equilibrium position in the ground state of the molecule (V′).
The energy of the classical vibrator that represents vibrations in the ion is proportional to the square of the vibra-
tional amplitude at the turning of the classical vibrational motion can be set equal to the experimental vibration
energy in the ion at the vertical ionisation potential.
For a stretching vibration in a diatomic molecule.
⎛ 1⎞
2π 2 μ m v 2o ( r ) 2 = V " a hhvv o I vertt − I ad (6.47)
⎝ 2⎠
"
where, vion is the vibrational frequency in the ion, μm is the reduced mass appropriate for the vibration, Vmax is the
vibrational quantum number at the maximum intensity in the band and Δr is the difference in equilibrium bond
distance between the ion and molecule, i.e. Δr = re (ion) – re (molecule). The same equation is valid for a bending
vibration also.
For a bending vibration, Δr = l dq (6.48)
Here, l is the bond length and dq is the change in equilibrium bond angle.
Thus, the Eq. 6.47) can be written as
( ad )
( )2 ( ) 2 = 5.439 × 105 × vert (6.49)
μmνion
2
Here, Δr and l are in Å, dq in radians, Ivert and Iad in eV, μm in atomic mass units and νion in cm−1.
For BAB triatomic molecules, the energy of the classical vibrator is expressed in terms of the AB bond
length is 4p2 μm νion
2
(ΔrAB)2. In such cases, Eq. (6.48) becomes
( ad )
( )2 ( ) 2 = 2.719 × 105 × vert (6.50)
μmνion
2
The equation holds good only for large vibrational quantum numbers, i.e. when there is a large change in shape
and hence long vibrational progressions are observed.
A more general method to estimate the changes in molecular geometry on ionisation is due to Heilbronner,
Muszkat and Schaublin, i.e. HMS method. This is based on calculated Franck–Condon factors for transitions
between two harmonic vibrators which can be expressed in the analytical form.
The geometrical changes in some diatomic and triatomic molecules determined by different methods are compared
in Table 6.12.
Table 6.12 Geometrical change in some diatomic and triatomic molecules.

Molecule Orbital ionised Ionic state Change in bond length on ionisation (Å)
(a) (b) (c)

H2 sg (b) X2∑+g 0.33 0.35 0.32
O2 pu (b) a Pu
4
0.17 0.20 0.19
pg (ab) X2Pg −0.08 −0.09 −0.08
CO2 pg (nb) X2Pg 0.015 0.015 ⎯
pu (b) A Pu
2
0.066 0.074 0.062
su (b) B Pu
2
0.018 0.015 ⎯
CS2 pu (b) A2Pu ⎯ 0.08 0.07
(a) From microwave spectroscopy, (b) From PES by HMS method (c) from (6.49)
Problem 6.8: Calculate the change in bond length on ionisation of H2 molecule if the difference between the
2
+
vertical and adiabatic ionisation potentials is 0.57 eV and the vibrational frequency of the H 2+ ion (in ∑g state)
and that of the H2 molecule is 2455.70 and 4400 cm−1 respectively. The bond distance of the molecule is 0.741 Å.
Solution Substituting the given values into Eq. (6.49), we get

0.. (eV ) 2
( ) 2 = 5.439 × 105 × = 0.103 Å 2
( .70 cm −1 ) 2
or Δr = 0.32 Å
Since the vibrational frequency of the H2 molecule decreases on ionisation, the bond length of the H +2 ion will
+
be large as compared to that of H2 molecule. Thus r(H +2 2
∑ g ) = (0.741 + 0.32) Å = 1.061 Å.
Problem 6.9: Calculate the change in bond length on ionisation of O2 molecule (O2, X3 O +2 , 2 Πg), the g
difference between the vertical and adiabatic ionisation potentials is 0.32 eV and the vibrational frequency of the
O +2 ion (in 2Pg ionic state) is 1850 cm−1. Compare the bond distance so determined with that of the O2 molecule
(1.21 Å) if the vibrational frequency of the O2 molecule is 1580 cm−1.
Solution Substituting the given values into Eq. (6.49), we obtain

0.. ( V ) 32
( ) 2 = 5.439 × 105 × 1 2
= 6.356 × 10 −3 Å 2
( ) × 16 × 16
or Δr = 0.08 Å
Since the vibrational frequency of the O2 molecule increases on ionsation, the bond length of the O + ion will
be small as compared to that of the oxygen molecule. Thus, r ( O +2 2Pg) = (1.21 − 0.08) Å =1.13 Å.
2
6.14.6 Molecular Charge Distribution

The ionisation potentials of inner shell electrons vary slightly with the chemical environment of the atoms in a
molecule. Thus, the chemical shifts are due to changes in the electrical potentials of the inner electrons caused by
charges within the molecules. The changes in potential are defined as
q δq
ΔU = +∑ i (6.51)
r i Ri
where Ri is the distance from the atom of interest to the ith other atom carrying charge δ qi . The atomic charge
q can be computed provided that the constant ( 1r ) is known and the Madelung potential ΔU can be estimated.
Here, the ionisation potential shift is equated with the Madelung potential. Since the ionisation potential shifts
in molecules with respect to some standard can be measured, the charge distribution in
molecules can be accomplished by photoelectron spectroscopy, e.g. a comparison of Ols +
O
ionisation shift of the ring oxygen in flavylium compounds which are usually shown as
under (I) with that of known oxygen groups showed that oxygen is slightly negative and
the charge is strongly localised. This demonstrates the power of PES for probing the charge
on individual atoms in the complex molecules.
(I)
The photoelectron spectra of HgBr2 exhibit 5d−1 ionisation bands (above 15 eV) at 16.4
(sharp),16.8 and 18.3 eV while that of (CH3)2 Hg at 15.0 (sharp),15.4 and 16.9 eV. These
bands are assigned to Hg 5d –1 ionisations, since they appear in the spectrum of every mercury compound, the
relative spacing of the three bands remain almost constant i.e. ≈ 0.4 and 1.5 eV, the spacing of the two outer bands
is ≈1.5 eV which is close to 1.86 eV, the spin-orbit splitting of the 2D state of Hg+.
The shift between the sharp (J = 5/2) peak and the 2D5/2 ionisation potential of atomic mercury is a direct mea-
sure of the difference between mercury atom in a molecule and a free mercury atom. Thus the charge on mercury
atom in a molecule can be calculated from Eq. (6.51) provided the constant 1 is known. The constant is equal to
r
the change in ionisation potential of the d shell produced by unit positive charge in the outer 6s shell, and it can be
obtained from the atomic spectrum of mercury. The atomic charges from 5d−1 ionisation potentials in some typical
mercury compounds (II) are given here:
Compound HgCl2 HgBr2 CH3HgCl CH3HgBr (CH3)2Hg (C2H5)2Hg
Shift from 2D5/2 (eV) 1.87 1.56 0.95 0.84 0.11 –0.16
Charge on Hg (units of e) 0.425 0.332 0.218 0.182 0.028 –0.041
The negative charge on mercury in case of (C2H5)2 Hg shows that the electron-attracting power of mercury
atom is intermediate between that of the methyl and the ethyl groups.
−0.2125 −0.166
Cl Br
+0.425 +0.332
Hg Hg
Cl Br
−0.2125 −0.166
−0.014 +0.0205
CH3 C2H5
+0.028 −0.041
Hg Hg
CH3 C2H5
−0.014 +0.0205
6.14.7 Structural Elucidation

Since the photoelectron signal due to the atom of an element in a molecule is characteristic of its electronic
environment, the PES lines due to atoms of the same element in a molecule will be different and discernible,
consequently this can be fixed as criterion for resolving the
O
structural problems in organic and inorganic molecules. More-
over, correlation tables could help in the final selection of Cy S S Cy Cy S S Cy
functional groups, e.g. N ls ESCA spectrum of sodium azide
(NaN3) shows two peaks with relative areas under the peaks in O O O
the ratio 2:1.This suggests that there are two types of nitrogens Thiosulphonate structure Disulphoxide structure
in sodium azide. The two similar-end nitrogens are different
from the central nitrogen. Thus, ESCA spectra can also impart (I) (II)
information about the number of each kind of atom in a mol-

ecule. Doubly oxidised cystine (CySO2SCy) can either have thiosulphonate (I ) or disulphoxide structure (II).
The thiosulphonate structure is expected to give two signals of equal intensity corresponding to the two oxida-
tion states of S atom in the molecule while only one peak is expected in the disulphoxide structure since both the
S atoms in the molecule exists in the same oxidation state. S (2p) ESCA spectra of cystine and doubly oxidised
cystine shown in Fig. 6.31 suggest that oxidised cystine has a thiosulphohate structure (I) since two signals corre-
sponding to two oxidation states of S are observed at 162 and 166 eV. The former corresponds to the low oxidation
while the latter to the high oxidation states of sulphur since
the S (2p) signal in Cy SS Cy is observed at 164 eV.
The oxidation states of central metal in oxides like W3V5O20,
WV2O6 can also be established by comparing ESCA spectra
of these compounds with their simple oxides. It is found that
in the former, V is in +5 and +4 oxidation states while in the
latter, V is in +5, +4 and +3 oxidation states.
intensity (counts/sec)
(a)
Further, the change in electronic structure takes place in
going from a linear molecule to a bent molecule since the
degenerate orbitals in the former become nondegenerate
in the latter. Thus, the number of photoelectron bands will
increase and their shapes will also change markedly. This is
attributed to the excitation of bending vibrations in ionisa- (b)
tion from the orbitals which are angle determining in a bent
molecule. This can be fixed a criterion to resolve any doubt
about the linearity of a molecule. Photoelectron spectrum is
also useful to resolve the problem regarding the planar and
nonplanar nature of the molecules, e.g. butadiene and 1, 1, 170 165 160 155 150
4, 4-tetrafluorobutadiene are planar molecules. The two P −1 Binding Energy (eV)
and one s−1 bands of butadiene appear at 9, 11.5 eV and 12.2 Fig. 6.31 S (2p) ESCA spectra of (a) doubly oxidised cystine,
eV respectively while at 10.5, 11.5 eV and 14.5 eV in case of and (b) cystine.
pentafluorobutadiene. Thus the shift of s−1 bands to higher
ionisation potential is greater as compared to P−1 bands due to perfluro effect. The strong band near 17 eV in the
spectrum of pentafluoro compound is due ionisation of the fluorine lone-pair electrons. The difference in energy
between the first (outermost p-orbital) and the second ionisation bands is above 2 eV in case of butadiene and
1, 1, 4, 4-tetraflurobutadiene. However, this difference is about (11.5−10.5) eV ≈ 1 eV in pentafluorobutadiene. This
suggests a nonplanar nature of the molecule since in a bent molecule, conjugation across the central bond will be less.
Consequently, ionisation-energy difference between two p-orbital bands will also be reduced. Optical absorption data
in conjunction with photoelectron spectral data suggest that pentafluorobutadiene has a cis-nonplanar configuration
having a dihedral angle of 42 degrees. This conclusion conforms to the x-ray diffraction studies on the molecule.
The partial x-ray photoelectron spectrum of ethylchloroformate is shown in Fig. 6.32. The presence of C and Cl
is established by the electron peaks at 284 and 270 eV respectively. Similarly, the presence of O could be established
by electron peak at 532 eV. The chemical shifts apparent in the Fig. 6.32 give evidence for the presence for C in
different valence states. The C (1s) peak intensity ratio 1:1:1 indicates that the compound contains equal number
of carbon atoms in three different groups. Since the XPS chemical shifts is governed by the electron-withdrawing
power of atoms attached to the atom whose chemical shift is under consideration, the peak at 284 eV is attributed to
O
C(ls) CI C O in groups and the following two peaks to C (ls) in ⎯O⎯CH2 and ⎯CH3 groups respectively.
6.14.8 Nature of Chemical Bonding

Based on the photoelectron spectral data on pentacarbonyl manganese derivatives, it has been accomplished that
the BE of pure ligand orbitals >the metal to ligand bonding orbitals >the metal d-orbitals, i.e complexation stabi-
lises both the ligand s-orbitals and the metal d-orbitals as compared to their energies in the free ligand molecule
and the metal atom as expected. This rule aids in interpreting the photoelectron spectra and conforms to the fact
that metal d-orbitals determine many of the chemical properties of the complexes.
Molecule BE(eV)
PF3 12.2, 16.2, 17.7, 19.7

Mo(PF3)6 9.2, 12.4, 13.5, 14.5, 16.2, 17.7, 19.7
Mo(CO)6 8.4, 13.2, 14.8
Let us now examine the HeI photoelec- O H H

tron spectral data of Mo (PF3)6 and Mo Cl C O C C H
(CO)6 given below in relation to the nature
of chemical bonding in these compounds H H
in the light of this rule.
The lowest ionisation energy band in C 1s
1000
the PES of Mo (PF3)6 appears at 9.2 eV
while an 8.4 eV in the PES spectrum of Mo
Intensity(counts/sec)
(CO)6. The shift of 0.8 eV on the change

of ligand from PF3 to CO is attributed Cl 2s
to the ionisation from metal d-orbitals.

In octahedral low-spin complexes, the
six d-orbitals reside in t2g orbitals, so the 500
d-electron ionisation bands are single.
Ionisation potential of the PF3 complex
is greater than that of the CO complex
showing thereby that PF3 causes greater sta-
bilisation compared with the CO ligand and
it is a better p-acceptor than CO. The PES 0
spectra of PF3 and Mo (PF3)6 are almost
identical in the ionisation region of 15−20 1190 1200 1210
Kinetic energy (eV)
eV suggesting thereby that the fluorine
lone-pair orbitals or the P−F bonding orbit- 290 280 270
als are weakly affected by complexation. Binding energy(eV)
The lowest-energy ionisation band of PF3 Fig. 6.32 Partial photoelectron spectrum of ethylchloroformate.
at 12.2 eV which disappears in the complex
is attributed to the ionisation from the lone pair of electrons of P atom, the orbital from which s-donation occurs. In
octahedral symmetry, the six ligand lone-pair orbitals become tlu, eg and a1g, molecular orbitals, and are stabilised by
the empty Mo p, d or s orbitals respectively on complexation. The three bands at 12.4, 13.5 and 14.5 eV having area
ratio of 3:2:1 in the PES spectrum of Mo(PF3)6 correspond to ionisation from tlu, eg and a1g orbitals respectively as
expected. All these three orbitals are stabilised compared with lone-pair orbitals of free PF3 .In case of Mo(CO)6, the
band at 13.2 eV corresponds to tlu orbital while at 14.8 eV is from orbitals located in the CO molecule (bands due
to uncomplexed CO orbitals fall in the range of ≈ 14−20 eV). The bands due to other s-bonding orbitals, i.e. eg, a1g
might have overlapped with the bands from orbitals located in the CO molecule.
Let us now correlate the photoelectron spectra of Ni(PF3)4: Ionisation energy bands [(9.6, 10.6), 12.3, 13.1,
14.5, 15.8, 17.4, 19.3 eV)] and Ni (CO)4 [(8.9, 9.8), 14−20 eV] in order to understand the nature of chemical bond
in these complexes. Both these tetrahedral complexes have completely filled t2 and e orbitals as nickel in these
complexes is in zero oxidation state. The lowest ionisation energy band in both the complexes appear as a doublet,
the bands at 9.8 and 8.9 eV are the components of the doublet of CO complex while at 10.6 and 9.6 eV are the
components of doublet of PF3 complex. The ratio of areas of bands constituting a doublet in both the complexes
is 3:2 as expected for t2−1 and e−1 ionisations. The splitting between the 2T2 and 2E states which is a measure of the
strength of the ligand field is almost the same, i.e. about l eV in both the complexes but the ionisation potential of
the PF3 complexes (10.6 eV, 9.6 eV) is higher than that of CO complex (9.8 eV, 8.9 eV). This indicates that though
ligand-field strength in both these complexes is about the same but PF3 is a better p-acceptor as compared to CO
ligand, i.e. PF3 causes greater stabilisation than CO. Just like the low-spin octahedral complex Mo (PF3)6, the PES
spectra PF3 and Ni (PF3)4 are almost identical in the ionisation reigon of 15−20 eV indicating that the P–F bonding
orbitals are weakly affected by complex formation. The lowest ionisation band of PF3 at 12.2 eV is attributed to
the ionisation from the lone pair s-donating orbitals of P-atom which are stabilised in the complex as is evident
from the ionisation bands at 12.3, 13.1 and 14.5 eV.
To summarise, the bonding in these complexes involves the donation of s-electrons from the ligands into empty
metal d orbitals accompanied by back donation from metal d orbitals into empty ligand p orbitals of appropriate
symmetry, the latter, i.e. P ← M p back bonding seems to be the most important for the stability, i.e. neutrality of
the complexes.
He (I) photoelectron spectrum of chloroform exhibits bands at 11.5, 11.9, 12.8, 16 and 17.1 eV whereas its
silicon analogue, i.e. SiHCl3, shows a broadened band (not all) at 11.8, 12.3, 13.1, (14.8 shoulder, 15.2 eV) and
18.1 eV. The ionisation energies from equivalent lone-pair orbitals of SiHCl3 are at higher potentials as to that
of CHC13 while the ionisation bands involving s-orbitals are observed at lower ionisation energies, as expected.
The broadening of some of the bands in going from CHC13 to SiHCI3, i.e. halomethanes to halosilanes, is attrib-
uted to the presence of dp−pp interaction between chlorine and silicon which can stabilise certain orbitals and
increase their bonding character. Si-F bonding in fluorosilanes has less dp−pp character than Si-Cl bonding in
chlorosilanes. These observations conform to theoretical models in which dp−pp interaction is incorporated.
Photoelectron spectra of thiophene and its derivatives, halophosphenes also exhibit dp−pp interaction in the
molecules.
6.14.9 Substituent Effects

PES, like all other spectroscopic methods, provides a tool of measuring substituent effects via ionisation potential
in organic molecules. The high accuracy of measurement of PES ionisation potentials (either adiabatic or vertical)
compared with other spectroscopic methods makes it a better tool to study the substituent effects. The ionisation
potentials of lone-pair electrons may be influenced by the following:
(i) Conjugative (bonded or nonbonded) interaction with other orbitals in the molecule may either raise or lower the ioni-
sation potential. Width of the lone-pair ionisation bands gives an estimate of the strength of conjugative interactions.
Many lone pairs being angle determining, give broad spectral bands so width test is of little practical use to measure
strength of conjugative, effects.
(ii) Spin-orbit coupling in halogen lone-pair ionisations splits the ionisation band symmetrically. Such an effect can be
distinguished from the other two by taking the averge from two peaks.
(iii) Local electric charges and electrostatic effect of the nearby dipoles shifts the body of the ionisation bands. Such effects
constitute the inductive effects. The simple presence of charge in large molecules with nonrigid skeletons may cause
changes in molecular geometry by polarisation effects and consequently, broadening of bands occurs. Thus, it is very
hard to separate the conjugative and inductive effects.
In spite of the limitations of separating out these effects quantitatively, correlations have been made between
ionisation potentials of lone pairs of heteroatoms in homologous series, between ionisation potentials and sub-
stituent constants and between ionisation potentials and electronegativities.
A linear relationship between the ionisation potentials of molecules belonging to several homologous series
and a new substituent parameter μR is given by the equation
PRX = IP
IP PMeX + μ R χ RX
MeX (6.52)
where, χ RX are the sensitivity parameters.
The sensitivity parameters for a series of compounds are obtained from the slope of the plot between lone-pair
ionisation potentials of alkyl iodides (RI) on the X-axis and those of other aliphatic homologous series (RX) on
the Y-axis. Alkyl iodides are used as standard because they exhibit sharp ionisation bands. The slope of RI is taken
as unity. The numerical values of sensitivities of adiabatic ionisation potentials to alkyl substitution are recorded
in Table 6.13 and those of alkyl group substituent parameters, μ R , in Table 6.14.
R–CN–n* – nitrogen lone-pair ionisations and R–CN–p* − n orbital ionisations of the alkyl cyanides.
The sensitivities grow as we move from iodides to bromides and from thiols to alcohols. The sensitivities are
higher for lone pairs on atoms adjacent to the alkyl groups as compared to lone pairs which are farther away from
the alkyl groups. The sensitivity of the p-electron ionisations in cyanides is largest of all and that of n-electron
ionisations in the same compounds is moderate.
Table 6.13 Sensitivities of lone-pair adiabatic ionisation potentials in different homologous series.
Series RX Sensitivity, χRX Ionisation potential of methyl homologue

IP MeX(eV)
RNCS ≈ 0.57 9.25

RNH2 0.66 8.97
RCO. t-Bu 0.84 9.14
RSH 0.94 9.44
RI (standard) 1.00 9.54
HCOOR 1.00 10.81
RNO2 1.00 11.08
RCOMe 1.12 9.71
RCN-n* ≈ 1.2 13.14
RBr 1.25 10.53
RCHO 1.40 10.22
ROH 1.72 10.85
RCN-p* 2.00 12.22
R-CN-n*-nitrogen lone-pair ionisations and R-CN-P*-P orbital ionisations of alkali cyanides
Table 6.14 Substituent parameters for the alkyl-group substituents.

Alkyl group Substituent parameter, mR
Methyl 0.0 (standard)

Ethyl −0.20
n-Propyl −0.29
n-Butyl −0.34
n-Pentyl ~ −0.35
Isopropyl −0.36
Isobutyl −0.38
t-Butyl −0.52
The new substituent parameter, μR reflects the effects of different alkyl groups on the ionisation potentials and
bears a close linear relationship with Tafts’ s * constants, indicating thereby that the alkyl substitution affects the
ionisation potential mainly through inductive effects; the role of other two effects, if any, is negligibly small.
In case of aromatic compounds, the following deductions are based on the PES spectrum of monosubstituted
and para-disubstituted benzenes. The first ionisation band at 9.40 eV in the PES of benzene is due to ionisation
from the two degenerate orbitals, p2 and p3. The introduction of an electron donating group such as —OH, —OR
and—NR2 cause splitting of the first band into two components (i.e. 8.75 eV, 9.45 eV; 8.54 eV, 9.37 eV; and 8.04
eV, 9.11 eV respectively) by conjugative mechanism. The splitting is due to the nondegenerate character of the p2
and p3 orbitals in the substituted benzenes. One of the bands remains near the original position whereas the other
band shifts to the lower ionisation potentials. The shift in the latter case is due to the interaction of substituent with
a new nondegenerate p-orbital withiout a node at the substituent position. The other nondegenerate p-orbital with
a node at the substituent position does not interact with the substituent, so the position of the band remains unaf-
fected. On the other hand, electron-withdrawing groups such as CF3, NO, NO2, CN, COCH3’ COOCH3’ —CHO,
etc., operates through inductive mechanism and does not produce detectable splitting but shifts the first band of
benzene (i.e. 9.40 eV) to higher potentials, e.g. —CF3 (9.90 eV) —NO (9.97 eV), —NO2 (10.26 eV), —CN (10.02
eV), − CHO (9.80 eV), etc. When two substituents are introduced into the benzene ring at the p-position, the
experimental splittings and shifts are approximately additive. Thus, photoelectron spectrum of benzene derivatives
enables to distinguish some conjugative and inductive effects. Halogen substituents operate via both the effects and
cause splitting and shifting of both the bands, e.g. C6H5F (9.5 eV, 9.86 eV), C6H5C1 (9.31 eV, 9.72 eV), C6H5Br
(9.25 eV, 9.76 eV), C6H5I (8.78 eV, 9.75 eV). The splitting of the ionisation band follows the order: I >Br >Cl >F,
while the shifts (i.e. mean ionisation potential of the first two bands minus the first ionisation potential of benzene)
follows the reverse trend of splitting, i.e. F >Cl >Br >I. When all the hydrogens in benzene molecule are replaced by
fluorine, s electron ionisation bands of the unsubstituted molecules are shifted to much higher ionisation energy as
compared to those of p electron ionisation bands. A clear separation between p − and s − ionisation bands is seen in
the PES spectrum. Both aliphatics and aromatics exhibit such an effect called the perfluoro effect.
Problem 6.10: Calculate the lone-pair adiabatic ionisation energies of C2H5NO2, n-C3H7NO2 and iso-
C3H7NO2.
Solution Substituting the values of the needed parameters from tables 6.5 and 6.6 into Eq. (6.52), we get
IE (C2H5NO2) = 11.08 − 1 × 0.20 = 10.88 eV (10.88 eV)
IE (n-C3H7NO2) = 11.08 − 1 × 0.29 = 10.79 eV (10.81 eV)
IE (iso-C3H7NO2) = 11.08 − 1 × 0.36 = 10.72 eV (10.71 eV)

The quantities with in brackets are the literature values.
6.14.10 Nonbonded Interactions

Interactions between electrons of atoms or groups within molecules which are not represented by bonds are called
nonbonded interactions. Such interactions are generally ignored in simple molecular orbital models. But in some
cases they are strong enough to influence the electronic structure, particularly the nature of the outermost orbitals
in molecules, Examples are mutual interactions between the halogen lone pairs in the halides of carbon and sili-
con, interactions of lone pairs of heteroatoms in large organic molecules, hyper conjugation (one form of through
bond interaction), homo conjugation (through space interactions), e.g. nonboradiene, barrelene, cis, cis, cis-1,
4, 7-cyclo nonatriene—the interactions in these compounds are called isolated double bond interactions, the last
compound being the representative of homoaromatic compounds.
PES is most sensitive to nonbonded interactions as compared to other physical techniques and such interactions
manifest themselves in the form of splitting and shifting of ionisation bands under observations in the photoelectron
spectrum of the molecule. The transitions due to lone pair on heteroatoms in cyclic organic compounds have low
ionisation energies and relatively narrow bands as compared to that of bonded electrons. However, if the molecules
are conjugated, interactions between occupied p-orbitals also exhibit narrow bands. In such a situation, the band due
to lone-pair interaction can only be identified by molecular orbital calculations on model compounds and compari-
son with photoelectron spectrum. Thus, experimental studies together with empirical molecular orbital calculations
provide an aid to understand the nature of molecular electronic structure. Note that PES deals with ions but not with
the neutral molecules; conjugation being much stronger in ionised form compared with the neutral molecule.
The most interesting deductions regarding the nature of nonbonded interactions in cyclic compounds contain-
ing oxygen, nitrogen and sulphur as heteroatoms based on molecular orbital calculation on model compounds and
comparison with the photoelectron spectrum and vice versa are the following: The interactions between lone pair
of electrons on heteroatoms in a series of organic compounds may occur through space or bonds. The magnitude
of both these effects in cyclic compounds containing heteroatoms such as O or N or S depend markedly on molec-
ular geometry and on different accessible conformations in open-chain molecules. The different interactions in
different conformations result in overlapping of bands in the lone-pair region of the photoelectron spectrum. Due
to this very reason, it is preferred to study lone-pair interactions in the relatively rigid-ring compounds.
When the interaction is through space, the symmetrical combination of atomic orbitals, i.e. positive overlap
integral will have higher ionisation potential (lowerr energy) but the unsymmetrical combination of orbitals will
have lower ionisation energy if the overlap integral is negative. The reverse is true for the through-bond interac-
tion. The through-bond interaction makes a new orbital containing mainly a symmetrical combination of lone
pairs energetically outermost, giving the opposite ordering to that expected on a through-space mechanism and
it also shifts the centre of gravity of the bands. The effects of the through-space and through-bond interactions on
the lone-pair orbital energies are shown in Fig. 6.33.
The extent of dominance of either of these interactions, i.e. through space and bonds depends upon not only on the
number of carbon centres separating the heteroatoms but also on the nature of the heteroatom. The above mechanism
also holds good for through-space interactions, e.g. in case of 2 (1,4-diazabicyclo [2,2,2] octane) (DABCO), the
through-bond interaction is dominant. The centre of gravity of the doublet (7.52 eV, 9.65 eV) at 8.6 eV undergoes a
shift of + 0.58 eV relative to the lone-pair band of quinuclidine at 8.02 eV. The energy difference between two lone-
pair ionisations is 2.13 eV. The first-order splitting, i.e. through space interaction is expected to be small due to the
large separation between the two nitrogen centres of about 3 Å. The number of bands in the photoelectron spectrum
may or may not correspond to the number

σ*(a)
of equivalent lone pairs even if the heteroa-
toms are separated by a number of bonds, N (a) N (s)
e.g. only a single doublet at 8.6, 9.0 eV
and at 8.8, 9.4 eV has been observed in 1,
4-dithiane and in 1, 3, 5-trithiane respec- na and nb
Binding Energy
tively. The splitting of the lone-pair band
in a series of compounds passes through
a minimum and rises again. The alteration
in the magnitude of splitting is the char-
acteristic of the transition from through- N (s) N (a)
space to through-bond interaction and the
Through Space Through Bond σ(s)
negative sign indicates the reversal in the
order of the symmetrical and antisym-
metrical orbitals combinations in pass-
ing from through-space to through-bond Fig. 6.33 Schematic molecular-orbital-energy diagram illustrating the effects of
through-space and through-bond interaction on the ionisation potential of lone-pair
interactions, e.g. the splittings in case of orbitals: s-symmetrical, a-asymmetrical, N-nitrogen.
o-, m- and p-diazines are 2.0 eV, 1.5 eV
and −1.7 eV respectively.
The through-space interaction is dominant when the two heteroatoms are neighbours and it remains dominant
even if they are separated by one carbon centre, e.g. the magnitude of splitting in a doublet (8.6 eV, 9.0 eV) of
1,4-dithiane is 0.4 eV whereas in a doublet (8.8 eV, 9.4 eV) of 1,3, 5-trithiane, it is 0.6 eV.
The spectral data of some model cyclic compounds having N, O and S as heteroatoms has been summarised
in Table 6.15.
Table 6.15 Characteristic spectral parameters of heteroatom lone pairs of some cyclic systems.
Molecular Heteroatom lone pair Splitting Centre of grav- Shift of centre of Dominant interaction
system ionisation potential (eV) ity of doublet gravity relative to
band (eV) (eV) original band
S
8.5 ⎯ ⎯ ⎯ ⎯
S 8.6, 9.0
0.4 8.8 0.3 Bond
S (Doublet)
S 8.8, 9.4
0.6 9.1 0.6 Space
S S (Doublet)
N
N
⎯ 2.0 ⎯ ⎯ Space
N
⎯ 1.50 ⎯ ⎯ Space
N
N
Transition from space
⎯ −1.70 ⎯ ⎯
to bond
N
⎯ ⎯ ⎯ ⎯ ⎯
8.02 ⎯ ⎯ ⎯ ⎯
N
7.52, 9.65 2.13 8.6 0.58 Bond

N
N (Doublet) [1.6] [0.37]
The quantities within square brackets are from the molecular orbital model.
The splittings between lone-pair ionisation potentials of heteroatoms in some typical molecules have been
given here.
O
CH3 CH3
3.6 eV 0.55 eV t-Bu . S—S . t—Bu 0.64 eV
N N
O
CH3 O H
N N 3.3 eV C C −1.6 eV S S −0.45 eV
CH3
H O
O S
NH. CH2. CH2. NH2 0.5 eV O 0.25 eV S 0.41 eV
N N −0.3 eV
N N −2.1 eV O O −1.22 eV
Lone-pair interactions occur in halogen compounds also but are complex relative to lone-pair interactions in cyclic
compounds containing heteroatoms such as N, O and S. A single halogen atom has two lone pairs and will give
two lone-pair bands in the photoelectron spectrum either because of spin-orbit interaction or because one halogen
p-orbital interacts more strongly than the other with the remainder of the molecule, e.g. in case of methyliodide (CH3I)
four distinct bands in the form of two doublets at 9.5 eV, 9.6 eV (doublet with peak height ratio 15:1) and at 10.2 eV,
10.3 eV (doublet with peak: height ratio 13:1) are observed in the spectrum. The strong peaks of equal intensities at
9.5 and 10.2 eV are due to spin-orbit coupling, i.e. first-order effect, the spin-orbit splitting being ≈0.7 eV whereas
the two doublets are because of Jahn–Teller effect, i.e. second-order effect; the Jahn–Teller splitting being ≈0.1 eV.
CH3Br also a shows a similar spectrum but the intensities of the strong bands at 10.6 eV and 10.8 eV are quite differ-
ent, the intensity ratio of the two peaks is ≈1.5. This is attributed to the removal of degeneracy of orbitals not only by
spin-orbit coupling but also by Jahn–Teller effeet, as casual, the latter being still smaller than the former. Further, in a
molecule with two halogen atoms each having two lone pairs, four lone-pair ionising bands are expected and the same
has been observed in the experimental photoelectron spectrum of dichloroethylene. The photoelectron spectrum of
diiodomethane also exhibits four distinct bands at 9.4, 9.8, 10.3 and 10.7 eV while that of dichloromemane (CH2Cl2)
shows only two bands at 11.4 and 12.3 eV. The separation of two bands in the spectrum of CH2C12 into four in the
spectrum of CH2I2 may be due to spin-orbit splitting which is partially true since spin-orbit splitting can arise from
accidental degeneracy as well as symmetry-based degeneracy, i.e. when two orbitals are nearly degenerate, the second
order effects become as strong as the first-order effects. The degeneracy is removed by distortions brought about by
excitation of one or more degenerate vibrational modes of the molecule, which are called the Jahn–Teller vibrations.
Let us now consider the nonbonded interactions between isolated double bonds, i.e. p-orbitals instead of lone-pair
electrons on heteroatoms in cyclic compounds. Both types of nonbonded interactions, i.e. through space as well as
through bond are also operative between isolated p-orbitals. The through-space interaction between isolated p-orbitals
is called homo-conjugative interaction while it is called hyper-conjugative interaction when it operates through bond.
Just like nonbonded interactions between lone pairs on heteroatoms, the isolated double-bond interactions are
also accomplished on the basis of molecular orbital calculation data in conjunction with the observed photoelectron
spectral data on some model compounds. The model candidates for through-space interaction are norbornadiene (I)
barrelene (II); cis, cis, cis-1,4,7-cyclononatriene (III), etc., and for through-bond interaction are l, 4-cyclohexadiene
(IV); 1,4,5,8-tetrahydronaphthalene (V), etc.; the typical overlap integrals for the former and similar molecules in an
Huckel molecular treatment is about 1 eV as compared to 2–3 eV for normal double bonds and the unsymmetrical
p-orbitals combination, i.e. negative overlap integrals has a lower ionisation potential than the symmetrical combi-
nation as required by the through-space interaction while the reverse is true for through-bond interaction, i.e. sym-
metrical combination of p-orbitals has lower ionisational potential compared with unsymmetrical combination.
Let us now examine the photoelectron spectrum of the candidates for through-space interaction. Both the p-orbitals
of norbornadiene (I) are singly degenerate but its PES exhibits two bands at 8.69 and 9.55 eV. Thus the splitting of
p-levels which does occur in norbornadiene and its 7-isopropylidene derivative (VI) is the result of p-bond interaction.
This p-bond interaction is through-space since through-space splitting is expected to be small (0.86 eV) due to the large
p-bond separation of about 2.5 Å compared with that in butadiene (3.15 eV) with a p-bond separation of 1.483 Å.
(I) (II)
(III)
(V) (IV)
(VI)
The through-space interaction between p-orbitals in norbomadiene (I) and such molecules can also be
accomplished by the general method proposed by E Heilbronner. In this method, the PES of the system
under observation, i.e. the parent molecule, is compared with the PES of a similar system having the parent
molecule in its structure also. Here, the spectrum of 7-isopropylidene derivative of norbornadiene (VI) is
compared with that of norbornadiene (I), the parent molecule. Isopropylidenenorbornadiene (VI) has three
p-bonds out of which the two p-bonds in the parent structure are equivalent and the dihedral angle between
them is almost the same as in the parent molecule itself. The new p-bond because of its symmetry interacts
with the unsymmetrical combination of the two equivalent p-orbitals giving rise to three ionisation bands
at 9.54, 9.25 and 7.97 eV in the photoelectron spectrum of the compound (VI). Thus, the higher ionisation
energy band of the parent compound at 9.55 eV remains unperturbed while the lower energy band at 8.69
eV splits into two separate bands at 7.97 and 9.25 eV in its 7-isopropylidene PES spectrum. This effect
proves that the outermost p-orbitals in norbornadiene has an unsymmetrical combination as demanded by the
through space interactions.
6.14.11 Study of Biological Systems

XPS proved to be a convenient tool for studying the presence of metals in metalloproteins. This technique is espe-
cially suitable for those metals which cannot be studied by means of Mössbauer or ESR spectroscopy. In some
cases, even the site of the metal in metalloproteins can be located by XPS. This is demonstrated here by compar-
ing x-ray photoelectron spectral data of Cu 2P3/2 and Zn 2P3/2 in erythrocuprein with the corresponding values of
different Cu/Zn amino acid chelates.
The intensity ratio of the signals in Cu/Zn amino acid chelates is approximately the same (Zn:Cu). However,
in case of erythrocuprein, a marked change has been noticed. The intensity ratio of the signals has been found
to be 1:2.5 (Zn: Cu). Since two gram atoms of each metal are present in erythrocuprein, it is inferred that Cu is
located in the outer sphere of the protein portion while Zn must be less accessible to the Mg Ka radiation. The Zn
2
P3/2 signals are observed at 1021.5 eV in the amino acid chelates and the aquo complex, and an identical value of
the 1020.5 eV is measured in erythrocuprein. On the other hand, the Cu 2P3/2 signal at 934.2 eV in the amino acid
chelates is observed at 931.9 eV in erythrocuprein. But in case of Cu [Cu (Asp)2], two signals have been observed
at 932 and 934.2 eV. The former signal at (932 eV) is attributed to extraneous Cu2+ while the latter (at 934.2 eV)
is attributed to the binding energy of inner Cu 2P3/2. It is interesting to note that the copper in Cu [Cu (Asp)2] is
identical to that of copper in erythrocuprein.
System BE(eV) Intensity ratio of Zn:Cu in Intensity ratio of Zn:Cu in

Cu/Zn amino acid complexes erythrocuprein
Cu[Cu (Asp)2] 934.2 Same 1:2.5

(Cu 2P3/2) 932
Bovine erythrocuprein 931.9
(Cu 2P3/2)
Zn[Zn (Asp)2] 1021.5
(Zn 2P3/2)
Bovine 1020.5
erythrocuprein
(Zn 2P3/2)
PROBLEMS
1. (a) Draw the 3p photoelectron spectra of Kr and Xe atoms.
(b) The 3p PES of Kr and Xe atoms using He (I) 21.22 eV V″ IP(eV)
radiation source show that the photoelectron kinetic ener- 0 20.29
gies for the two different states of Kr are 7.23 and 6.57 eV
and that of Xe are 9.093 and 7.793 eV. Calculate 1 20.42
(i) the BEs for the different states of Kr and Xe, and 2 20.55
(ii) the weighted mean of the ionisation energies of Kr
3 20.67
and Xe atoms as well.
4 20.78
[Ans:(b);(i )Kr. 13.99 eV, 14.65 eV;Xe:12.127 eV, 13.427
eV(ii) Kr 14.210 eV, Xe: 12.560 V] 5 20.88
6 20.96
2. The figure given below shows an energy-level diagram for
sodium with approximate BEs for the core levels as deter- (b) Compare the force constant of O2+ ion with that of
mined from XPS spectrum of NaCl obtained by using the oxygen molecule if the vibratlonal frequency of
MgKa (hv = 1253.6 eV) radiation. At what kinetic energy will the latter is 1580 cm–1.
the Na, Is, 2s and 2p photoelectron peaks be observed? +
8. The vibrational frequency of O2 4pu ion as determined
Vacuum level from the vibrational progression in the PES of O2 molecule
(using He (I) resonance line hv = 21.22 eV) is 1129 cm–1.
VB
Calculate the change in bond length on ionisation of O2
2p (BE = 31 eV) molecule if the difference between the vertical and adia-
2s (BE = 64 eV) batic ionisation potentials is 0.67 eV. Also, determine the
+
1s (BE = 1072 eV)
bond length of O2 ion if the bond length of O2 molecule
is 1.21 Å and its vibrational frequency is 1580 cm–1.
[Ans: 2p(1222.6 eV, 2s (1189.6 eV); and 1s (181.6 eV)] [Ans: 0.19 Å; 1.40 Å]
3. Calculate the probability of escape of a photoelectron 9. The argon 2P3/2 peak is often used as a standard test
through a solid without undergoing scattering for λ = 20.0 Å resolution. Calculate the observable widths at 300 K and
10 K when ionisation of Ar is carried out by He (I):21.22
d eV and He (II) 40.81 eV radiation lines respectively. The
and d= 100 Å. At what value of d or will the probability
λ d natural line width of PES signal is 6.5 meV and atomic
of escape be maximum? [Ans: 67 10–4, d or ] mass of Ar is 39.95 amu.
λ
4. Calculate the probability P(d) of escape of an electron [Ans: At 300 K;9.5meV; 13.1 meV, At 10K;
through a solid without undergoing scattering when d = 1.7 meV; 2.4 meV]
d +
1λ, 2λ, 3λ …8λ. Draw a graph between P(d) and and 10. The width of ionised band of H2, i.e. H 2 , when
λ recorded using He (I); 21.22 eV radiation line is 22
then from the graph find out at what value of d will the meV and is 46 meV when He (II) 40.81 eV radiation
probability of escape of the electron tend to be zero? line is used. Calculate the theoretical line widths of H2
[Ans: d =5λ] ionised bands obtained by He (I) and He (II) radiation
5. Calculate the escape depth of an electron in a solid if its lines at 298 K if the natural PES line width is 6.5 meV.
kinetic energy is 1000 eV. [Ans: He (I), 42.17 meV; He (II), 58.48 meV.)
[Ans: 31⋅ 6 Å] 11. (a) Calculate the thickness of the covering layer of
6. Using the data in Problem 6.6, draw a graph of the vibra- some material on the surface of a catalyst if the pho-
+
2
+ toelectron signal intensity of the catalyst surface in the
tional interval of H2 ion (X Σ g ) against the upper-state
absence of the covering layer is reduced to 1 th of its
vibrational quantum numbers, i.e. V≤. From the graph, 4
find the maximal value of vibrational quantum number at value in the presence of the covering layer, and (b) find
which the vibrational interval is zero and using this value at what value of the thickness of the covering layer on
of V≤, determine the dissociation energy of hydrogen the surface of the catalyst is I = I0? λ= 20 Å.
molecule ion and compare it with the one computed by [Ans: (a)1.38 nm, and (b)r = 0 Å]
nongraphical method in Problem.
12. The BEs of valence electrons of Pd metal with respect
[Ans: V≤max = 16 De = 24051 cm-1] to the vacuum level lie in the energy range 0–8 eV.What
will be the BE range with respect to the Fermi level if the
+ −
7. The fourth ionisation band of O2 (B g ) in the energy gap between the two scales is 4 eV?
photoelectron spectrum of oxygen (O2) exhibits a [Ans: 4 –12 eV]
vibrational progression. The ionisation potentials in the
13. What will be the ratio of He (I) excitation radiation line
vibrational progression on different vibrational quantum
width to that of He (II) excitation radiation line width?
numbers are given here:
[Ans: ≈1:7]
(a) Calculate the dissociation energy of the O2+ ion 14. By making use of the data in tables 6.13 and 6.14,
by the graphical and nongraphical methods and generate the lone-pair ionisation energies of C2H5NH2’
compare the values so obtained with that of the n-C3H7NH2, iso-C3H7NH2, n-C4H9NH2! iso-C4H9NH2 and
oxygen molecule listed in the appendix of the book. tert-C4H9NH2.
[Ans: 8.84 (8.86); 8.78 (8.78); 8.73(8.72) 8.74 (8.71); 16. Calculate the Fermi energy of a free electron in the
8.72 (8.70) and 8.63 (8.64) eV. The quantities within highest energy state in (a) Cu and (b) Ag, and (c) Al.
parenthesis are the liteature values.] Density (Cu) = 8.96 g/Ml, Density (Ag) = 10.5 g/Ml; Den-
sity (Al) = 2700 kg/m3; Molecular mass (Cu) = 63.54 g/
15. The partial ESCA spectra of a mixture of carbon diox-
mole; Molecular mass (Ag) = 107.870 g/mole; Molecular
ide, carbon monoxide and methane is presented in the
mass (Al) = 27g/mole.
figure 6.34. From the spectra, determine the percent-
age composition of the gas mixture. [Ans: (a) 7.0 eV, (6) 5.5 eV]
[Ans: CO2;28.4 percent CO;36.3 percent CH435.3
percent]
O 1s C 1s
CO
CO CH4
CO2 CO2
Intensity (Counts/sec)
300
200
100
545 540 295 290

Binding Energy (eV)
Fig. 6.34 O1s and C1s ESCA spectra of a mintane of CO2, CO and CH4 .
CHAPTER 7 NUCLEAR MAGNETIC
RESONANCE SPECTROSCOPY
Discoveries and inventions are not terminals. They are fresh starting points from which we can climb to new
knowledge.
—Whitney
It is the intuition of unity among diversity which impels the mind to form science.
—Hoffman
7.1 INTRODUCTION
Nuclear Magnetic Resonance (NMR) spectroscopy is based on the measurement of absorption of electromagnetic
radiation in the radio-frequency region of roughly 4–900 MHz by the analyte placed in an external magnetic field.
The purpose of the external field is to cause the nuclei to develop the energy states required for the absorption of
radiation.
The theoretical basis for NMR spectroscopy was proposed by W Pauli in 1924. He suggested that splitting of
energy levels of atomic nuclei will take place under the influence of an external magnetic field provided these
nuclei have properties of spin and magnetic moment.
The first nuclear magnetic resonance experiment in the strict sense was performed on hydrogen in 1939 by
II Rabi, S Millman and P Kusch. However, the most successful experiments in the field of NMR were conducted
independently in 1946 by F Bloch, W W Hansen and M E Packard, and by E M Purcell, H C Torrey and R V
Pound. The impact of this phenomenon was so great that within five years after the discovery, when it was found
that NMR frequencies depend upon chemical environment of the nuclei, commercial instruments capable of
resolving NMR lines separated by 0.1 ppm became available. The day of the chemists in NMR dawned when WG
Proctor and FC Yu in 1951 discovered two NMR lines in NH4NO3 corresponding to nitrogen in NH +4 and NO3−
, ions. This was the experimental evidence to the fact that frequency depends on the electronic surroundings in
which the nuclei are placed. In 1952, the independent discoveries of Bloch at Stanford and Purcell at Harvard
were recognised in the form of the Nobel Prize in physics. The technique, though discovered by physicists, proved
to be a boon for chemists as an analytical tool as well as for structural elucidation of molecules. The second Nobel
Prize in NMR was awarded in 1992 to Richard R Ernst of Switzerland for his contribution to Fourier transform
NMR spectroscopy. Currently, two broad classes of NMR are (i) high-resolution continuous wave (CW) NMR,
i.e. CWNMR; and (ii) pulsed or Fourier transform NMR, i.e. FTNMR. In CWNMR, the behaviour of nuclear
spins is monitored by frequency-sweep methods. The excitation is moved across the whole frequency range of
interest and the magnetic field is kept constant. But in FTNMR, the behaviour of nuclei is manipulated with the
aid of sequences of pulses of varying duration, phase, frequency and amplitude and the external magnetic field is
also maintained constant. The capabilities of NMR have been extended to two and three dimensions, hence trans-
forming NMR into a molecular structure determination tool competing with x-ray crystallography.
7.2 CHARACTERISTICS OF MAGNETIC NUCLEI

The nucleus of an atom is characterised by its mass, charge, spin quantum number or intrinsic angular momentum.
The angular momentum of the nucleus is due to its spin or rotation about an axis. The circulatory charge of the
spinning nucleus generates a magnetic field and hence such nuclei have magnetic moments. The nuclear magnetic
dipole moment (mI) is 1000 times smaller than that of an electron and is given by the relation
⎛ e ⎞
μI = g N ⎜ I (I + 1) (7.1)
⎝ 2M N ⎟⎠
where gN is the nuclear g factor, MN is the mass of the nucleus, e is the electronic charge, I (I + 1) . is the magni-
tude of the spin angular momentum, and I is the spin quantum number. Magnetic-dipole moment and spin angular
momentum act in the same direction in case of nuclei, whereas in opposite directions in the case of an electron.
The laws governing the vector addition of nuclear spins are not yet known, so the spin of a particular nucleus
cannot be predicted in general. However, on the basis of observed spins, certain empirical rules have been framed.
The rules are the following:
(i) For nuclei, I = 0, eQ = 0 where, Q is a measure of deviation of charge distribution from the spherical symmetry,
i.e. there is no net circulation of nuclear charge in these types of nuclei and hence their magnetic moment is
zero, e.g. 12C, 32S, 16O, 28Si, etc. Thus, all nuclei with even mass number (A) and even charge number (Z) do not
have magnetic moments and are NMR-inactive. The spherical nonspinning nucleus is shown in Fig. 7.1(a).
(ii) When I = 1/2, eQ = 0, these types of nuclei have spherical charge
distribution and hence have magnetic moments and are NMR-active, +
e.g. 13C, 1H, 3H, 19F, 31P, 15N, etc. Thus, all nuclei with odd mass num-
bers and even or odd charge numbers possess magnetic moments. − − −
+ +
The spherical spinning nucleus is shown in Fig. 7.1(b). + −
(iii) For I >1, eQ > 0 or eQ < 0. In such types of nuclei, there is also a net cir-
culation of either positive or negative charge and hence they have mag- (a) (b) (c) (d)
netic moments and are also NMR-active. The charge distribution in these Fig. 7.1 Various representations of nuclei;
(a) I = 0, eQ = 0; (b) I = 1/2, eQ = 0; (c) I ≥ 1,
types of nuclei is nonspherical and, therefore, besides nuclear magnetic eQ > 0; and (d) I ≥ 1, eQ < 0. A positive value
moment they also possess electric quadrupole moment and hence are of eQ indicates that the protonic charge is
NQR (Nuclear Quadrupole Resonance)-active also, e.g. 2H, 14N, 10B, 11B, oriented along the direction of the applied
35
Cl, 37Cl, etc. Thus, all nuclei with odd charge numbers and even or odd field while negative values for eQ indicate
mass numbers possess magnetic dipole as well as quadrupole moments. that the charge is oriented perpendicular to
the direction of the applied field.
The nonspherical spinning nucleus is shown in Figs 7.1(c) and (d).
7.3 CONCEPT OF MAGNETIC ENERGY STATES

When a tiny bar magnet is suspended, it will align itself parallel to the direction of the earth’s magnetic field (geo-
graphical). The magnet in this situation is said to be in low-energy state. In order that it should flip in a direction
anti-parallel to the direction of the earth’s magnetic field, some energy has to be supplied from an external source, and
this state of the magnet is termed as the excited state. This behaviour of a bar magnet in the earth’s magnetic field is
shown in Fig. 7.2 and it forms the basis of nuclear magnetic
N S Energy Supplied N S
resonance. Nuclei with nonzero spins also behave as tiny bar
ΔE
magnets under the influence of an external magnetic field. N N S S N S N S
The energy to flip the nuclei from a direction parallel to the N S N S
external magnetic field to that anti-parallel to the field is sup- Low-Energy High-Energy
plied in the form of radiation in the radio frequency region. State State
In the text, the low-energy state, i.e. parallel, is denoted by a Fig. 7.2 Behaviour of a bar magnet in the earth’s magnetic
field.
or +1/2 while the high-energy state, by b or −1/2.
7.4 LARMOR THEOREM

Applied field
According to this classical theorem, a spinning bar magnet in its H
N-S direction, free from all frictional forces, under the influence of
Precessional
a steady magnetic field, precesses about the direction of the applied orbit
magnetic field as the spinning mechanical top or gyroscope precesses Component of μI μH Magnetic dipole
μI (Magnetic moment)
in the direction of the gravitational field, shown in Fig. 7.3. in direction of H
θ Rotational axis
It is to be noted that because of gyroscopic effect, the force
applied by the applied magnetic field to the axis of rotation of the Nucleus spinning
spinning magnet does not move the magnet in the plane of the θ
force but perpendicular to this plane. Consequently, the rotational
axis of the spinning magnet moves in a circular path, i.e. pre-
cesses around the axis of the applied magnetic field. The angular
velocity of this rotation, w, in radians per second is given by
Fig. 7.3 Precession of a spinning top of spinning nucleus
w = gH (7.2) in a magnetic field.
Nuclear Magnetic Resonance Spectroscopy 509
where w = 2pv, n is the frequency of precession, also called Larmor frequency, g is the gyromagnetic constant
also known as gyromagnetic ratio, i.e. the ratio of magnetic dipole moment and spin angular momentum. Thus,
Eq. (7.2) can also be expressed as
γH
ν= (7.3)
2π
The classical Larmor frequency is identical to the frequency of absorbed radiation derived quantum mechani-
cally. This theorem is helpful in understanding the adsorption of radio frequency radiations by a magnetic nucleus
under the influence of external magnetic field.
7.5 ENERGY OF THE MAGNETIC STATES

We know that all nuclei with nonzero spin behave as tiny bar magnets. Thus when placed in a uniform magnetic
field of strength H, the nuclear magnetic dipole will precess about the axis of the applied filed. The frequency of
precession of a spinning magnetic moment is given by
γH
ν=
2π
μI 1
where, γ = and is different for different nuclei.
I (I + 1)
The potential energy E of the precessing nuclei which is due to the interaction between the applied magnetic
field and magnetic moment of the nucleus is given by
E = –mIH cos q (7.4)
where q is the angle between the direction of the H H
magnetic dipole and the direction of the applied field H.
On the basis of classical mechanics, q can take on any
value so that energy varies continuously. But according Ih = h θ I(I + 1)h = 3 h
2 4
to quantum mechanics, q can take on only certain per-
missible values since angular momentum is quantised in
space. Thus, when radio frequency energy is absorbed Ih = −1 h I(I + 1)·h = 3 h
2 θ 4
by the nucleus, its angle of precession changes. It is to
be noted that since magnetic moment of a circularly
polarised radiation of a suitable frequency has the char-
acteristics of a precessing nucleus, it is absorbed by the Fig. 7.4 Nuclear spin angular momentum and its components along
nucleus for its flipping. The relation between angular the applied magnetic field.
momentum for I = 1/2 and its component along the axis
of external magnetic field, i.e. its direction of quantisation, is shown in Fig. 7.4.
For I = 1/2, q can take on only two values, i.e. either 350 15' or 1440 45' as shown here. The projection of the
spin angular momentum on the axis of applied field
–
= I MI (7.5)
2
The spin magnetic quantum number MI can have (2I + I) values, i.e. MI = I, I – 1, I – 2,…, –I. Thus, for I = 1/2,
the two values for MI are +1/2 and –1/2. Therefore, q in this case is either 350 15' or 1440 45' since
3 1
sin θ sin 35 15′
2 2 3
The same is true for magnetic moment.

Let mH be the component of mI in the direction of the field; then
μH
cosq = (7.6)
μI
Form Eqs (7.4) and (7.6), we obtain
E = − μH H (7.7)
where, mH = gN bN MI (7.8)
Here, gN is the nuclear g-factor (for proton gH = 5.5837) and bN is the nuclear magneton.
eh
βN = = 5.046 × 10 −27 JT −1 (7.9)
4π m p c
where c is the velocity of light, c = 2.997925 × 1010 cms−1, mp is the mass of the proton = 1.67252 × 10−24 g, e, the
electron charge = 4.803 × 10−10 esu and the Planck’s constant h = 6.623 × 10−27 erg s = 6.623 × 10−34 J s.
E = −gN bN H MI (7.10)
Thus, the total number of the energy levels in the magnetic system is governed by MI for a given value of I, and
each energy level is characterised by its magnetic quantum number.
For I = 1/2, MI can take on two values, i.e. MI = ±1/2
Further, for MI = +1/2, Eq. (7.8) becomes
−1
E g N βN H (7.11)
+
2 2
This corresponds to low-energy orientation in which the magnetic field of the nucleus is aligned with the exter-
nal field.
Similarly, for MI = −1/2,
1
E g N βN H (7.12)
−
2 2
This corresponds to the high-energy orientation in which the magnetic field of the nucleus is anti-parallel to
that of the external field.
In the absence of an external magnetic field, the two orientations, i.e. substates, are said to be degenerate. This
degeneracy is removed with the aid of an external applied field (Zeeman effect) as shown in Fig. 7.5(a). Similarly, for
I = 3/2, MI can have four values, i.e. 3/2, 1/2, −1/2, −3/2. The Zeeman splitting in this case is shown in Fig. 7.5(b).
E = −gNbNH MI (a)
MI = −1/2
Higher Energy
State
I = 1/2
ΔE = h v
No External
Magnetic Field
MI = +1/2
Zeeman Splitting Lower Energy
State
(b)
E = −gNbNHMI MI = −3/2
I = 3/2 MI = −1/2
No External MI = +1/2
Magnetic Field
MI = +3/2
Zeeman Splitting
H
Fig. 7.5 The splitting of magnetic energy level under the influence of an
external magnetic field (a) I = 1/2, and (b) I = 3/2.
7.6 GENERAL SELECTION RULES FOR THE TRANSITION BETWEEN

MAGNETIC STATES
The transition probability is proportional to the integral
∫ (Ψ I Ψ dτ )
2
n m
The selection rule evolved from the solution of the integral is ΔMI = ±l. A nonresonant nucleus does not change
its quantum stage, hence absorption (ΔMI = +l) rather than emission (ΔMI = − l) is considered which of course is
not so in vibrational and rotational spectra. The general selection rule is supplemented by the condition that the
resultant MI of one and only one basic group, i.e. ΣMI, must change by unity in every observable transition, e.g.
ΣMI for CH3 group = 3/2. In other words, we can say that chemically equivalent nuclei are magnetically equivalent
also and hence they are characterised by only one quantum number which is the sum total of quantum numbers
of individual nuclei, e.g. for CH2,
1 1
M I = + = 1.
2 2
7.7 SPACING BETWEEN TWO CONSECUTIVE MAGNETIC LEVELS

The energy of any magnetic state is given by E = −gN bN H MI (Eq. 7.10)
For MI = I, I −1, I − 2,..., −I
EI = − gN bN HI (7.13)
EI−1 = −gN bN H (I −1) (7.14)

Subtracting Eq. (7.13) from Eq. (7.14),
ΔEd = EI−1 − EI = gN bN H (7.15)
Thus, the space ΔEd between two consecutive states is independent of magnetic quantum number but does
depend upon the strength of the external field and g-factor of the nuclei. For a specific nucleus, ΔEd depends only
on the strength of the external field.
Units of field strength = 1 tesla T = 104 Gauss = 1 kg s−2 A−1
Problem 7.1: Determine the distance between any two successive states of the 1H nuclei at a field strength of
13,500 gauss.

ΔEd = gN bN H
For hydrogen nuclei, gN = gH = 5.5854, H = 13500 gauss = 1.35 T, bN = 5.0504 × 10−27 J T−1.
Thus, ΔEd = 5.5854 × 5.0504 × 10−27 (JT−1) × 1.35 (T)
= 38.08 × 10−27 J = 19.17 × 10−4 cm−1
7.8 TRANSITION BETWEEN MAGNETIC LEVELS

The transition from low energy to higher energy state can occur only if the nuclei under the influence of external
magnetic field absorb a suitable quantum of radio frequency energy ΔE = hv. The nuclei in the lower energy state
absorb energy so as to flip in a direction anti-parallel to that of the applied field as already shown in Fig. 7.2.
The energy difference between two energy levels as determined from Eqs (7.11) and (7.12) and selection rule
ΔMI = ±1 comes out to be
ΔE = gN bN H (7.16)
which is equal to the gap between any two successive magnetic states given by Eq. (7.15).
Thus the energy difference for a specific nucleus depends upon the strength of the applied field. The system of
nuclei in a magnetic field is capable of absorbing small quanta if the condition hv = bN gN H is satisfied. The fre-
quency of radiation at which absorption occurs is called resonance frequency, and the phenomenon itself is known
as nuclear magnetic resonance. When the frequency exceeds the resonance frequency, the quantum condition is
not qualified and the absorption stops. Consequently, the NMR signal appears in the form of a narrow peak. NMR
spectrum is represented by plotting the absorption intensity on the vertical axis versus the strength of the external
magnetic field on the horizontal axis. The resonance frequency of different nuclei, say IH, 19F, 31P, 13C, etc., sharply
differ from one another because of different gN factors and hence their NMR spectra never overlap.
The characteristic parameters of some common nuclei are listed in Table 7.1. When the nucleus under consid-
eration is hydrogen, the nuclear magnetic resonance is called proton magnetic resonance and is abbreviated as
PMR or simply IH-NMR. Similarly, 13C-NMR, 19F-NMR, 31P-NMR, etc., stand for 13C, 19F- and 31P, etc., nuclear
magnetic resonance respectively.
Problem 7.2: (a) Calculate the value of gN for 13C if its resonance frequency in MHz at a field strength of
10 kilogauss is 10.705.
(b) Determine the resonance frequency of 19F in MHz at a field strength of 1 T if its gN value is 5.2532.
Solution (a) bN = 5.05 × 10−27 J T−1,

H = 10 kilogauss = 1 T
h = 6.623 × 10−34 J s
v = 10.705 MHz = 10.705 × 106 cps
Substituting these values in the Eq. (7.16), we get
6.623 × 10 −34 (J s ) × 10.705 × 106 (cps )
g ( C) =
13
0 −27
5.05 10 ( ) × 1( T )
= 1.4039
(b) Equation (7.16) may be expressed as
g β H
v=
h
Using the value of the respective parameters from part (a) and part (b) of the problem:
v=
5.2532 × 5.05 × 1 −27
( ) × 1( T )
6.623 × 10 −34
(J sec)
= 4.0055 × 107 cps
= 40.055 MHz
Problem 7.3: Show that as the field strength grows, the energy difference between magnetic levels also grows.
Solution This can be shown by determining the energy difference, ΔE, between the levels at different values
of H.
For H = 10,000 G
ΔE = gN bN H = (5.585) × (5.051 × 10−31 JG−1) × (10,000 G)
= 2.820 × 10−26 J
Since, ΔE = hv
2.820 × 10 −26 (J )
∴ v= = 42 Mcps
6.625 × 1034 (J s)
Table 7.1 Nuclear magnetic resonance data of some common nuclei.

Isotope Spin Relative Gyromagnetic Magnetic gN Relative NMR NMR Electric
Abundance Ratio g in Moment Sensitivity for Equal Resonance Quadrupole
in Nature units of 103 mI in Number of Nuclei, i.e. Frequency Moment Q
radian Multiple 1000 in MHz at a in Mul-
(Gauss)−1 of Nuclear Magnetic tiples of
(s)−1 Magneton At At Field of e × 10−24 cm2
(eh/4pmpc) Constant Constant 10 kG
Field Fre-
quency
H
I
1/2 99.9844 26.7519 2.79270 5.5837 1.0 1.0 42.5759
2
H/D 1 0.0156 4.106 0.85738 0.8571 9.64×10−3 0.409 6.53566 2.77×10−3
6
Li 1 7.46 − 0.82191 − 8.51×10 −3
0.392 6.265 4.6×10−4
7
Li 3/2 92.57 − 3.2560 − 0.294 1.94 16.547 −4.2×10−2
10
B 3 18.83 2.875 1.8006 0.6000 1.99×10−2 1.72 4.575 0.111
11
B 3/2 81.17 8.853 2.6880 1.7914 0.165 1.60 13.660 3.55×10−2
13
C 1/2 1.11 6.726 0.70216 1.40394 1.59×10−2 0.251 10.705
−3
14
N 1 99.64 1.933 0.40357 0.4034 1.01×10 0.193 3.076 2.0×10−2
15
N 1/2 0.365 −2.711* −0.28304 −0.5659 1.04×10−3 0.101 4.315
17
O 5/2 0.037 −3.627* −1.8930 −0.7504 2.91×10−2 1.58 5.772 −4.0×10−3
19
F 1/2 100.0 25.167 2.6273 5.2532 0.834 0.941 40.055
−2
23
Na 3/2 100.0 7.0789 2.2161 1.476 9.27×10 1.32 11.262 0.1
25
Mg 5/2 10.05 − −0.85471 − 2.68×10 −2
0.714 2.606
27
Al 5/2 100.0 6.971 3.6385 1.4549 0.207 3.04 11.094 0.149
29
Si 1/2 4.70 5.317 −0.55477 −1.1095 7.85×10 −2
0.199 8.460
−2
31
P 1/2 100.0 10.829 1.1305 2.2605 6.64×10 0.405 17.235
33
S 3/2 0.74 2.052 0.64274 0.4283 2.26×10 −3
0.384 3.266 −6.4×10−2
35
Cl 3/2 75.4 2.261 0.82089 0.5472 4.71×10−3 0.490 4.172 −7.97×10−2
37
Cl 3/2 24.6 2.182 0.68329 0.4553 2.72×10 −3
0.408 3.472 −6.21×10−2
39
K 3/2 93.08 − 0.39094 − 5.08×10−4 0.233 1.987
43
Ca 7/2 0.13 − −1.3153 − 6.39×10 −2
1.41 2.865
47
Ti 5/2 7.75 − −0.78712 − 2.10×10 −3
0.659 2.400
49
Ti 7/2 5.51 − −1.1023 − 3.76×10−3 1.19 2.401
50
V 6 0.24 − 3.3413 − 5.53×10 −2
5.58 4.245
51
V 7/2 ~100.0 − 5.1392 − 0.383 5.53 11.193 0.3
53
Cr 3/2 9.50 − −0.4735 − 1.0×10 −4
0.29 2.406
55
Mn 5/2 100.0 − 3.4610 − 0.178 2.89 10.553 0.5
57
Fe 2.245 − ≤0.05 − − − − −
59
Co 7/2 100.0 6.350 4.6388 1.3132 0.281 4.83 10.103 0.5
63
Cu 3/2 69.09 − 2.2206 − 9.38×10−2 1.33 11.285 −0.15
65
Cu 3/2 30.91 − 2.3790 − 0.116 1.42 12.090 −0.14
67
Zn 5/2 4.12 − 0.8735 − 2.86×10 −3
0.730 2.635
75
As 3/2 100.0 − 1.4349 − 2.51×10 −2
0.856 7.292 0.3
79
Br 3/2 50.54 6.704 2.0990 1.3989 7.86×10−2 1.26 10.667 0.33
81
Br 3/2 49.46 − 2.2626 1.5957 9.84×10 −2
1.35 11.498 0.28
95
Mo 5/2 15.78 − −0.9099 − 3.22×10 −3
0.761 2.774
97
Mo 5/2 9.60 − −0.9290 − 3.42×10 −3
0.776 2.833
107
Ag 1/2 51.35 − −0.1130 − 6.69×10−5 4.03×10−2 1.722
(Continued)
Table 7.1 Nuclear magnetic resonance data of some common nuclei—Cont'd.

109
Ag ½ 48.65 − −0.1299 − 1.01×10−4 4.66×10−2 1.981
111
Cd ½ 12.86 − −0.5922 − 9.54×10 −3
0.212 9.028
113
Cd ½ 12.34 − −0.6195 − 1.09×10−2 0.222 9.444
115
Sn ½ 0.35 − −0.9132 − 3.50×10 −2
0.327 13.22
117
Sn ½ 7.67 − −0.9949 − 4.53×10 −2
0.356 15.77
119
Sn ½ 8.68 −9.971* −1.0409 − 5.18×10 −2
0.373 15.87
121
Sb 5/2 57.25 − 3.3417 − 0.160 2.79 10.19 −0.8
123
Sb 7/2 42.75 − 2.5334 − 4.57×10 −2
2.72 5.518 −1.0
127
I 5/2 100.0 − 2.7939 1.1173 9.35×10 −2
2.33 8.519 −0.75
133
Cs 7/2 100.0 − 2.5642 − 4.74×10 −2
2.75 5.585 ≤ 0.3
199
Hg ½ 16.92 4.783 0.4993 0.9983 5.72×10−3 0.179 7.612
201
Hg 3/2 13.24 − −0.607 − 1.90×10−3 0.362 3.08 0.5
207
Pb ½ 21.11 − 0.5837 − 9.13×10 −3
0.209 8.899
Elec-
½ − 17.6×103 −1836 2.0023192# 2.85×108 658 27944
tron
* The angular momentum vector and the vector of the magnetic dipole are anti-parallel.
# Lande g factor for the electron. The positive value of gN indicates that the magnetic moment and angular momentum are parallel and
anti-parallel for negative value of gN.
For H = 14000 G, ΔE = 3.949 × 10−26 J, and v = 60 Mcps

For H = 24500 G, ΔE = 6.911 × 10−26 J, and v = 104 Mcps
Thus, as H increases, the energy difference between the magnetic levels also increases. Same is true for fre-
quency also or since gN and bN are constants of a nucleus, so ΔE/v a H.
7.9 NUCLEI POPULATION IN DIFFERENT STATES

The ratio of nuclei population in the two magnetic states is governed by Boltzmann’s law, i.e.,
N +1
2
= e ΔEE / T (since population density ∝ e−E/kT) (7.17)
N −1
2
where, ΔE = gN bN H.
The population difference, Δn is given
N ΔE Ng N βN H
Δn = N +1 − N −1 = − (since kT < 1) (7.18)
2 2 2 T 2κT
where, N −1 N 1 is the total number of nuclei in both the populations.
+
2 2
N 0e − E1/ T
Hint: N1 =
e −EE1 T e E2 / T
N 0e − E2 / T
N2 = ; N 0 N1 + N 2
e −EE1 T e E2 / T
The lower magnetic level has more nuclei than the higher level. Thus, if there were no thermal motion, all
nuclei would be at the lower energy level. The energy of thermal motion tends to bring the populations of the
levels into accord with Boltzman law. The population difference Δn varies inversely to absolute temperature and
directly to field strength, H.
Problem 7.4 Show that the distribution of magnetic nuclei between two magnetic states can be expressed as
N −1
⎡ g β H ⎤ ⎡ hv ⎤
2
= ⎢1 − = 1−
N +1 ⎣ κT ⎥⎦ ⎢⎣ κT ⎥⎦
2

N −1
2
= e − ΔEE /κT e −gg β H /κ T
N +1
2
x2 x3
Applying the expansion, e x 1+ x + + + ... and considering only the first two terms, we obtain
2 3
N 1
− ⎡ g β H ⎤ ⎡ hv ⎤
2
= ⎢1 − = 1− (since gN bN H = hv)
N 1 ⎣ κT ⎥⎦ ⎢⎣ κT ⎥⎦
+
2
Problem 7.5: Determine the ratio of the proton nuclei population at 300 K when the field strength is 13500 G.
k = 1.3806 × 10−23 JK−1.
Solution The population ratio can be determined as follows.
ΔE = gN bN H = 5.585 × (5.051 × 10−31 JG−1) × (13500 G)

= 3.808 × 10−26 J
N +1
−26 ( J)) / (1.3806 × 10 −23 −
−11 ) × 300 K
2
= e 3.808 × 10 e 0.000009 = 1.000009
N −1
2
The lower energy level has nine proton nuclei more than the higher energy level if its population scale is kept
1,000,000.
Problem 7.6: Determine the population difference of proton nuclei between two magnetic states at 300 K if
the strength of the external field is 13500 G and the total population of the nuclei is 1,000,000.
Solution The population difference between two states can be calculated with the aid of Eq. (7.18).
Ng N βN H
Δn =
2κT
Using the value of gN bN H from Problem 7.5, we get
1, 000, 000 × 3.808 × 10 −26 J
Δn = = 4 59 = 5
2 × 300 K × 1.3806 × 10 −23 ( −11 )
Problem 7.7: Using the expression for the distribution of magnetic nuclei between two states from Problem 7.4
and data of Problem 7.5, calculate N−1/2/N+1/2.
Solution From Problem 7.4, we write

N −1
⎡ g β H⎤
2
= ⎢1 −
N +1 ⎣ κT ⎥⎦
2
Substituting the needed data from Problem 7.5:

N −1
⎡ 3.808 × 10 −26 (J ) ⎤
2
= ⎢1 − −23 −1 ⎥
N +1 ⎣ (1.3806 × 10 JK ) × 300 K ⎦
2
= 1 – 0.0000091 = 0.9999909
Since N–1/2/N+1/2 < 1 the upper state is thinly populated as compared to the lower state.
7.10 LINE SHAPE AND LINE WIDTH

There are two important types of line shape in the NMR absorption curves: (i) Lorentzian line (L) shape, and
(ii) Gaussian line (G) shape. The main difference between the two types of line shapes is that the outerwings of
the L-line are much longer and drop more slowly than those of the Gaussian lines [see Figs 1.5(a) and (b)]. The
shape of the absorption line in some cases is an admixture of these two types of lines. These two types of spectral
lines are characterised by the measurable experimental parameters recorded in Table 7.2.
Table 7.2 Properties of Lorentzian and Gaussian Lines.
Parameter L-shape G-shape
Half width at half height ΔH1/2 ΔH1/2
12
1 ⎛ ln 2 ⎞ 1
Peak amplitude I0 = I0 = ⎜
πΔ
ΔH
H 1/ 2 ⎝ π ⎟⎠ πΔ
ΔH
H1 2
⎡ ⎤
I 0 ( ΔH
H )2 ⎢ − ln 2 ( − r )2 ⎥
Equation for absorption line I= I = I O exp ⎢ ⎥
( ΔH
H ) (H − H r )
( )
2 2 2
⎢ ΔH 1 ⎥
⎣ 2 ⎦
Hr is the field resonance at the line centre, I is the intensity of the absorption line as a function of field H for L and
G-line shapes.
7.11 FACTORS INFLUENCING THE INTENSITY OF NMR LINES

The NMR emission or absorption can only be detected when there is a population difference between the two
spin states. The transition between two spin states E−1/2 and E+1/2, can occur either by emission or by absorption.
The rates of both processes are proportional to (i) the population of the state considered, (ii) the radiowave energy
density, and (iii) the square of the transition matrix element.
The radiowaves of the oscillating magnetic field (hv) induce two types of transition: (i) from upper state to the
lower state (the emission process), and (ii) from the lower state to the upper state (the absorption process). When
the nuclear-spin population is equally divided between the upper population and the lower population, the emis-
sion of radiowave quanta is higher than the absorption of these quanta, and no observable signal is obtained. When
the population of the nuclear spin is higher in the lower state than in the upper state, a radio frequency absorption
is observed. We know that the ratio of the two populations in equilibrium is determined by
N −1
⎡ −g β H ⎤
2
= exp ⎢ ⎥ (Eq. 7.17)
N +1 ⎣ κT ⎦
2
The NMR signal recorded in an NMR spectrometer increases with increasing ratio between N−1/2 and N+1/2 because
the absorption more and more predominates over the emission. The signal intensity is also increased when the tem-
perature decreases. We have already shown that the population difference between two magnetic levels is given by
⎛ N g N βN H ⎞
Δn = ⎜ ⎟⎠ (Eq. 7.18)
⎝ 2κT
The NMR signal intensity is directly proportional to Δn, which varies inversely to absolute temperature and
directly to field strength H and concentration of nuclei in the analyte, N. Thus for any nuclei, the intensity of the
absorption signal at low temperature will be higher than at room temperature. The dependence of signal sensitiv-
ity on magnetic field strength has led manufacturers to produce magnets with field strength as large as 14T.
The intensity (I) of the absorption signal as a function of the external magnetic field of strength, H for Lorentzian
and Gaussian line shapes is given by
I0
L-line shape: I = (7.19)
T2 (H − H r )
2
1 +T 2
I 0 exp ⎡ −b ( H − H r ) T 22 ⎤
2
G-Line shape: I (7.20)
⎣ ⎦
Here, T2 is the spin-spin relaxation time, also called the line-width parameter, and will be discussed under spin-
relaxation processes, and b is a constant.
For a given experimental set-up at constant temperature, the signal strength at a given frequency v is propor-
tional to
N (I + 1) mI n 2 7.21)
(
and at a given field H,
⎡ I + 1⎤
N ⎢ 2 ⎥ μI3 H 2 (7.22)
⎣ I ⎦
These expressions indicate that for the same value of H, proton nuclei are ≈4000 times more sensitive than
C13 nuclei.
7.12 MEASUREMENT OF NMR SIGNAL INTENSITY

The absolute intensity of an NMR line is a complex function of sample parameters, instrument parameters and
operating conditions and hence difficult to measure. However, the relative intensities of different bands in a
typical NMR spectrum often follow general rules. The integrated intensity of NMR signal represents the total
energy absorbed by the nuclei in an analyte at resonance conditions and is expressed by the total area under
the resonance curve. This area is proportional to the number of nuclei, say protons, of a particular functional
group in the sample and hence can be used in their determination. The area under the curve can be measured by
(i) paperweight method, and (ii) integration method. The latter, however, is widely used in NMR spectroscopy.
All modern NMR recorders are equipped with electronic or digital integrators to measure areas under absorp-
tion peaks.
7.13 SPIN-RELAXATION PROCESS

The probability of transition between allowed energy states is proportional to the radio-frequency energy density.
The signal intensity grows to a maximum with the growth of incident radiowave energy. The further increase of
power of radio frequency decreases and broadens the signal and ultimately it results in the disappearance of the
signal. This is due to the saturation process where the population in the lower and upper nuclear magnetic levels
become equal, i.e. N–1/2 = N+1/2. At this stage, the probability of upward and downward transition is equal and hence
no further absorption can take place. The return of the spin system to equilibrium is known as relaxation process.
The relaxation processes are exponential and are characterised by ‘relaxation times’.
MT = Me−T/l (7.23)
MT
where MT is the bulk magnetisation after time T and l is a relaxation constant. So for T = l; = 1 e . Thus, the
M
time during which the nuclei spin in the excited state dissipate a fraction 1/e = 0.37 of the excess spin energy to
the environments or other spins in order to return to the ground state without emission of any energy is called the
relaxation time.
7.13.1 Types of Relaxation Processes

The two spin relaxation processes which influence the equilibrium ratio N−1/2/N + 1/2, and are of interest to chemists
are: (i) spin-lattice (s-l) process, and (ii) spin-spin (s-s) process.
(a) s-l Process The s-l relaxation process is due to the interaction of the spin of the system with its environ-
ment, i.e. lattice in case of solids. The term ‘environment’ also refers to the solvent, electrons in the system and
other different types of atoms or ions in the system. The s-l mechanism involves the thermal equilibrium of the
spin system with the atomic vibrations of crystal lattice or tumbling motion of liquids or gases. Consequently,
the transference of energy from the excited state to the environments occurs and the spins return or relax back
to the lower state and the population difference is maintained. In other words, we can say that the radiationless
transition from equilibrium is called s-l relaxation since the excess energy passes from the spins to lattice, i.e.
surroundings, as heat.
The s-l process is characterised by s-l relaxation time, also called longitudinal relaxation time T1 which is a
measure of the average lifetime of the nuclei in the higher energy state. It is the time taken for a dissipation of
fraction 1/e = 0.37 of the excess energy to the environments. T1 for specific nuclei not only depends on the isotope
and its chemical environment but also on the physical state of the analyte. T1 is of the order of 10−2−104 second for
solids and 10−4−10 second for liquids. The shorter time for liquids is because of free motion of molecules which
causes large fluctuations of magnetic field in the neighbourhood of the nuclei. The various nuclei constituting the
gaseous or liquid lattice are in violent vibrational and rotational motions. Such a motion generates a complex field
around each magnetic nucleus. The resultant complex field has a continuum of magnetic components. The phase
and frequency of some of the components correspond to the precessional frequency of the nuclei in the analyte.
Thus, these vibrationally and rotationally developed components exchange energy with the spins in the excited
state and the nuclei move to the lower spin state, the absorbed energy then simply increases the amplitude of the
thermal vibrations or rotations. This change produces a very small temperature rise in the analyte. In crystalline
solids and viscous liquids, where the mobility of the lattice is low, T1 is large.
(b) s-s Relaxation Process The s-s relaxation process occurs when two
neighbouring nuclei of the same kind having identical precessional frequen-
cies but different quantum numbers, interact with each other. In the process, the
nucleus of one atom in the higher energy level imparts its energy to an atom in ΔE
the lower energy level. Consequently, the nuclei in the lower state moves to the
higher state while the excited state nuclei relaxes back to the lower energy state.
Nevertheless, the net number of nuclei in the excited state remains unchanged,
i.e. population ratio is not affected. However, the average lifespan of a particular
excited nucleus is shortened. Fig. 7.6 Spread of nuclear spin energy
The s-s process is characterised by s-s relaxation time, also called transverse of the upper and lower states due to
relaxation time, T2. Because of s-s interaction, the energy levels corresponding spin-spin interaction.
to the upper and lower states are not well defined but they spread over a certain
value of spin energy which is represented in Fig. 7.6. The time required for the
spin energy to spread over a certain value is called the s-s relaxation time. For solids, T2 is of the order of 10−4
second and for liquids T1 = T2.
Both the s-s and s-l mechanisms increase the line width of the NMR signals and are related to line width parameter
t by the expression
1 1 1
= + (7.24)
τ T1 T 2
In addition to these two types of relaxation processes, another type which is of little importance to chemists is
the spin-rotation (s-r) relaxation processes.
(c) s-r Relaxation Process The interaction between the nuclear magnetic moment and the magnetic field gen-
erated at the nucleus by the rotation of a molecular magnetic moment arising from the electron structure of the
molecule is called spin-rotational interaction. The relaxation process in this case is due to the energy transfer from
the nuclear spin system to the molecular rotation. This mechanism is important in spin ½ nuclei in the gas phase
and in liquid phase generally with small molecules at relatively higher temperatures. As already has been pointed
out, s-r relaxation process is of little use to chemists. However, the nuclear spin-internal rotation (s–int-rot) is of
interest to chemists. Here, the internal rotation of a group relative to the larger molecule to which it is attached is
the source of relaxation. The s-r and s-int-rot mechanisms become more important at higher temperatures. The
s-int-rot relaxation process is important in 1H–NMR of freely rotating −CH3, 31P−NMR of freely rotating–OPO3,
proton decoupled 13C–NMR of freely rotating-CH2X and 19F– NMR of free rotating –CF3, etc., groups attached
to larger molecules.
7.14 WIDTH OF NMR SIGNAL AND RELAXATION TIMES

According to the Heisenberg uncertainty principle,
h
ΔE Δt = = 10 −34 J s
2π
On rearrangement, we get
ΔE 1 01
= Δv = = cps
h 2πΔt Δt
where ΔE is the energy spread over an excited state (not well-defined) and Δv is the corresponding frequency spread,
Δt is the lifetime of the state; Δt = 1/2pΔn. Δn is the uncertainty in energy of the excited state and manifests itself
in the width of the NMR signal. This Heisenberg bandwidth of the NMR signal is influenced both by the s-l and s-s
relaxation times, i.e. T1 and T2. When T1 and T2 are large, i.e. lifetime of nuclear spin state is large; Δn is large. For
a typical liquid, T1 = T2 = 1, Δn = 0.1 cps and for solids T2 = 10−4 second and hence Δn = 1000 cps. Thus, inherent
NMR bandwidth is large in solids as compared to liquids. Hence, solids rarely give high-resolution spectra. Some
liquids may give spectra having broad lines, provided T1 and T2 are small. Such liquids are generally viscous or con-
tain paramagnetic ions which increase the efficiency of relaxation processes. In general, we can say that the natural
width of NMR spectral line varies inversely to the average time the system spends in the excited state.
In addition to T1 and T2 the applied field H also contributes to line broadening in two ways provided it differs
slightly from nucleus to nucleus. In such cases a band of frequencies, say v ± δν1 v ± δν 2 , v ± δν 3 ,.... rather than
a single frequency v is absorbed by the nucleus of interest. First, the presence of other spatially fixed neighbouring
magnetic nuclei present in the analyte whose spins generate local magnetic fields that may interact with H to increase
or decrease its impact on the nucleus under consideration. In gases and liquids, these local magnetic fields cancel
each other since the nuclei producing them are in violent motion. However, in solids and viscous liquids, a range of
field strengths, i.e. H δ H 1 H ± δ H 2 , H δ H 3 ,.... .is generated by the local magnetic fields,. δ 1 δ H 2 , δ H 3 ,.....
and thus a corresponding range of absorption frequencies, v + δ v 1 v + δ v 2 , v + δ v 3 ,.... . Second, the variations in
the static magnetic field may occur if the applied field, H, is not homogeneous. This affect is minimised by spin-
ning the analyte in the magnetic field.
7.15 FOURIER TRANSFORM NMR (FTNMR) SPECTROSCOPY

Our discussion on FTNMR is based on a rotating reference frame rather than a fixed laboratory reference frame.
The rotating reference frame is one in which the Z-axis is the field direction but the X- and Y-axes rotate about the
Z-axis at a frequency ω γ H . The angular component due to precession in the laboratory frame becomes fixed;
we have, in effect, switched off the H-induced precession.
Due to thermal motion, the magnetic moments associated with nuclei spins in a sample are randomly oriented.
Since the magnetic moment is quantised on the application of an external magnetic field of strength H, the magnetic
moments of nuclei (denoted by narrow arrows) orient themselves randomly in the direction of the applied field which
is defined as the Z-axis as shown in Fig. 7.7(a). The net resultant magnetic moment, i.e. bulk magnetisation M (shown
by broad arrow) of these spins is static with the field direction [Fig. 7.7(a)]. Now when a p/2 pulse of radio frequency
field of strength Hrf (also static in the rotating reference frame) is applied in a direction perpendicular to the Z-direction
which is taken as the X-axis, it will strike the sample, due to which the bulk magnetisation will sense with each pulse
a torque and is tipped from the Z-axis into the Y-Z plane by an angle qf (the tip angle or the flip angle) given by
θf γ H rf τ (7.25)
where qf is the angle of rotation in radians and t is the duration of the pulse. The urge of the net magnetic moment to
move away from the Z-axis at the instant the pulse just strikes the sample is shown in Fig. 7.7 (b), whereas the rota-
tion of M due to torque of radio frequency field around X-axis in the Y-Z plane is presented in Figs. 7.7(c) and (d).
We can determine the relationship between duration of the pulse or pulse width t and the tip angle by varying
the length of the pulse and examining the resulting signal intensity. The signal intensity will be maximum for a
tip angle of π 2, a null at p when magnetisation is along Z-axis, a negative maximum at 3π 2 and return to null
at 2p with magnetisation back at Z. For a given pulse, τ 1 10 μs , = π 2 radians.
Further, on the termination of the pulse, the torque due to the applied magnetic field causes the relaxation of
the nuclei in the Y-direction to return to their equilibrium position instantaneously as shown in Fig. 7.7(e) and
then to the original state after several seconds via s-s and s-l relaxation processes operating independent of each
other as depicted in Fig. 7.7(a). The relaxation of the nuclei is recorded as a function of time and the signal due
to relaxation process is called free induction decay (FID)—free of the influence of radio frequency field along
X-axis and decaying back to the original state. The FID is then converted to the normal frequency domain signal
by a mathematical tool, of Fourier transformation.
When a nucleus returns to its equilibrium state as shown in Fig. 7.7(e), there is a decrease in the magnetic
moment My along the Y-axis and an increase in the magnetic moment Mz along the Z-axis. The mechanisms of
Z Z
M Z
M
M
θf
Y
Y Y
X H rf
X X H rf H
H H
(a) (b) (c)
Z
M
Mz θf
M
Y Y
MY M
X H rf H =0
X H rf H
(d) (e)
Fig. 7.7 Behaviour of magnetic moments of nuclei in rotating field of reference.
s-s and s-l processes now in the stationary frame of reference are shown in Fig. 7.8. In the first-order relaxation
process, i.e. s-l relaxation process, the magnetic moment along Z-axis grows until it returns to its original value
as shown in Fig. 7.8(a). On the other hand, in s-s relaxation mechanism, exchange of energy takes place between
the nuclei spins, due to which some nuclei precess faster than the Larmor frequency while the others precess
slowly. Consequently, the spins begin to fan out or dephase in the X-Y plane as shown in Fig. 7.8(b). Ultimately,
the magnetisation along Y-axis reduces to zero. There is no residual component of M in the X-Y plane by the time
relaxation is complete along the Z-axis, which means T2 ≤ T1.
The s-s relaxation rate constant R2 in the X-Y plane is related to the relaxation time constant T2 by the relation
1
R2 = (7.26)
T2
The factors which contribute to the signal decay may be intrinsic to the molecules, such as exchange processes,
or may be purely experimental such as poor skimming of magnetic field. The intrinsic relaxation rate and time con-
stants are denoted by R2 and T2 respectively, while the actual rate and time constant of decay observed in FID are R *2
and T 2* respectively. For large R *2 or small T 2*, the line in the spectrum is broad. Conversely, if the line is sharp, R *2
is small and T 2* is long, the s-s relaxation is slow. The resonance line width Δv 1/ 2 and R *2 are related by
R* 1
Δv 1 2 = 2
= (7.27)
π T 2*π
1
or T 2* =
πΔv 1 2
7.16 TIME PERIODS INVOLVED IN MULTIPLE PULSE FT

In multiple pulse sequences, three distinct time periods: (a) preparation period, (b) evolution period, and (c) detec-
tion period are involved. The spin system is perturbed from its normal position by a pulse or pulses or by some other
effect such as decoupling in the preparation period. After perturbation, the system is allowed to evolve by relaxation
processes for some period called evolution period (t1) and then detected with a pulse called the detection or acquisi-
tion pulse (t2) to find out what happened to the system in the evolution period. The evolution is not detected directly
but may be mapped by studying the system with different evolution periods. The spectrum that is finally obtained
depends on the nature of preparation and on the length of the evolution period. Thus, the mechanisms affecting the
magnetisation during the evolution period and those affecting it Z

during the acquisition period can be studied by multiple pulse
sequences.
Y
7.17 THEORY OF MEASUREMENT OF
RELAXATION-RATE CONSTANTS X
BY MULTIPLE PULSE FT
Z
Let us now examine the effects of multiple pulse sequence
Z
and intervening delays on simple spin systems to detect the T2
rates at which the magnetization returns to its normal position T1
after perturbation by a pulse. Based on the nature of multiple
pulse sequences, the methods to determine the s-l and s-s rate M Slower
constants are termed as inversion recovery and spin echo, Y Y
respectively.
Faster
X
X
Inversion Recovery When a p pulse is applied in a direc-
tion perpendicular to the bulk magnetization of the sample
along Z-axis, i.e. X-axis, it will rotate M through p-radians,
the bulk magnetisation is inverted from M intially aligned
along Z-axis to –M along –Z as shown in Figs 7.9(a) and (b)
respectively. The excited nuclei will give up their energy to
the surroundings because of which M will become smaller,
pass through zero and then grow upwards until it attains its
original position. This relaxation of bulk magnetisation from
the inverted (–M) state to the intial (+M) state is exponential
and is depicted in Fig. 7.9(c). There is no magnetization in
the X-Y plane and hence no signal will be observed by the
receiver coil along Y-axis.
In order to detect the relaxation process of bulk magnetisation
from –M to + M, we must have its component in the X-Y plane.
This is generated by a p/2 pulse which rotates any Z- magneti-
sation into X-Y plane producing a signal amplitude A, which is
proportional to the remaining Z-magnetisation. The p/2 pulse
transfers information from the Z-axis where it is not observable
into the X-Y plane where it can be detected by the receiver coil.
Now the s-l relaxation mechanism may be detected with a pulse
sequence represented as π t D π 2 − Acquire (AQ) and shown
in Fig. 7.10. Here, p is the preparation pulse, p/2 is an acquisition
pulse and tD is the delay which is under the control of the operator.
(a) (b)
The width of p pulse is twice that of the p/2 pulse. The p pulse
Fig. 7.8 Relaxation processes in the X-Y plane: (a) Longi-
is a nonselective, ‘hard’ pulse applied with the high-power trans- tudinal/spin-lattice relaxation, and (b) transverse/spin-spin
mitter, and tD is of the order of milliseconds to seconds while the relaxation.
length of the p pulse is of the order of a few milliseconds. Note
that the notations (π )x d (π )− x indicate that p pulses are 180° out of phase with each other.
Let us now visualise the inversion-recovery experiment through vector diagrams. For short tD, the magnitude
of inverted M will almost remain unchanged and the p/2 pulse will rotate it to the Y-axis as shown in Fig. 7.9(d).
The signal is inverted compared with the signal produced by a p/2 pulse on the Z-magnetisation. Now if tD is
lengthened before applying the p/2 pulse, the magnitude of M will decrease as shown in Fig. 7.9(e). The p/2 pulse
will rotate the magnetisation in the –Y-axis producing an inverted signal with smaller amplitude as shown in Fig.
7.9(f). Further, if the delays between the p and p/2 pulses are prolonged, the magnetisation, will pass through
zero [Fig. 7.9(g)] so that p/2 pulse produces a + Y magnetisation, a positive signal [Fig. 7.9(h)] and ultimately
magnetisation will recover its original value M [Fig. 7.9(i)] and the p/2 pulse will produce a positive signal with
maximum amplitude as shown in Fig. 7.9(j). The effect of inversion-recovery sequence on bulk magnetisation
may also be presented by Fig. 7.9(c).
Z
M
π
π/2
Y
X
−Z
(a) (b) (d)
l/tD
π/2
Time
(c)
(f)
(e)
tD
π/2
(g) (h)
tD
π/2
(i) (j)
Fig. 7.9 Vector diagram for the inversion recovery pulse sequence.
π π/2
tD
AQ
Fig. 7.10 The inversion-recovery pulse sequence.
Now, if tD between p and p/2 pulses is longer, the system will relax back to its original state by the time the
p/2 pulse is applied. The amplitude A∞ of the signal will be the same when the p pulse is not applied. On the
other hand, if tD is small, the magnetisation will be inverted and its amplitude will be –A∞. For intermediate delays
between p and p/2 pulses, the exponential decay of the magnetisation is given by
At = A∞ [1 – 2 exp(– R1 t)] (7.28)
This expression may be used to determine the s-l relaxation rate constants.
These days, a computer attached to the spectrometer is used to control the whole process of measurement. It
times the original p pulse, selects increasing tD values, times the p/2 pulse, and collects and measures the signal
intensity. A hypothetical spectra of a single-spin system produced by inversion-recovery pulse sequence is shown
in Fig. 7.11 while in Fig. 7.12 is displayed the inversion recovery spectra of m-dinitrobenzene.
7.18 MEASURING OF S-L RELAXATION RATES

Relaxation rates depend upon solvent, temperature, concentration, paramagnetic impurities and method of mea-
surement. So, care should be taken about these parameters while measuring the relaxation rates. We will describe
here only the inversion recovery and saturation recovery methods for the estimation s-l of relaxation rates.
H2 H4,6 H5
M
Time of
Inversion
Time of the Pulse
tD (in s)
M
tD = 0(in sec)
Fig. 7.11 Hypothetical spectra of a single-spin system produced by an inver-
sion recovery pulse.
(a) Inversion Recovery This is the classical method

of measuring constants. The system is observed under the
influence of p–tD–p/2 – Acquire pulse sequence. tD is varied
in a series of experiments, If the original bulk magnetisation
is M, the magnetisation after p-pulse will be –M. The return of Fig. 7.12 250 MHz 1H inversion recovery spectra for
Z-magnetisation to the original state is governed by m-dinitrobenzene.
Mt = M [1 – 2exp (–R1 t)] (7.29)

At = A0 [1 – 2exp (–R1 t)] (Eq. 7.28)
Here, Mt/At, is the Z-magnetisation/amplitude at time t after the p-pulse. The signal amplitude At, detected in
the X–Y plane after the p/2 pulse, will grow exponentially from –A to A.
When, Mt = 0, Eq. (7.29) yields
R1 t = ln 2 or tn = 0.693 T1 (7.30)
where, tn is the null point. Thus, the relaxation time, T1 is about 1.5 times the time taken
CH2
for the signal to decay to zero intensity. The relaxation rates may be obtained from the DO
plot of ln(A0 – At) against tD. The slope of the straight line is –R1. However, this method of Ho Hm
determining R1 is not preferred since for latter points when the relaxation is least, A0 – At
will be small due to which the relaxation is likely to be non-exponential. So, it is better to fit the exponential curve
directly to obtain R1. Trial values of A0 are selected in order to obtain the best fit between observed peak intensities
and Eq. (7.28). The s-l relaxation rates for the two aromatic signals corresponding to H0 and Hm protons of peptide
tyrala, (I) are 0.48 s−1 and 0.91 s−1 , respectively.
(b) Saturation Recovery The preparation step used to measure s-l relaxation time constants may be saturation
rather than inversion of the original magnetisation. The saturation may be carried out by a series of hard pulses with
only short delays between them so that there is not enough time for the system to return to its original position. The
whole spectrum becomes saturated by this technique. This type of saturation is achieved by using p /2 pulses and a
delay scheme in which the time interval is decreased during the train pulses as shown in Fig. 7.13. The saturation of
proton magnetisation may also be done by a burst of broad band 1H decoupling for 1H relaxation measurements.
1
H
tD
AQ
Fig. 7.13 Pulse sequence for saturation recovery using a converging pulse train to ensure
saturation.
Once the saturation is attained, the system is allowed to evolve during the delay tD, and then detected with a p/2
pulse. The evolution is measured as a function of varying evolution delay, tD. The signal intensity is given by
At = A0 [1 – exp (–Rl tD)] (7.31)
The slope of the line between ln(Ao −At) and tD will give the value of Rl. The saturation-recovery methods are
less time consuming than inversion-recovery method. This method is particularly useful for slow relaxing nuclei,
i.e. nuclei with low sensitivity such as 15N.
7.19 SPIN ECHO

In order to measure s-s relaxation time constant T2, the s-s relaxation process is to be disentangled from other pro-
cesses which can cause the spin to precess at different frequencies, mainly the field inhomogeneities. The effect
of field inhomogeneity is that the chemically identical spins M
in different parts of the sample tube will experience different
effective magnetic fields. They will, therefore, precess at aver- π/2 tD
S
age of frequencies rather than a single frequency. The effect of M
inhomogeneity which appears in the form of line broadening is F δ
removed using a pulse sequence represented as p/2−tD−p−tD− (a) (b) (c)
Echo, where terms and symbols have their usual meanings.

When, a p/2 pulse is applied along the X-axis, via the trans-
mitter coil, it tips the bulk magnetisation M along Z-axis [Fig.
7.14(a)] to the Y-axis [Fig. 7.14(b)]. During the evolution time tD π
S
tD, the magnetisation decoupled into fast (F) and slow (S) com- S
ponents in the X-Y plane as shown in Figs 7.14(c) and (d), and δ F F δ
(e) (f) (d)
the signal observed in the receiver coil will result from the sum
of the components of magnetisation. As the individual com- Fig. 7.14 Vector diagram of the effect of the spin-echo
sequence on the signal which is broadened by field inhomo-
ponents fan out, the sum decreases and the intensity of the net geneity.
signal decay rapidly. At this stage, a p pulse is applied along the
X-axis and its effect is to rotate the components of magnetisation by 1800 about the X-axis [Fig. 7.14(e)]. The sum of
components and the resulting signal intensity will increase to form an echo of maximum amplitude at the end of delay
period equal to the first [Fig. 7.14(f)]. After that moment, the decoupling of magnetisation will start and the signal will
decay but at the peak of echo, all effects of field inhomogeneity have been removed. The decay process continues up
to 3tD and further application of p pulse will cause rebunching of the F and S components at 4tD. The whole process is
shown in Fig. 7.15, where (a) represents the spin-echo pulse timing, and (b) the detector output.
π/2 π π
On
Pulse
Off
tD 3tD Time
(a)
Signal intensity
2tD 4tD Time

(b)
Fig. 7.15 The pulse timing of p/2–tD–p sequence, and (b) the resultant spin echoes.
7.20 CHEMICAL SHIFT AND SHIELDING CONSTANT

The resonance condition hv = bN gN H is derived on the assumption that the interaction between nuclear magnetic
moment of bare nuclei, say proton, and applied field of strength H occurs. However, the bare nucleus seldom
exists. The nuclei in molecules are surrounded by electrons. The external field H induces currents in the electron
clouds of the molecules. These currents, in turn, pro- E = − gNbN HMI

E = −gN bNH(1−σ)MI
duce their own secondary magnetic fields at the sites MI = −1/2
of the magnetic nuclei which shift the resonance away
for the bare nucleus. This shift from the bare nucleus MI = ±1/2
is called chemical shift. The magnitude of the induced
Energy
ΔE = 0 ΔE = hν ΔE′ = hν′
secondary field is proportional to the external field H = g N β N H = g N bN H(1−σ)
H=0
and in isotropic environments, its direction is along
the external field. H. An Isotropic environment is
MI = +1/2
effectively produced in gases, liquids and solutions.
Under these conditions, the effective magnetic field Bare Nucleus—Shielded Nucleus
H
experienced by the nucleus is given by Fig. 7.16 A sketch depicting the comparison of the resonance
Heff = H −sH = H (1− s) (7.32) condition for a bare and shielded nucleus.
⎛ 1⎞
Thus, the resonance condition for a nucleus ⎜ I = ⎟ , i.e. proton, in going from a bare nucleus to the shielded
⎝ 2⎠
nucleus is given by
ΔE′ = hv′ = gN bN H (I − s) (7.33)
and is also shown in Fig. 7.16.
The dimensionless parameter s is known as screening or shielding constant of the nucleus in the environment
considered. The value of s for hydrogen nuclei in diamagnetic samples varies over a range of some 10–20 ppm. s
is somewhat greater for heavier nuclei because currents can be easily induced in their electron clouds. The shield-
ing constant is, therefore, a function of molecular environment of the nucleus of interest, say proton, and is largely
determined by the density of the surrounding electron cloud.
Let us now consider two resonance lines due to two magneti-
cally non-equivalent protons, say A and B, as shown in Fig. 7.17.
The difference between resonance frequencies or fields of A and
B is called chemical shift. In other words, we can say that the
measurable distance between two resonance signals in terms of HA HB H
frequency or magnetic field is the chemical shift.
Fig. 7.17 Hypothetical resonance signals due to two
Applying Eq. (7.32) to both the nuclei, we get magnetically non-equivalent protons, A and B.
Heff (B) − Heff (A) = H (sA −sB) = HdAB (7.34)

H efff ( B ) − H eff
eff ( A )
where δ A B = × 106 is called the chemical shift.
H
Similarly, in terms of resonance frequencies of the two nuclei, i.e. vr (A) and vr (B), the chemical shift is
expressed as
v r (B ) − v r (A )
δAB = × 106 ppm (7.35)
v
Here, v (in MHz) is the operating frequency of the NMR spectrometer, a factor of 106 is introduced so that dAB
can be expressed in convenient numbers. dAB is measured in dimensionless units—ppm and is independent of the
strength of the applied magnetic field , H.
Quantitatively, the chemical shift should be measured with respect to some standard reference, and for protons,
chemical shifts are usually reported with reference to tetramethylsilane (TMS), (CH3)4 Si. TMS has twelve equiv-
alent protons and hence will give narrow intense line at a higher field than the signals of hydrogen atoms present
in most of the organic compounds. TMS is chemically inert, has low BP (300 K) and thus can easily be removed
from most of the analytes after use. Generally, 5 per cent TMS is added to the analyte. For aqueous solutions,
sodium 2,2–dimethy1-2 silapentane-5-sulphonate, i.e. (CH3)3 Si(CH2)3 SO3− Na +, called DSS is preferred since
TMS is insoluble in water. The CH3 groups of DSS resonate in the same position as those of TMS. Weak peaks due
to methylene protons interfere so in order to eliminate the undesirable peaks, DSS with methylene groups deuter-
ated is used. The other standards for protons are cyclohexane, tetramethylammonium ion, hexamethyldisiloxane
(HMDS). The chemical shift relative to TMS is given by
v v
δ analyte TMS
analyte TMS = × 106 ppm (7.36)
v
The chemical shifts of most of the protons fall in the range of 1–15 ppm. The chemical shifts of H-bonded protons,
NH, OH, SH, etc., are temperature, concentration, pH and solvent dependent. The NMR signals are often broadened.
H-bonding causes low field shift relative to unbonded state. Thus increase in d value is because of deshielding effect.
In other words, we can say that hydrogen bonding decreases the electron density around the proton, and thus moves
the proton to lower field, i.e. higher d values. Intramolecular hydrogen bonds are less affected by their environment
than are intermolecular hydrogen bonds. The up-field shift of the hydroxyl proton resonance of ethanol with increase
of temperature or on diluting the ethanol with CCl4 is attributed to breaking of intermolecular H-bonds. At present,
there are two scales for measuring chemicals — the delta (d) and the tau (t) scales. The d-scale is recommended for
PMR by American Society for Testing and Materials (ASTM, Manual of Feb. 1966, sponsored by Committee E-13).
The sign convention used in d-scale is that the frequency and chemical shift increase down the field. The other scale,
called t-scale, is contrary to that of d-scale is due to Tiers (GVD Tiers, J. Phys. Chem. 62 (1958), 1151) and is recom-
mended by the Chemical Society of London. The parameter t is defined as
τ δ (7.37)
Standard Reference and Chemical Shift of Some Other Nuclei Standard references for 13C-nucleus are CS2
and TMS. 13C-chemical shifts with respect to TMS can be converted to some other standards where the up-field
shifts have been regarded as positive (contrary to the convention for proton chemical shifts) by
d (TMS) = −[d (C6 H6)ext −129]
= −[d (CS2)ext −194] (7.38)
= −[d (CH3 C* OOH)ext− 179)]

CF4 and trifluoroacetic acid are used as standards for 19F while 85 per cent phosphoric acid is used for 31P
nucleus.
The range of chemical shifts for other nuclei is greater because of associated 2p-electrons, e.g. 13C-chemical
shifts lie in the range 0–220 ppm. In some cases, the range may extend to 400 ppm or even more. For 19F, the range
may be 800 ppm while 300 ppm or more for 31P nucleus. Chemical shift for 35Cl and 37Cl cover a range of about
1000 ppm. The range of chemical shifts of 205Tl and 207Pb exceeds even 14000 ppm or more.
The chemical shifts of some typical protons are listed in Table 7.3 Chemical shift is taken as the centre of the
peak or of the multiplet (see Figs. 7.21 and 7.22).
Table 7.3 d and t values of some typical protons.
Types of Proton τ (δ ), ppm Types of Proton τ (δ ) , ppm

Si(CH3)4 10(0) F−CH3 5.7(4.3)
CH4 9.8(0.2) Cl−CH3 7.0(3.0)
R−CH3 9.1(0.9) Br−CH3 7.3(2.7)
R2−CH2 8.7(1.3) I−CH3 7.8(2.2)
(R)3 C−H 8.5(1.5) R−CH2−OH 6.4(3.6)
O
CH3 C 7.9(2.1) R2−CH2−OH 6.4(3.6)
OH
CH3NO2 5.7(4.3) (R)2−CH−OH 6.1(3.9)

CH3NH2 7.9(2.1) R2C = CH2 5.4(4.6)
CH3−Ar 7.7(2.3) (R)2C = CHR 4.8(5.2)
R−C ≡ C−H 7.7(2.3) Ar−H 2.7(7.3)
O C CH3 H
7.7(2.3) R C 0.1(9.9)
R O
R−O−CH3 6.7(3.3) RSH 9−8(1−2)

ArSH 7−6(3−4) ArOH 5.5−3.5(4.5−6.5)
9.5−5.5 ArNH 2 ⎤
ROH ⎥ 7−4(3−6)
(0.5−4.5) ArNHR ⎥⎦
Table 7.3 d and t values of some typical protons—Cont'd.
RCONH 2 ⎤
RNH2, RNHR 9−5(1−5) ⎥ 5−(−2)(5−12)
RCONHR ⎥⎦
C N OH 1−(−2)(9−12) RCOOH 1−(−3)(9−13)
C O… H O C 3−(−3)(7−13)
Induced
Benzene Acetylene
Fields
Electronic
Current H
C
H C C H
C C
H H
σH
σH
H
H
(b) (a)
Fig. 7.18 (a) Shielding of protons of acetylene, and (b) deshielding of protons of benzene.
In the case of metal derivatives the proton attached, to the carbon atom is shielded much more than the proton
of TMS. Thus the chemical shift has a negative value, but the influence of metal atom on the neightbouring
carbon atoms decreases sharply in comparison to the electronegative element. For examples, the d for the protons
attached to the carbon atom immediate to the metal atom is having a negative d with respect to TMS. But these
values become positive for the protons attached to the second carbon atom, as shown below:
CH3Li CH3CH2Li (CH3−CH2)2 Mg
−1.30 1.33−0.99 1.26−0.64
There are three types of contributions to the screening constant, i.e. diamagnetic (sd), paramagnetic (sp) and
solvent (ss). Thus,
σ = σd + σ p + σs
s is related to the induced precession of electrons bonding the nucleus say proton, around the nucleus in a
plane perpendicular to the magnetic field. This induced circulation of electrons generates a secondary magnetic
field opposed to the external field H. This diamagnetic effect results in shielding the nucleus by electrons from the
external field. Consequently, the external field is increased to cause resonance.
Diamagnetic shielding of protons of acetylene is shown in Fig. 7.18 (a). In acetylene bond, the symmetrical
distribution of p-electrons about the bond axis permits electrons to circulate around the bond. sd is difficult to be
computed but qualitatively it is proportional to the charge densities. The electron-loving (electronegative) groups
or atoms adjacent to protons reduce the density of the electron cloud around the protons of interest and hence
the shielding constant. The resonance, therefore, occurs at lower modified external field or higher frequency. The
most common electron attracting groups are NO2, F, Br, I, Cl, CN, OR, COOR. The reverse effect is exhibited
by electron repelling (electropositive) groups such as CH3, NH2. Protons of tetramethylsilane, i.e. (CH3)4 Si are
highly shielded and, therefore, are observed at high magnetic fields. Si is relatively positive.
sp is related to anisotropy in the electron distribution around the nucleus. The electron circulation around the nucleus
in such a case produces a magnetic field in the direction of the applied field. This is a paramagnetic effect and it results
in deshielding the nucleus. Consequently, the external field is decreased to cause resonance. Thus sp is negative. The
p-electrons of benzene may be thought of precessing around the ring and thus give rise to ring current which causes
deshielding. Deshielding will occur only if the external field is perpendicular to the plane of the ring. This effect is either
absent or self-cancelling in other orientations of the ring. Deshielding of protons of benzene is shown in Fig. 7.18(b).
Deshielding is also exhibited by ethylenic or carbonyl double bonds. p-electrons in these bonds are not sym-
metrically distributed along the bond axis as in the case of acetylene. Thus, the circulation of p-bonds around the
bond axis is prohibited by the nodal plane in the distribution of a double bond. p-electrons in these types of bonds
can be thought of circulating in a plane along the bond axis, H

σH Electronic Current
when the molecule is oriented with the field as shown in Fig.
7.19. Further, it is observed that the observed chemical shift
(d) for a given nucleus in a molecule in dilute solution is dif-
ferent from the shift in the gaseous phase (dgas). The difference H H
between these two chemical shifts is called the solvent screen- C C
ing constant (ss). It depends on the nature of the solvents and
causes deshielding. H H
ss = dgas −d (7.39) Induced Field
ss in part has contributions from the bulk diamagnetic sus- σH

ceptibility difference of the solution and reference samples Fig. 7.19 Deshielding of protons of ethylene.
(sb), the solvent magnetic anisotropy (sa) and weak dispersion,
i.e van der Waals forces between solute and solvent molecules (sw). In case of polar solutes, a polar effect (se) also
contributes to the solvent screening constant. This effect arises due to the secondary electric field in the solvent
owing to its polarisation by the permanent dipole moment of the solute. The specific complex formation in solu-
tion, say solute–solvent complex (sc) may also contribute to ss . We may, therefore, write
ss = dgas − d = sb + sa + sw + se + sc (7.40)
sb can be eliminated using unreactive internal reference. In favourable cases, sw, sa and se are small as com-
pared to sc. sc is eliminated by measuring the chemical shift of solute molecules over a wide range of concentra-
tions and extrapolating to zero.
7.21 ESTIMATION OF PROTON CHEMICAL SHIFTS IN COMPOUNDS

OF KNOWN STRUCTURES
The chemical shift of a proton bonded to carbon follows the additivity rule; the value of shift is the sum of the
contributions along the other three bonds to the carbon, together with the effects of nearby unsaturation. Thus, the
NMR shifts of protons attached to carbon in compounds of known structures can be estimated from the deshield-
ing values of atoms, bonds and rings recorded in Table 7.4.
Table 7.4 Deshielding values of atoms, bonds and rings.
System Deshielding
values (ppm)
(a) B, Si, Sn −0.3
H 0.2
C, P 0.7
N, S 1.7
O, Cl, Br, I 2.7
F 3.7
(b) Double bond If the carbonation participates DB 4.2
If one carbonation intervenes dB 0.8
Benzene ring; also, pyridine, and diazines
If the carbonation participates BZ 6.0
If one carbonation intervenes bZ 1.2
Additional values, more subject to steric effects than those

(c) TB 2.1
above:
tb 0.8
Triple bond If the carbonation atom intervenes TB 2.1
If the carbonation intervenes tb 0.8
Four-membered ring R4 0.7
Three-membered ring R3 –0.7
Table 7.5 Calculated and observed chemical shifts for some selected molecules.
System Proton(s) Contributors to Chemical Shift (d)

Chemical Shift (in ppm)
Calculated Observed
Methanol: CH3OH
(a) H, H, O 3.1 3.6
(a)
Ethanol: CH3CH2OH
(a) H, H, C 1.1 1.2
(a) (b)
(b) H, C, O 3.6 3.7
Tetramethyltin:(CH3)4 Sn
(a) H, H, Sn ( = B) 0.1 0.1
(a)
Sodium 2, 2-dimethyl-2-silapentane (a) H, H, Si 0.1 0.0
sulphonate: (b) H, C, Si 0.6 0.6
(CH3)3SiCH2CH2SO3(−)Na( + 1) (c) H, C, C 1.6 1.9
(a) (b) (c) (d) (d) H, C, S 2.6 2.8
1, 2, 5, 6-Tetrahydropyridine: 2.4 2.1
(a) H, C, C, db
H (e) 2.6 3.0
C (b) H, C, N
(a) H2C CH (d)
3.4 3.3
(c) H, C N, db
(b) H2C CH2 (c)
N 5.6 5.7
(d) C, C, DB
H 5.6 5.8
2-Ethyeprrole (e) C, C, DB 5.6 5.8
(a) H, H, C 1.1 1.2
(c) H H (d)
(b) H, H, C, db 2.4 2.6
CH3CH2 N H (e) (c) C, C, DB, db 6.4 6.0

(a) (b) H
(d) C, C, DB, db 6.4 6.2
Tolnene (e) C, N, DB 6.6 6.7
CH3 CH3C6H5
(a) (b)
(a) H, H, C, bz 2.3 2.3
(b) C, C, Bz 7.4 7.2
Diphenyl-methylbromide (a) C, C, Br ( = O) 6.5 6.3

(C6H5)2CHBr
bz, bz
(a)
1-propyl-4pyridylthioether (a) H, H, C 1.1 1.1
H H
(b) H, H, C 1.6 1.7
CH3CH2CH2 S N (c) H, C, S ( = N) 2.6 2.9

(a) (b) (c)
H H (d) C, C, BZ 7.4 7.1

(d) (e)
(e) C, N, BZ 8.4 8.4
2-Pyropyne-1-ol
(a) C, TB 2.8 2.5
HC CCH2OH
(b) H, C, O, tb 4.4 4.3
(a) (b)
-Methoxypropionitrile (a) H, C, C, tb 2.4 2.6
CH3OCH2CH2CN (b) H, H, O 3.1 3.4
(a) (b) (c) (c) H, C, O 3.6 3.6

Table 7.5 Calculated and observed chemical shifts for some selected molecules.
System Proton(s) Contributors to Chemical Shift (d)
Chemical Shift (in ppm)
Calculated Observed
Trimethy-leneoxide
(a) H2C CH2 (b) (a) H, C, C, R4 2.3 2.7
(b) H, C, O, R4 4.3 4.7

(b) H2C O
Phenylcyclo-propane (a) H, C, C, R3 0.9 0.7

(a) H2C H (b)
(b) C, C, C, bz, R3 2.6 2.8
C
(a) H2C C6H5 (c) (c) C, C, BZ 7.4 7.1
Ethtyleneimine
(a) H2C
(a) H, C, N, R3 1.9 1.6
NH
(a) H2C
*in CDCl3
Within a saturated compound, the d value for a proton is the sum of the deshielding values of the other three
substituents on the carbon atom. For alkenes and arenes, the chemical shift of a proton on an unsaturated carbon
atom is the sum of the contributions from the other two substituents, plus the effect of p-electrons at the carbon.
Based on these rules, the computed values of chemical shifts are compared with the observed values of some
selected molecules in Table 7.5.
7. 22 DISPLAY OF NMR SPECTRUM

Equations (7.35) and (7.34) show that the NMR spectrum of an analyte can be recorded either by field sweep
or by frequency-sweep method, respectively. In the field-sweep method, the external field is varied and the
frequency is maintained constant while frequency is varied and applied field is kept constant in the frequency-
sweep method. The universally accepted mode of displaying NMR spectrum is a left (L) to right (R) arrangement
in which H, in a field-sweep spectrum increases from L-R and consequently n in a frequency-sweep spectrum
decreases from L-R. The NMR spectrum is obtained by plotting d against absorption intensity on the horizontal
and vertical axis respectively. The zero value of d scale corresponds to TMS signal. The hypothetical spectrum
of a proton is shown in Fig. 7.20.
Higher Frequency Lower Frequency
Paramagnetic Shift, Diamagnetic Shift

(Deshielding Effect) (Shielding Effect)
TMS
10 δ (in pprn) 0
Lower Field τ (in pprn) Higher Field
Fig. 7.20 A hypothetical NMR spectrum of a proton.
7.23 DIPOLE–DIPOLE (D-D) INTERACTION

Dipole–dipole interaction takes place directly through space. Let us consider two protons, say A and B, separated
by a distance r. The interaction between the two magnetic dipoles is given by
⎡I I 3(I A .r )(I B .r ) ⎤
H g N2 βN2 ⎢ A 3 B − ⎥ (7.41)
⎣ r r5 ⎦
where IA and IB are the nuclear spins of protons A and B respectively, Hd-d is the Hamiltonian for d-d interaction.
As a result of interaction between the two magnetic dipoles, there will be an additional magnetic field, Hadd, at the
proton A due to proton B and vice versa, and is given by
3g N βN 1 − 3 2
θ
H add = ± ⋅ (7.42)
4 r3
where q is the angle between the applied field H and the line joining the nuclei. The sign ± appears due to the fact
that the magnetic dipole of the proton B can be parallel or antiparallel to that of the proton A. Same is true for
proton A. Thus the effective field at each nucleus is
3g N βN 1 3 2
θ
Heff = H ± Hadd = H ± ⋅ (7.43)
4 r3
Consequently, the absorption spectrum will be a doublet with a separation of
3 1 − 3 cos 2 θ
ΔH = H eff − H = g N βN (7.44)
2 r3
e.g. the NMR spectrum of the single crystal of gypsum, CaSO4 . 2H2O, gives rise to four lines (two doublets). The
interaction between each pair of equivalent protons should only give rise to a doublet. However, we get two dou-
blets. This will lead us to infer that two water molecules in each unit have different orientations.
This type of magnetic interaction vanishes in nonviscous liquids due to the rapid tumbling motion of the mol-
ecules. When a molecule tumbles in solution, the angle ‘q’ changes continuously, and if the frequency of tumbling
is equal or greater than the energy due to d-d interaction (in frequency units), this effect will be averaged out to
zero. However when the correlation time for molecular reorientation is of the order of l0−11 second, only a very
weak line broadening effect remains of that d-d interaction. The broadening due to this effect becomes uncertain
or obscure normally if the external magnetic field is not homogeneous. d-d interaction in solids produces a struc-
ture having a line width of 10–20 kHz. This width is sufficient to smear out all structures caused by chemical
shifts and internuclear s-s coupling, i.e. J coupling. Thus, high-resolution NMR spectrum in solids is impossible
unless the d-d interactions are annihilated or reduced to minimum. If a nonspherical molecule is dissolved in a
liquid crystal matrix, the motional averaging may become slightly anisotropic and in that case d-d interactions
may produce additional structure in a high-resolution NMR spectrum. This forms the basis of applications of
liquid crystals in the study of d-d interactions and will be discussed later on under the heading ‘liquid crystals in
high-resolution NMR spectroscopy’.
7.24 SPIN–SPIN (S-S) COUPLING AND SPIN–LATTICE (S-L) COUPLING

CONSTANTS
So far we have considered only the interaction between the magnetic moment of a nucleus and applied field H.
However, in molecules, the protons may be closely spaced and the net interaction between these types of protons
may be due to (i) direct d-d interaction through space, and (ii) indirect s-s interaction through bonds. As we have
already discussed, the former averages out to almost zero in case of isotropic fluids. The isotropic s-s interaction
in liquids is mediated by the electron-nuclear and electron-electron interactions in the molecule. This interaction
imparts a fine structure to the NMR signals. The energy of s-s interaction is proportional to the product of spins
of the nuclei considered and is given by
E = JAB IA IB (7.45a)
where IA and IB are the spins of the coupled nuclei A and B respectively. JAB is the spin coupling constant between
A and B and has dimensions of frequency. The coupling constant is independent of applied magnetic field while
532
Table 7.6 Characteristic values of s-s coupling constants for some typical fragments.
Fragment J, cps Fragment J, cps Fragment J, cps
H2 (CH4)
HB J23 = 1.6–2.0
280 (12.4)
HA
CHA C 1–3 (4.6–5.8)
C
HB
12.5
OR J24 = 0.6–1.0
HA CHA C 2–3
CHB
C C (1.0–1.8)
Ha
HB 0.35–2 Ha
Jaa = 9–13 J25 = 1.3–1.8
He 3 4
CHA CHB
He He
Jae = 2–4 (2.1–3.3)
2–9 5
C C 2
He J34 = 3.2–3.8
HA HB He O (S)
Jee = 2.7–4
6–14 Ha He
HB (3.0–4.2)
Ha
C C
HA HB JAB = 7 –10
HA′ HB HA JAB = 3–4
11–18
CHB JAA' = 2–3 JAC = 0.6–1
C C HA HB′ HC R
HA 0.5–2 JAB' = < 1.5 O JBC = 1.8–2
HA JAB = 7.5–8
HB′ HB JAA' = 1.4–2
C CHA CHB C 10–13 H C C C H JHH = 0
JA'B = 5.2–5.5
HA′
JA'B' = 0.9
JHH = 6.5 JHD (provided the two protons are in the same position)
chemical shift is proportional to it. Theoretically, both posi- Br2CHCH2Br

a b
tive and negative values of JAB are possible. However, the b
appearance of NMR spectrum depends only on the absolute
magnitude of JAB, i.e. |JAB|. The sign and magnitude of JAB a
depends upon the number of bonds and the type of bonds JAB
through which nuclei interact in addition to the geometrical
orientations of those nuclei, e.g.
(i) J (C C) J (C C ) > J (C C )
Jab Jab
(ii) J (trans) > J ( cis )
(iii) J (axial-axial) > J (axial-equatorial) and J (equatorial-
equatorial)
(iv) J (FHCCl2) > J (HCCl2CCl2F) > J (HCCl2CCl2CCl2F)
The s-s coupling constants of some fragments of organic
molecules are listed in Table 7.6. Note that the coupling con-
Fig. 7.21 NMR spectrum of 1,1,2-tribromoethane.
stant is taken as the distance between any two adjacent peaks
in a multiplet as shown in Figs 7.21 and 7.22.
Further, it is to be noted that spin-coupling constants JAB (in Hz) between different combinations of nuclear
species can be meaningfully compared only when signs and magnitudes of the magnetogyric ratios of the nuclei
are accounted for. Thus, it is useful to consider a reduced coupling constant KAB which may be defined as
⎛ 2π ⎞ ⎛ 4π 2 ⎞
KAB = ⎜ J = ⎜⎝ γ γ ⎟⎠ J A B (7.45b)
⎝ γ γ B ⎟⎠
A B
B
where gA and gB are the magnetogyric ratios of nuclei A and B. In the cgs system, the unit of KAB is cm−3, in the SI
it is N A–2m−3, e.g. the reduced Cl coupling constant for
⎡J 1 35 = 41 ± 2 ⎤ −3
HCl ⎢ H Cl ⎥ is 35 ± 2 × 10 cm
20
J
⎣ H Cl
1 37 = 35 ± 2 ⎦
and is 19 ± 1 × 10 20 cm−3 for
⎡J 1 79 = 57 ± 3⎤
HBr ⎢ H Br ⎥
⎣ J 1 H81Br = 62 ± 3⎦
400 300 200 100 0 Hz
d c b a
d CH2 CH2 CH3
Jab ≈ Jbc
a
b
Jab
c
Jbc
Jbc
Jbc
8 7 6 5 4 3 2 1 0
δ (pprn)
Fig. 7.22 NMR spectrum of 3-phenylpropane.
7.25 TYPES OF S-S COUPLING CONSTANTS AND FACTORS

AFFECTING THEM
We know that s-s interaction between two nuclei takes place through bonds and is measured in terms of a constant
called coupling constant (J). In general, J decreases as the number of intervening bonds increases. Depending
upon the number of intervening bonds, s-s coupling is of two types: (i) short range, and (ii) long range. Long-
range interactions are common in p-systems. The number of intervening bonds in short-range interactions is two
or three and the magnitude of J in such type of interactions is of the order of 0–20 Hz. On the other hand, the
number of bonds through which long-range coupling occurs is four or five and the value of J lies in the range of
0–3 Hz. Further, on the basis of nature and geometrical orientation of nuclei (protons), short-range interactions
are characterised as vicinal, geminal and vinylic whereas long-range interactions are classified as substituted
benzenes (J = 0–10Hz), allylic (0–3 Hz) and homoallylic (0–2 Hz) couplings. In vicinal coupling (I), protons are
attached to the adjacent carbon atoms and the coupling constant Jvic is of the order of 5–10 Hz,e.g. in ethylene
JHH (gauche) = 2–4 Hz and JHH(anti) = 5–12 Hz. On the other hand, in geminal coupling (II, III), the protons are
attached to the same carbon atom and Jgem lies in the range 10–20 Hz. The vinylic coupling in the fragment, (IV)
is 4–10 Hz whereas 6–14 Hz for cis protons and 11–18 Hz for the trans protons in the (V) fragment. The coupling
constants are sensitive to various parameters such as bond angle, bond length, electronegativity, etc., and the
behaviour of these factors towards coupling constant is reported here.
HA θ HB HA HA H H H
θ′ 120°
θ′ C θ C C C C C
C C C H H
HB HB
(I) (II) (III) (IV) (V)
7.25.1 Vicinal Coupling ( Jvic)

(a) Effect of Dihedral Angle The vicinal coupling is very sensitive to the dihedral angle j between the planes
containing the protons and obeys the relation called Karplus equation:
JHH′ = A + B cos j + C cos 2j (7.46)
where A, B and C are constants.
or JHH′ = 8.5 cos2j − 0.28, 0 ≤ j ≤ 90 ( 7.47)
9.5 cos 2 ϕ − 0.28, 90 ≤ ϕ ≤ 180
The behaviour of JHH′ with j is shown in Fig. 7.23 which shows that JHH′ (transplanar protons, j = 1800) > JHH′
(cis-coplanar protons) > JHH′ (j = 900).
Equation 7.47 when applied to the chair form of cyclohexane indicates
aa (ϕ aa = ) = 9 Hz > J (ϕ ae ae = ) = 1.8 Hz = J (ϕ ee ee = ) = 1.8 Hz

10 10
ϕ H′ H
The observed values of J for cyclohexane in the chair
form are: J aa = 10–13 Hz, Jae = J ee = 2.5 Hz. A modified 8 C C 8
version of the Karplus equation when applied to vicinal
coupling in olefins, H C C H′ shows, that JHH′ (cis j = 0) 6 6
JHH′
h = 7–11 Hz > JHH′ (trans, j = 180°) = 12.18 Hz. (Hz) 4 4
2 2
(b) Effect of Electronegativity In a freely rotating
alkyl chain, CH3CH2CH2X, where X is an electronegative
substituent atom, i.e. X = Cl, F, etc., the behaviour of J 0 20 40 60 80 100 120 140 160 180°
with electronegativity is governed by the equation ϕ
Fig. 7.23 Behaviour of coupling constant (JHH′) with dihedral
JHH′ = 7.9 – 0.7n Δχ (7.48) angle (j).
Here, n is the number of substituent atoms, and Δc is the difference between the electronegativities of the sub-
stituent atom and proton, i.e. Δc = χ χ H . Here, cH = 2.10, and cCl = 3.0 and cF = 4.0. The equation fairly holds
good for di-and tri-substitution by electronegative atoms. According to this equation, the value of JHH′ should
decrease with the increase in the electronegativity of the substituent atoms in an alkyl chain provided their number
is maintained constant (n = constant), e.g.
JHH′ (FCH2 CH2F) < JHH′ (ClCH2CH2Cl)
and JHH′ (FCH2 CHF2) < JHH′ (ClCH2CHCl2).

On the other hand, if Δx is constant and the number of substituent atoms of the same kind are varied in the alkyl
chain, the coupling constant decreases with the growth of substituent atoms, e.g.
JHH′ (Cl−CH2−CH2−Cl) < JHH′ (Cl−CH2−CH−Cl2)
JHH′(F−CH2−CH2−F) < JHH′ (F−CH2−CH−F2)
Problem 7.8: With the aid of Eq. (7.48), compute the value of JHH¢ in (a) 1, 1, 2-trichloroethane, (b) 1, 1, 2-
tri-fluoroethane, (c) 1, 2-dichloroethane, and (d) 1, 2-difluoroethane.
Solution We know that cF = 4.0, cCl = 3.0 and cH = 2.10.

By Eq. (7.48),
JHH′ = 7.9 − 0.7 n Δc
For (a) and (b); n = 3,
Hence,
JHH′ (1, 1, 2-trichloroethane) = 7.9 − 0.7 × 3 (3.0 − 2.10) = 6.01 Hz
The observed value is 6 Hz.
(b) JHH′ (1, 1, 2-trifluroethane) = 7.9 − 0.7 × 3 (4.0 − 2.10) = 3.91 Hz
For (c) and (d); n = 2,
Therefore,
JHH′ (1, 2-dichloroethane) = 7.9 − 0.7 × 2 (3.0 − 2.10) = 6.64 Hz
JHH′ (1, 2-difluaoethane) = 7.9 − 0.7 × 2 (4.0 − 2.10) = 5.24 Hz
(c) Effect of Bond Length For constant bond angles and hybridisation, Jvic decreases as the C–C bond length
increases, e.g. cis olefinic coupling constant follows the reverse trend of p-bond order, i.e.
Jethylene(B.O = 1) = 11.5 Hz > Jbenzene (B.O = 0.67)
⎛ H 2 − H 3 B .O = 0.60⎞ 6 4 Hz ⎫
= 8 H > J nnaphthalene ⎜ = ⎬
⎝ H 1 − H 2 B .O = 0.72 ⎟⎠ 8 1 Hz ⎭
⎛ H 2 − H 3 B O = 0.70⎞ 5 5 Hz ⎫
> J ppyridine ⎜ = ⎬
⎝ H 3 − H 4 B O = 0.64⎟⎠ 7 5 Hz ⎭
(d) Effect of C−C−H Bond Angles The vicinal coupling constant is also very sensitive to the angles (θ d θ ′)
between the olefinic protons and in cyclic olefins obeys the reverse trend of the bond angles, i.e.
J(cyclopropene) = 0.5 – 2.0 Hz
< J(cyclobutene) = 2.5 – 4.0 Hz
< J(cyclopentene) = 5.1 – 7.0 Hz
< J(cyclohexene) = 8.8 – 10.5 Hz
In other words, as the ring size increases, the vicinal coupling also increases.
Problem 7.9: Predict the trend of vicinal coupling constants J23 in five-membered heterocylic ring compounds,
i.e. furan, pyrrole and thiophene. What will happen to J23 if carbon atom at the position 4 is replaced by nitrogen
atom in the thiophene ring?
Solution C = C bond length in five-membered heterocyclics is almost constant whereas the angle q at the posi-
tion 2 decreases slightly and the electronegativity of the hetero atom also decreases as we move from furan (cO,
134°) → pyrrole (cN, 131°) → thiophene (cS, 128°). The vicinal coupling constant follows the opposite trend of
electronegativity as well as of the bond angle. Therefore, J23 obeys the following order:
J23 (furan) < J23 (pyrrole) < J23 (thiophene).
When nitrogen is also present in the thiophene ring, the total electronegativity increases, consequently J23
decreases. These predictions conform to the observed values of J23 in the ring systems, i.e. J23 (furan) = 1–2 Hz,
J23 (pyrrole) = 2−3 Hz, J23 (thiophene) = 4.5−5.5 Hz and J23 (thiophene with N at position 4) = 3.4 Hz.
7.25.2 Geminal Coupling ( Jgem)

The geminal coupling is negative and opposite to that of vicinal which is nearly always positive.
(a) Effect of Electronegativity Coupling constant of methylene protons (CH3X) increases algebraically with the
electronegativity of the atom X, i.e.
J CH3 F (−9.6 Hz) > J CH3OH (−10.8 Hz) > CH4 (−12.4 Hz).
Direct determination of Jgem (HH) is not possible in this system. However, by deuteration of one of the equiva-
lent protons, Jgem (HD) is observed and the H-H coupling constant is then computed from the equation
JHH = 6.55 JHD (7.49)
On the other hand, Jgem follows algebraically the reverse trend of electronegativity of a substituent X, attached
to the carbon atom adjacent to the geminal protons in a series of monosubstituted 1, 1-dichloro-cyclopropanes, i.e.
Jgem(OAc) = –9.7 Hz < Jgem(COOH) = –6.8 Hz < Jgem(Si(CH3)4) = –4.9 Hz. The opposite effects of electronegativity
of atom X on Jgem, in these two different types of systems indicate that changes in Jgem cannot simply be attributed
to the direct inductive effect of the substituent.
(b) Effect of p -bonds The magnitude of coupling constant is very sensitive to the number of p-bonds adjacent
to the methylene or methyl group and decreases algebraically as the number of p-bonds increases, e.g.
CH4(0) = − 12.4 Hz > C6H5 – CH3 (1) = −14.5 Hz > CH3CN (2)
= −16.2 Hz > CH2 (CN)2 (4) = –20.3 Hz
The quantities within brackets correspond to the number of p-bonds in the respective systems.
(c) Effect of Bond Angle The geminal coupling constant is very sensitive to the angle q between the inter-
acting geminal protons, i.e. (I) Jgem varies from 0 at 125° to 20 Hz at 105°. In cycloalkanes, when q ′, i.e. ring
size, decreases, q increases, and consequently Jgem increases. Hence, Jgem (Cyclohexane) = −(12−14) Hz < Jgem
(Cyclopentane) = −(10−14) Hz < Jgem *(Cyclobutane) = −(7−14) Hz < Jggem(Cyclopropane) = −(4−9) < Jgem(II) =
0−3 Hz. The terminal methylene group in Jgem” which carbon atom is sp 2 hybridised, q = 120°. In strain-free cy-
clohexane and cyclopentane, the angle q = 109° is close to the tetrahedral angle in methane. Consequently, Jgem
of these compounds is very close to that of methane. Further, when one of the carbon atoms in cycloalkanes ring
is replaced by the oxygen atom then (i) Jgem, also increases with decrease in ring size and even becomes positive
for cyclopropane, and (ii) Jgem of the respective rings also increases as expected, since the coupling constant of
methylene protons of a system RCH2X increases algebraically with increasing electronegativity of the atom X.
H H
θ′ C θ C C θ
H H
(I) (II)
(d) Long-range Coupling Constant As mentioned earlier, in long range s-s interactions, the number of inter-
vening bonds between the interacting protons is four or five and the coupling constant lies in the range 0–3 Hz.
Long-range coupling occurs in substituted benzenes as well as in allylic and homoallylic systems.
Hp HA
(i) Substituted Benzenes The coupling constant in substituted benzenes decreases as the number
of intervening bonds between the interacting protons increases. Hm Ho
Thus
JH (Ao) (ortho) = 6 −10 Hz > JH(Am) (meta) = 1−3 Hz > JH(Ap) = 0 −1 Hz.
(ii) Allylic Coupling Allylic (H−C−C = C−H) coupling constants are around 0−3 Hz. In an unsaturated sys-
R1
R2
tem, H2 C H3, J1,3 is known as transoid allylic coupling constant whereas J2,3 is a cisoid allylic coupling
C C
H1 X
constant.
(iii) Homoallyic Coupling Homoallylic (H−C−C−C = C−H) couplings are usually negligible and may have
value about 1.6 Hz.
(iv) Coupling through Conjugated Polyalkyne Chains This may occur even through nine bonds.
(v) JAB Coupling Constant (≈7 Hz) in the bicyclo [2.1.1] hexane system, is attributed to the ‘W conformation’
of the four s bonds between HA and HB, i.e.
HA C HB
C C
7.26 CONCEPT OF MAGNETICALLY EQUIVALENT PROTONS

The nuclei which are indistinguishable in all their NMR interactions and can be permuted at will without
changing the physical state of the spin system are said to be magnetically equivalent. In other words, we can
say that the magnetically equivalent nuclei have the same chemical shift and the same s-s coupling constant
with every other nucleus that does not have the same chemical shift, e.g. all the protons of ethylene (H2C =
CH2) are magnetically equivalent. The NMR spectrum will exhibit only one sharp resonance line, centered at
the ethylene proton Larmor frequency. However, the protons of 13C-ethylenes (H213C = 12CH2) are not mag-
netically equivalent. The asymmetric isotopic substitution reduces the symmetry of the molecule, since the
directly bonded = 13CH2 protons belong to one group of symmetrically equivalent nuclei and the vicinal protons
(= CH2) belong to the second group of this type. The effect of 13C isotope on proton chemical shift is very
small, i.e, ±15 x 10−4 ppm. So, all the four protons will show one absorption peak. In spite of this, protons are
not magnetically equivalent since 13C (I = 1/2) affects the coupling constants of the two types of protons, i.e.
H(1)
H(3)
J( = 13CH2) = 156.4 Hz and J (= CH2) = 2.5 Hz. In case of
13 13
C C protons, H(1) and H(2) are not mag-
H(2) H(4)
netically equivalent with protons H(3) and H(4), because these two pairs of nuclei are unequally coupled to a
particular 13C nucleus. By definition, protons H1 and H2 will also not form a group of magnetically equivalent
nuclei since JH(1),H(4) (19 Hz) ≠ JH(2),H(4) (11.7 Hz). Two protons of F2C = CH2 are magnetically non-equivalent
since JHF (cis) ≠JHF (trans). All the three protons of methyl the group of ethanol are magnetically equivalent if
F
the rotation about the C--C bond is rapid. On the other hand, two protons of CH2 Ι are not equivalent even
Br
Cl
if the rate of rotation about the C–C bond is fast since both of them have different atoms as spatial neighbours.
Magnetic equivalence which arises due to free rotation about a single bond is temperature dependent. At room
temperature, all the three protons of CF3–CH3 are magnetically equivalent since the internal rotation about the
C−C bond produces averaged H−F spin couplings that are the same for all the possible HF pairs. However at
low temperatures, one particular rotational conformer of the molecule is in the frozen state. Thus, the three
protons and the fluorine nuclei would not form groups of magnetically equivalent nuclei since the vicinal H−F
spin interactions depend upon the dihedral F−C−C−H angle. None of the protons of pyridine are magnetically
equivalent since J23≠J36, J25≠ J56, etc.
Protons of furan and thiofuran are also non-equivalent since they have different coupling constants (see
Table 7.6).
7.27 MECHANISM OF S-S INTERACTION

In order to understand the mechanism of s-s interaction, consider a molecule with two different interacting protons,
⎛ 1⎞
say A and B. The proton A will have only two orientations in the external magnetic field, i.e. parallel ⎜ M 1 = + ⎟
⎝ 2⎠
⎛ 1⎞
and antiparallel ⎜ M 1 = − ⎟ to the applied field. The proton B thus will experience two fields (Heff − x) and
⎝ 2⎠
(Heff + x) where x is the correction due to the magnetic moment or field due to the nucleus A. Consequently, the
signal of the proton B will appear as a doublet. Similarly, the signal of the proton A will also split into two lines of
equal intensity. This may be represented in terms of spin and spin quantum number in the tabular form as follows:
MI Proton A Degeneracy MI Proton B Degeneracy
1 1
+ a I + a 1
2 2
1 1
− b I − b 1
2 2
For protons A and B, I = 1/2, M1 = (2I + 1) = 2 and the states corresponding to M1 = 2 are + 1/2 and –1/2.
Thus, the spectrum of AB system consists of two doublets, each of equal intensity, since the statistical weight
of +1/2 and –1/2 states is equal.
1
Let us now consider the s-s interaction in HD molecule, Here, we know that I(H) = and I(D) = 1 . Hence,
1 1 2
MI(H) = + , − and MI(D) = + 1, 0, −1.
2 2
The possible combinations of spin orientations of H and D in HD are
MI Proton A Degeneracy MI Deutron Degeneracy
1
+ a 1 +1 aa 1
2
1
− b 1 0 ab,ba 2
2
−1 bb 1
Thus, the signal due to a proton will consist of a triplet with intensity ratio 1 : 2 : 1 while the deutron will be a
1 : 1 doublet. The values of net spin moment ‘0’ are two times more probable than the values of + 1 and −1 in D.
On the other hand, the probability of + l/2 and −l/2 states of H is equal.
Let us now apply the mechanism of s-s interaction to C2H5NO2. It contains two different groups of magnetically
equivalent protons, i.e. –CH3 and −CH2. Thus, in this case, I(CH3) = 3/2; I(CH2) = 1. Consequently, ∑MI (CH3) =
2I + 1 = 2 x 3/2 + 1 = 4; + 3/2, + 1/2, –1/2, –3/2 and ∑MI (CH2) = 2I + 1 = 2 x 1 + 1 = 3; + 1, 0, −1. The possible
combinations of spin and spin quantum number for –CH2 and –CH3 groups are:
∑M I −CH2 Degeneracy ∑M I −CH3 Degeneracy
3
+1 aa 1 aaa 1
2
1
0 ab,ba 2 ααβ ,αβα , baa 3
2
1
−1 bb 1 − 3
2 ββα , βαβ , abb
3
− bbb 1
2
Thus, the signal due to protons of −CH3 group will split into triplet with intensity ratio of 1 : 2 : 1 while that of
–CH2 group will appear as a quartet with intensity ratio of 1 : 3 : 3 : 1.
Based on the above pattern of splitting of NMR signal into multiplets of same or varying intensity, some
empirical rules are deduced to interpret the first-order NMR spectra, J v δ . The theory of first and different
ordered NMR spectra will be discussed later on.
(a) Rule 1 Let nA, nB and nC be the number of magnetically equivalent protons on atoms A, B and C respectively
in a hypothetical chemical molecule A−B−C. If these three types of protons are magnetically non-equivalent, i.e.
they are in different chemical environments, the multiplicity of protons on A = (nB + 1) on B = (nA + 1) (nC + 1)
and on C = (nB + 1). If the neighbouring protons of B, i.e. A and C are magnetically equivalent, the multiplicity of
the signal is given by (nA + nC + 1).
(b) Rule 2 The relative intensities or areas of peaks that are symmetrical around the mid-point of the multiplet
are proportional to the coefficients of the bionomial expansion
n ( n − 1) 2 n ( n −1
1)( 2)
(1 x ) = 1 + nx +
n
x + x 3 + ...
2 3
Here, n is the number of protons on the atom adjacent to atom x.
Problem 7.10: Deduce theoretically the NMR spectra of the following: (a) CH3−CHCl−CH3; (b) CH2Br−CHBr2;
(c) ClCH2CH2Cl; and (d) CH3CH2OCH2CH3.
Solution
(a) In CH3−CHCl−CH3, both the CH3 groups are magnetically equivalent. Thus, the signal due to proton of
CH group will appear as a [(6 + 1) = 7] septet with intensity ratio 1 : 7 : 21 : 35 : 21: 7: 1 while that of protons
of end CH3 groups will be a doublet (1 : 1). CH protons being deshielded will be observed at higher d values as
compared to protons of CH3 groups, i.e. d CH > dCH3.
(b) In CH2Br–CHBr2, protons of −CH2 and −CH groups are in different chemical environments, hence they
are magnetically non-equivalent. Thus the signal due to protons of −CH2 group will appear as a doublet (1 : 1)
while that of −CH group will be a triplet (1 : 2 : 1). Protons of −CH group being highly deshielded as compared
to protons of −CH2 group, will be observed at higher d value, i.e. dCH > δ CH2.
1 1
(c) In ClCH2CH2Cl, the spin of protons on both the carbon atoms is the same, i.e. + = 1, hence they are
2 2
magnetically equivalent. Consequently, a single line spectrum without any multiplicity will be observed.
(d) Protons of −CH2 and –CH3 groups of identical ethyl groups of C2H5OC2H5 are magnetically non-equiva-
lent. Therefore, the signal of −CH2 protons will consist of a quartet (1 : 3 : 3 : 1) while that of −CH3 protons will
be triplet with intensity ratio 1 : 2 : 1. Protons of –CH2 group being highly deshielded as compared to protons
of −CH3 group, will be observed at low field, i.e. high d values, or d CH2 > d CH3.
Problem 7.11: What will be the number of signals and their intensity ratio in (a) toluene, (b) p-xylene, and
(c) mesitylene?
Solution
(a) There are two groups of magnetically non-equivalent protons in toluene, i.e. three methyl protons and
five benzene protons. Thus, the NMR spectrum of toluene will exhibit two signals, one corresponding to methyl
protons and the other corresponding to benzene protons. Since the intensity of an NMR signal is directly propor-
tional to the number of protons that cause it, so the ratio of areas under these signals will be 3(−CH3) : 5(−C6H5).
The methyl protons will be observed at lower d values than the protons of the benzene ring, i.e. d (−CH3) < d
(−C6H5).
(b) p-xylene will exhibit two lines spectrum, The ratio of areas under these two lines will be 2 (−C6H4) : 3
(−CH3−CH3) and d (−C6H4) > d (−CH3).
(c) Two bands corresponding to methyl and benzene protons will be observed in the PMR spectrum of mesi-
tylene. The ratio of areas of these two bands will be 3(−CH3)3 : 1 (−C6H3), and d (−C6H3) > d (−CH3).
Problem 7.12: Predict the number of signals in (a) ethylchloride, (b) isopropyl chloride, (c) n-propylchloride,
(d) isobutylene, (e) 2-bromopropane, (f) vinyl chloride, (g) methylcylopropane, and (h) 1,2-dichloropropane.
Solution According to the concept of magnetic equivalence of nuclei, the number of signals in (a), (b), (d) and
(e) is two while in (c), (f), (g) and (h) is three, three, four, and four respectively.
Problem 7.13: Predict and explain the NMR spectra of ethanol in the presence and absence of a mineral acid
or a base and compare it with the observed spectra displayed in Fig. 7.24.
Solution Hydroxyl protons in ethanol (CH3−CH2−OH) are hydrogen bonded, i.e. O−H...O. In the presence of
a mineral acid or a base, the movement of H-bonded proton gets accelerated and becomes too fast to be experi-
enced by the methylene group (−CH2). So in such a situation, the signal corresponding to the protons of methylene
group will appear as a quartet (3 + 1 = 4) with an intensity ratio of 1 : 3 : 3 : 1 due to the coupling of −CH2 group
with the protons of CH3 group. The signal of the methyl group will consist of a (2 + 1 = 3) triplet with an intensity
ratio of 1 : 2 : 1 due to the interaction of protons of −CH3 group with that of −CH2 group. A singlet is expected
corresponding to the proton of –OH group. Due to the inductive effect of −OH group, the signal corresponding to
a methyl group will appear up-field as compared to that of methylene group and −OH group, i.e. dCH3 < dCH2 < dOH.
The same has been observed experimentally [Fig. 7.24(a)]. In the absence of a mineral acid or a base, the spins of
protons of −CH2 group and of −OH group interact with one another. Consequently, the −OH group will appear as
a (2 + 1 = 3) triplet with intensity ratio of 1 : 2 : 1 whereas an [(3 + 1) (1 + 1) = 8] eight-line multiplet is expected
for the protons of −CH2 group. However, at moderate resolution a quintet has been observed experimentally
[Fig. 7.24 (b)].
1
These rules are also equally applicable to nuclei with spin I = , i.e. 13C, 31P, 19F, etc., other than protons in
2
AB and ABC (where A and C are magnetically equivalent groups) types of systems, e.g. 31P-NMR spectrum of
P(OCH3)3 consists of a ten-line multiplet with intensity ratio 1 : 9 : 36 : 84 : 126: 126 : 84 : 36 : 9 : 1 while that
(a)
H3C CH2 OH (In presence of HCI)
CH3
OH
CH2
TMS
4 5 6 7 8 9 10 τ
(b)
H3C CH2 OH (Pure)
CH3
OH CH2
TMS
6 5 4 3 2 1 0 δ
Fig. 7.24 NMR spectrum of ethanol (a) in presence of hydrochloric acid, and (b) in absence of hydrochloric acid.
of tripolyphosphate is a doublet (1 : 1) and a triplet (1 : 2 : 1). 13C spectrum of (CH3)2O is a quartet with intensity
ratio 1 : 3 : 3 : 1 while that of 13C6H6 is a doublet (1 : 1).
7.28 QUANTUM MECHANICS OF S-S INTERACTION

In the presence of an external magnetic field of strength H, the various magnetic interactions are given by the
Hamiltonian
H = Hzeeman + Hd-d + Hs- s (7.50)
Here, HZeeman represents the interaction of nuclear spin with the external magnetic field and is given by
H g N βN I z (A ) H σ ) g N βN I z (B ) H ( − σ B ) (7.51)
where Iz(A) and Iz(B) are the nuclear spin quantum numbers of nuclei A and B in the spin system AB in the direc-
tion of the applied magnetic field, i.e. Z. The Hamiltonian Hs-s represents the interaction between nuclear spins IA
and IB and is represented by
Hs-s = JAB I(A) I(B) (7.52)
The interaction between the two magnetic dipoles is represented by the Hamiltonian
⎡ I ( A )I ( B ) 3 (I ( A )r )(I ( B )r ) ⎤
Hd-d = g N2 βN2 ⎢ − ⎥ (7.53)
⎣ r3 r5 ⎦
where the terms and symbols have their usual meanings. In case of liquids, the d-d interactions averages out to
zero, i.e. Hd-d = 0.
Thus, Eq. (7.50) can be written as
N βN I z (A ) H σ ) g N βN I z (B ) H σ B ) J A B I (A ) I (B ) (7.54)
In frequency units, Eq. (7.54) becomes
H v( − A )I z ( A ) − v ( B )I z ( B ) J A B I ( A ) I ( B ) (7.55)
where I (A ) I ( ) = I x (A ) I x (B ) + I y (A ) I y (B ) + I z (A ) I z (B ) (7.56)
The energy of various magnetic levels due to s-s interaction can be worked out from the Schrodinger wave
equation
Hψ Eψ (7.57)
where y is the exact wave function and can be written as a linear combination of the basic spin function j, i.e.
ψ ∑C ϕ
i
i (7.58)
1
For a two-nuclei spin system having I(A) = I(B) = ,the four possible combinations of spins are
2
ϕ α (A )α (B ) ϕ 2 = α ( ) β (B )⎫⎪
⎬ (7.59)
ϕ β (A )α (B ) ϕ 4 = β ( ) β (B ) ⎪⎭
If Ψ is normalised,
E ∫ ψ H ψ dτ (7.60)
E is determined by solving the determinant (7.62) with the aid of raising and lowering ladder operators I + and
−
I respectively given by
I + = Ix + iIy
I − = Ix− iIy (7.61)
The determinant is
H 11 E H 12 H 13 H 14
H 21 H 22 − E H 23 H 24
= 0. (7.62)
H 31 H 32 H 33 E H 34
H 41 H 42 H 43 H 44 − E
The properties of ladder operators are

I + a = 0, I + b = a.
I − a = b, I − b = 0. (7.63)
Consequently,
I ( A + ) I ( B − ) = {I x ( A ) I x ( B ) + I y ( A ) I y ( B )} + i {I y ( A ) I x ( B ) I x ( A ) I y ( B )} (7.64)
I ( A − ) I ( B + ) = {I x ( A ) I x ( B ) + I y ( A ) I y ( B )} − i {I y ( A ) I x ( B ) I x ( A ) I y ( B )} (7.65)
and
1
Ix (A) Ix (B) + I y(A)I y (B) = [I(A +)I(B-) + I(A-)I(B +)] (7.66)
2
Combining Eqs (7.55), (7.56) and (7.66), we get
H v (1 − σ A ) I z ( A ) − v (1 σ B ) I z ( B ) (7.67)
+ J ⎡I (A ) I (B ) 1
{I (A + ) I (B ) + I (A − ) I (B + )}⎤⎥
AB z z
⎣ 2 ⎦
If ji and jj have different values of MI then
∫ ϕi ϕ j τ = ∫ ϕ j ϕi =0 (7.68)
The various Hij terms of determinant (7.62) are evaluated as
v v J
H 11 = − x A − x B + A B
2 2 4
v v J
H 22 = + x A − x B − A B
2 2 4
v v J
H 33 = − x A + x B − A B
2 2 4
v v J
H 44 = x A + x B + AB
2 2 4
JAB
H 23 H 32 =
2
other terms, H12, H21, H13, H31, H14, H41, H24, H42, H34, and H43 are all zero.
Here, 1 − σ A = A d 1 − σ B = x B
Thus, the determinant (7.62) becomes
v v J
− x A − x B + AB − E 0
2 2 4
v v J
0 x A − x B − AB − E
2 2 4
JAB
0
2
0 0
0 0
JAB
0
2
−v v J =0
x A + x B − AB − E 0 (7.69)
2 2 4
v v J
0 x A + x B + AB − E
2 2 4
Solution of determinant (7.69) gives

⎛ σ σ ⎞ J
E1 v −1 + A + B ⎟ + A B (7.70)
⎝ 2 2 ⎠ 4
⎛ σ σ ⎞ J
E4 v 1 − A − B ⎟ + AB (7.71)
⎝ 2 2 ⎠ 4
E2 and E3 are determined from the solution of the determinant (7.72).
v v J JAB
x A − x B − AB − E
2 2 4 2
=0 (7.72)
JAB v v J
− x A + x B − AB − E
2 2 2 4
JAB 1 2 2
E2 = − − ν δ A B + J A2 B (7.73)
4 2
JAB 1 2 2
E3 = − + ν δ A B + J A2 B (7.74)
4 2
where, δAB B A σA − σB
x x2
Using the expansion, (1 x )
1/ 2
= 1+ + + ..., Eqs (7.73) and (7.74) may be written as
2 4
JAB 1 ⎡ J2 ⎤
E2 = − − v δ A B ⎢1 + 2A B2 + ...⎥ (7.75)
4 2 ⎣ 2v δ A B ⎦
JAB 1 ⎡ J2 ⎤
and E3 = − + v δ A B ⎢1 + 2A B2 + ...⎥ (7.76)
4 2 ⎣ 2v δ A B ⎦
The negative and positive signs in the energy expressions correspond to attraction and repulsion respectively.
To evaluate the perturbation energy due to s-s interactions in the various levels, we proceed as follows
(i) For a bare nuclei, σ = 0, J A B = 0
∴ E1 v
E4 v (7.77)
E2 = 0
E3 = 0.
(ii) In case there is no s-s interaction, the coupling constant J A B becomes zero. This is called zero-order interac-
tion. Thus, the energy of the magnetic levels becomes
⎛ σ σ ⎞
E1 v −1 + A + B ⎟
⎝ 2 2 ⎠
⎛ σ σ ⎞
E4 v 1− A − B ⎟ (7.78)
⎝ 2 2 ⎠
−v δ A B ⎛σ σ ⎞
E2 = = −v ⎜ A − B ⎟
2 ⎝ 2 2 ⎠
v δAB ⎛σ σ ⎞
E3 = =v ⎜ A − B⎟
2 ⎝ 2 2 ⎠
Thus, the perturbation with respect to bare nuclei, when J A B = 0 is given by
1
δ 1 (σ A + σ B )
2
1
δ 4 (σ A + σ B )
2
1
δ 2
2
(σ A − σ B ) (7.79)
1
δ 3 (σ A − σ B )
2
(iii) When J A B v δ A B , there is weak interaction between the nuclei A and B and the system is known as AX instead
of AB. The spectra corresponding to this case are called first-order spectra. For a rigorous first-order spectra,
J A B /vv δ A B 0.05. When the nuclei are strongly coupled, J A B v δ A B and the spin system is said to be AB.
The energies of the magnetic levels for an AX spin system are given by
⎛ σ σ ⎞ J
E1 v −1 + A + B ⎟ + A B
⎝ 2 2 ⎠ 4
⎛ σ σ ⎞ J
E4 v +1 − A − B ⎟ + A B (7.80)
⎝ 2 2 ⎠ 4
JAB 1 J 1
E2 = − − v δ A B = − A B − v (σ A − σ B )
4 2 4 2
JAB 1 J 1
E3 = − + v δ A B = − A B + v (σ A − σ B )
4 2 4 2
The perturbations relative to bare nucleus in this case are
⎛ σ A σ B ⎞ JAB
δ E1 v + ⎟+
⎝ 2 2 ⎠ 4
⎛σ σ ⎞ J
δ E4 v ⎜ A + B ⎟ + AB (7.81)
⎝ 2 2 ⎠ 4
JAB 1
δ 2 δAB
4 2
JAB 1
δ 3 δAB
4 2
(iv) When J A B v δ A B or J A B v (σ A σ B ) , the s-s interaction is of second order and the energies of magnetic
levels are
⎛ σ σ ⎞ J
E1 v −1 + A + B ⎟ + A B
⎝ 2 2 ⎠ 4
⎛ σ σ ⎞ J
E4 v 1 − A − B ⎟ + AB (7.82)
⎝ 2 2 ⎠ 4
JAB 1 J2
E2 = − − v (σ A − σ B ) − A B
4 2 4v δ A B
JAB 1 J2
E3 = − + v (σ A − σ B ) + A B
4 2 4v δ A B
The perturbations with respect to bare nucleus/first-order interactions are
σA + σB J
δ 1 ν[ ] + AB 0
2 4
σA + σB J
δ 4 ν[ ] + AB 0 (7.83)
2 4
JAB 1 J A2 B J A2 B
δ 2 ν [σ A σB ] − −
4 2 4νδ A B 4ν
νδ A B
JAB 1 J2 J A2 B
δ 3 ν [σ A − σ B ] + A B
4 2 4νδ A B 4ν
νδ A B
The perturbations with respect to bare nucleus in the zeroth-, first- and second-order interactions/spectra are
shown in Fig. 7.25(a). In the second-order perturbation
case, the energy states originally corresponding to ab and
(a)
ba are mixed and are represented by linear combinations
Bare Nuclei 0th order 2nd order
of the type (αβ βα )). αα −(1/2)ν(σA + σB )
Ist order
4
JAB /4
(a) Selection Rule The transition between the magnetic
levels n and m is induced by applying radio frequency (1/2)ν(σA − σB )
−JAB /4 3
field (Hrf) to the sample placed in the external magnetic J 2AB /4νδAB
αβ,βα
field of strength H. The radio frequency field is adjusted −JAB /4
−(1/2)ν(σA − σB ) −J 2AB /4νδAB
to oscillate in the direction (X ) which is perpendicular 2
to the direction of the applied field H(Z). The interaction 1
JAB /4
between the field, Hrf and nuclear magnetic moment is
ββ (1/2)ν(σA + σB )
given by JAB < νδ JAB ∼ νδ
σ=0,JAB JAB = 0
Fig. 7.25 (a) The zeroth-, first-, and second-order perturbation
of an AB system. Transitions are: = 1→ 2, 1→3; 2→ 4; 3→4.
H ′ = μ x H rf (7.84)
(b)
AX(JAX = 0) AX(JAX > 0) A2(JAA > 0) AB(JAB > 0)
Fig. 7.25 (b) The four energy levels of a two-spin system for various relative values of
ν (σ σ ) d Jν σA σ B >> J for AX system; ν σ A − σ B = 0 for A2 and ν σ A − σ B ≈ J for AB.

where, μ x is the component of nuclear magnetic moment μI in the X-direction. The transition probability is pro-
portional to
2
⎡ ψ ( μ x H rff )ψ dτ ⎤
⎣∫ ⎦ (7.85)
Since, μ I γ (7.86)
Therefore, transition probability is proportional to simply
(ψ I ψ m dτ )
2
(7.87)
+ −
I I
Taking I x = , the solution of the integral results into the selection rule
2
Δ I = ±1 (7.88)
(b) Transition Between Allowed Energy Levels (Spectrum of AB System) According to the selection rule
Δ = ±1 , the permissible transitions (in frequency units) are
I
Eqs. (7.75–7.70)
JAB ⎛ σA σB ⎞ 1 ⎛ J2 ⎞
v 12 E 2 − E1 = +v 1 − v δ A B 1 + 2A B2 + ...⎟ (7.89)
2 ⎝ 2 2 ⎠ 2 ⎝ 2v δ A B ⎠
Eqs. (7.76–7.70)
JAB ⎛ σA σB ⎞ 1 ⎛ J2 ⎞
v 13 E3 − E1 = − +v 1 + v δ A B 1 + 2A B2 + ...⎟ (7.90)
2 ⎝ 2 2 ⎠ 2 ⎝ 2v δ A B ⎠
Eqs. (7.71–7.75)
JAB ⎛ σA σB ⎞ 1 ⎛ J2 ⎞
v 24 E4 − E2 = +v 1 + v δ A B 1 + 2A B2 + ...⎟ (7.91)
2 ⎝ 2 2 ⎠ 2 ⎝ 2v δ A B ⎠
Eqs. (7.71–7.76)
JAB ⎛ σA σB ⎞ 1 ⎛ J2 ⎞
v 34 E 4 − E3 = +v 1 − v δ A B 1 + 2A B2 + ...⎟ (7.92)
2 ⎝ 2 2⎠ 2 ⎝ 2v δ A B ⎠
Now, when
(i) JAB = 0
⎛ σ σ ⎞ 1
v 12 v 1 − A − B ⎟ − v δAB
⎝ 2 2 ⎠ 2
⎛ σ σ ⎞ 1
v 13 v 1 − A − B ⎟ + v δAB (7.93)
⎝ 2 2 ⎠ 2
⎛ σ σ ⎞ 1
v 24 v 1 − A − B ⎟ + v δAB
⎝ 2 2 ⎠ 2
⎛ σ σ ⎞ 1
v 34 v 1 − A − B ⎟ − v δAB
⎝ 2 2 ⎠ 2
Thus, the spectrum of AB spin system will consist of two lines positioned at v12 = v34 and v13 = v24 respectively
and is represented in Fig. 7.26(a).
(ii) If JAB < v δ A B , the system is weakly coupled and is called AX instead of AB system, e.g. ethanol at 40 MHz.
The frequencies of the lines under this condition are
JAB ⎛ σA σB ⎞ 1
v 12 = − +v 1 v δAB
2 ⎝ 2 2 ⎠ 2
v 13 = − +v 1 + v δAB (7.94)
2 ⎝ 2 2 ⎠ 2
v 24 = +v 1 + v δAB
2 ⎝ 2 2 ⎠ 2
v 34 = +v 1 v δAB
2 ⎝ 2 2 ⎠ 2
Thus, the spectrum in this case consists of two sets of lines, i.e. v12, v34 and v13, v24 and each separated by JAB
and is represented in Fig. 7.26(b).
(iii) If JAB >v δ A B , the spin-spin system is said to be strongly coupled. This is a true case of AB spin system,
e.g. ethanol at 60 MHz. Under this condition,
JAB ⎛ σA σB ⎞ 1 J2
v 12 = − +v 1 v (σ A − σ B ) − A B
2 ⎝ 2 2 ⎠ 2 4v δ A B
v 13 = − +v 1 + v (σ A − σ B ) + A B (7.95)
2 ⎝ 2 2 ⎠ 2 4v δ A B
v 24 = +v 1 + v (σ A − σ B ) + A B
2 ⎝ 2 2 ⎠ 2 4v δ A B
v 34 = +v 1 v (σ A − σ B ) − A B .
2 ⎝ 2 2⎠ 2 4v δ A B
The spectrum in this case also consists of two sets of lines, i.e. v12, v34 and v13, v24. The intensity ratio between
the outer to inner lines will always be < 1. The schematic spectrum for this case is shown in Fig. 7.26(c).
(iv) If ν δ A B = 0 then
JAB ⎛ σ σ ⎞
v 12 = − + v 1− A − B ⎟
2 ⎝ 2 2 ⎠
JAB ⎛ σ σ ⎞
v 13 = − + v 1− A − B ⎟ (7.96)
2 ⎝ 2 2 ⎠
JAB ⎛ σ σ ⎞
v 24 = + v ⎜1 − A − B ⎟
2 ⎝ 2 2 ⎠
JAB ⎛ σ σ ⎞
v 34 = + v ⎜1 − A − B ⎟
2 ⎝ 2 2 ⎠
JAB ⎛ σ σ ⎞
The spectrum in such a case will consist of only one line centred at + v 1 − A − B ⎟ and is shown in
Fig. 7.26(d), e.g. benzene exhibits only a one-line spectrum. 2 ⎝ 2 2 ⎠
The stick spectra of AB spin system reveal that it is better to record the spectra at the highest field possible
so that most spectra have the first-order appearance due to increase in chemical shift. The analysis of an NMR
spectrum is carried out by making tentative guesses of JAB s and v δ A B s and then drawing a theoretical spectra
comparable to the observed one.
The selection rules from Section 7.28 give allowed transitions 1→2, 1→3, 2→4 and 3→4 for non-equivalent
protons are shown in Fig. 7.25(b). The selection rules differ for non-equivalent and equivalent protons. The rules
for the latter are 1→3, 3→4 [Fig. 7.25(b)].
7.29 INTENSITY OF NMR LINES IN AB SPIN SYSTEM

The intensity and position of lines in an NMR spectrum depend upon the magnitude of interaction but the area
under the observed spectrum remains the same in all the cases. The general structure of NMR spectrum which
arises due to s-s interaction in the AB system is shown in Fig. 7.26.
n(σA – σB)
JAB JAB
n12,n34 n13,n24 n12 n34 n13 n24

(a) (b)
a
d2 d3
d1 b d4
n12 n34 n13 n24

(c) (d)
Fig. 7.26 Theoretical stick spectra of an AB system: (a) JAB = 0, (b) J AB < n dAB (c) J AB > n dAB (d) n dAB = 0.
The relative intensities of the outer to inner lines in an AB or AX system and the difference in chemical shifts
( Δδ ) = v A v B , are calculated from the following expressions:
⎛ Line1 ⎞ Q − J
Relative intensity ⎜ = (7.97)
⎝ Line 2 ⎟⎠ Q + J
1
Δδ = ( 2
− 2
)2 (7.98)
vA +vB
and v = (7.99)
2
Here, JAB d1 − d 2 o d 3 d 4 (in Hz) (7.100)
d1 d 2 d d4
vA = vB = 3 (7.101)
2 2
Q d1 − d 3 or d 2 d 4 (in Hz) (7.102)
Δδ = ( − )( − ) (7.103)
(d d ) − (d − )
JAB = (7.104)
2
Or the relative intensities of four resonance frequencies for a two-spin system are also expressed as
Frequency Relative Intensity*
(I r 1) (I r + 1)2
2
v 1→ 2
v 1→3 1
(I r 1) (I r + 1)2
2
v 2→ 4
v3 4 1
1/ 2
⎡ ( Δ 2 + J 2 )1/ 2 + Δ ⎤
*where Ir = ⎢ 2 ⎥
⎣ (Δ + J ) − Δ ⎦
2 1/ 2
and Δ =v( A − B )
Problem 7.14: In the PMR spectrum (60 MHz) of cis-b-phenylmercaptostyrene, dAB = 6.48 ppm for two
ethylenic hydrogens, JAB = 10.7 Hz and ΔdAB = 0.095 ppm. Compute the intensity ratios of outer to inner lines in
the spectrum.
Solution JAB = 10.7 Hz. Using Eq. (7.98), we can compute the value of Q, i.e. 60 × 0.095 = (Q2 –10.7 × 10.7 ) 1/2.
Q = 12.123. From Eq. (7.97), we get
I1 I 12.123 − 10.7 1.423

or 4 = = = 0.062
I2 I 3 12.123 + 10.7 22.823
The observed intensity ratio is 0.056.
Problem 7.15: The NMR spectrum of cis-b-ethoxystyrene, when recorded on a 60 MHz spectrometer gives
ethylenic protons at dA = 5.18 ppm for and dX = 6.06 ppm JAX = 7.2 Hz. Calculate the intensity ratios of the outer
to inner lines in the spectrum.
Solution In AX system, the lines become more equal in intensity as compared to AB system. Here, Δd = (6.06 –
5.18) = 0.98 ppm = 0.98 × 60 Hz = 58.8 Hz.
58.8 = (Q2 −7.2 × 7.2)1/2
Q = 59.23 Hz
I1 I 59.23 − 7.2 52.03
Thus, or 4 = = = 0 78
I2 I 3 59.23 + 7.2 66.43
The observed intensity ratio is 0.77.
Problem 7.16: The PMR spectrum of 2, 3-dibromothiophene on a 56.4 MHz spectrometer (15 per cent by
weight in CDCl3) exhibits four lines at 410.39, 404.65, 390.64 and 384.87 Hz down-field from TMS. Calculate
the intensity ratio between the outer and inner lines.
Solution Using Eq. (7.103), we get
Δδ = ( .39 − 384. )( .65 − 390. )
= 18.91 Hz
and from Eq. (7.104),
25.52 − 14.01
JAB = = 5.70 Hz
2
According to Eq. (7.98),
18.91 = (Q2 − 5.70 × 5.70)1/2
Hence, Q = 19.75 Hz
From Q and JAB , we get
I1 I 19.75 − 5.70 14.05
or 4 = = = 0.552
I2 I 3 19.75 + 5.70 25.45
which compares well with the experimental value of 0.606.
Problem 7.17: The PMR spectrum of trans-1-bromo-2-chloroethene on 56.4 MHz instrument (in hexafluo-
robenzene) gives d1 − d4 = 25.80 Hz, d2 – d3 = 0.89 Hz and JAB = 12.46 Hz. Calculate the intensity ratio between
the inner and outer lines.
Solution Using Eq. (7.103), we get
Δδ = 25.8
8 × 89 = 4.79 Hz

Q = ( .79) 2 + (12
2.. ) 2 = 13.35
Thus, the intensity ratio between the inner and outer lines
13.35 + 12.46 25.81
= = = 29
13.35 − 12.46 0 89
The observed value of intensity ratio is 20. This discrepancy is due to the fact that the two inner strong lines are
easily saturated because of their large transition probability. The outer lines are very weak and might even escape
detection. This is the collapse of AB pattern into an A2 pattern.
7.30 SECOND-ORDER EFFECTS

Perfect first-order spectra are seldom observed in homonuclear spin systems. The deviation (lAB) from the near
first-order spectra is given by
JAB
λA B = (7.105)
(v A v B )
The spacing between multiplets depends linearly on the perturbation parameter λ A B . Second-order corrections
become simplest in spectra of spin systems in which every basic group contains only one nucleus of spin 1/2,
4 3
e.g. 2-furan aldehyde 5 2 CHO . The spin coupling between two of the coupled nuclei, A and K, produces a
shift of the whole A multiplet pattern by J A2 K / 2(vA − vB ) and of the K multiplet pattern by J A K / 2(v K − v A ). The
2
second-order shift tends to increase the separation between the A and K multiplets beyond the value v A v K .
In a first-order spectrum, the multiplet pattern of any basic multiplet displays a perfect mirror image symmetry
relative to its centre frequency. On the other hand, when two basic groups are sufficiently strongly coupled to
make λ A B non-negligible, this mirror image symmetry is lost in such a way that the intensities of those parts of
the two bands which are nearest to each other increases at the expense of the intensities of the outer parts of the
multiplets. This is called roofing effect. The roofing effect is discernible in the H3 and H4 bands of 2-furanaldehyde
and is of great diagnostic value in the analysis of NMR spectra when certain resonances are hidden under broad
absorption bands. In a spin system containing several magnetically equivalent nuclei in a basic group, the second-
order effects result in further line splitting and the line spacing within a basic multiplet will no longer depend
linearly on the coupling constant.
Coupling constants are smaller than 20 Hz and chemical shifts are higher (of the order of several thousand
Hz) in a second-order spectra. Splitting can be explained by the rules of first-order spectrum. However, when
/ v .δ > 0.1 – 0.15, the rules of first-order spectrum hold no more and when v δ → J roofing effect appears in the
spectrum. These effects manifest themselves in the spectrum of 2-furanaldehyde, 5-methyl-4-thiolen-2-one and
ethylbromide, etc. The spectrum of ethylbromide on a 60 MHz instrument exhibits a quartet due to CH2 group at
3.46 ppm and a triplet due to CH3 group at 1.67 ppm. The coupling constant is 7.35 Hz. The perturbation parameter
( 7 35)
2
7 35
λA B = = 0.068 and the second-order shift is = 0.25. Thus, the second-order shifts are significant.
1 79 × 60 214.8
Consequently, the first-order estimates of the chemical shifts should have an analytical error of the order of 0.25 Hz.
Moreover, the basic groups of ethyl bromide contain groups of magnetically equivalent nuclei due to which addi-
tional splittings of the order of 0.25 Hz will arise in the multiplet lines through second-order effects. Asymmetry
and roofing effects are also observed in the two multiplets as shown in Fig. 7.27.
Spectrum of 2-furanaldehyde on 60 MHz instrument exhibits δ H4 (four doublets) = 398.5 Hz, δ H3 (two dou-
blets) = 438.9 Hz. δ H5 (pentet) = 467.4 and δ CHO (two doublets) = 581.6. J34 = 3.70 Hz, J45 = 1.70 Hz, J35 = 0.75
Hz, J CHO 4 = 0.30 Hz, J CHO−3 = vanishingly small and J CHO−5 = 0.65 Hz. The maximum value of perturbation
parameter in this case,
3 70 3 70 δ (−CH3) = 1.67 ppm

λ34 = = = 0 09
(438.9 − 398.5) 40.4 H H
H C C H
This value is large enough to make second-order effects appear vis- H Br
( 3 70 )
2
ible in the spectrum. The second-order shift is = 0 17 Hz and

80.8 δ (−CH2) = 3.46 ppm
is almost negligible.
The PMR spectrum of 5-methyl-4-thiolen-2-one on 40 MHz instru-
ment (in cyclohexane) exhibits a simple first-order pattern and even
60 MHz PMR spectrum exhibits no second-order effects. dCH3(three
doublets, 1 : 2 : 1) = 2.07 ppm, δ CH 2 (pentet, I : 4 : 6 : 4 :1 ) = 3.37
ppm and δ CH (triplet of quartets; intensity ratio of three quartets =
1: 2 :1; the lines within each quartet has intensity ratio = 1: 3 : 3 :1) =
5.54 ppm. J CH3 CH2 2 5 z ; J CH3 − CH = 1 5 Hz, and J CH2 CH2 = 2 5 Fig. 7.27 NMR spectrum at 60 MHz of neat
ethylbromide.
Hz. The largest perturbation parameter
25 25
λ= = = 0.048
(3.37 − 2.07) 40 52
and it causes second-order shift =
(2 5)2 = 0 06 Hz and is negligible.
104
7.31 EFFECT OF CHEMICAL ENVIRONMENT ON SHAPE OF NMR SIGNAL

The spectrum of a nucleus which exists in more than one chemical environments depends upon its lifetime in
these environments. The lifetime of nucleus on sites I and II, i.e. t1 and t2, are long as compared to the NMR
transition time. The populations on sites I and II are PI and PII respectively such that
τ1
PI = (7.106)
τ1 + τ 2
τ2
and PII = (7.107)
τ1 + τ 2
Thus, the probability of nucleus jumping from site I to site II will be 1/t1 while from II to I will be 1/t2. The
jumping of the nucleus at the same site has no effect on magnetisation. Let us suppose that the population and
lifetime of the nucleus in the two environments are the same, i.e. t1 = t2 = t and PI = PII. In such a case, the lifetime
t can be determined from the expression
1/ 2
(δ I δ II )obs ⎡ ⎤
1
= ⎢1 − ⎥ (7.108)
(δ I δ II ) ⎢ 2π τ δ
⎣
2 2
I ( δ II )
2
⎥
⎦
( )
Here δ I0 δ I0I is the difference between the chemical shifts of the two absorptions for large values of t [Fig.
7.28(a)]. (δ I δ II )obs is the chemical shift difference between the coalescing peaks [Fig.7.28(b)–(d)], and in this
case the mean lifetime t = t1 /2 or t2 /2. When the peaks have just submerged into a single line [Fig. 7.28 (e)], then
(dI – dII)obs becomes zero and the expression for mean lifetime is
1
t= (7.109)
(
2 π δ I0 − δ I0I )
This is the case of intermediate rates of exchange between two sites. The transverse relaxation times are large,
i.e. 1/T2 (I) = 1/T2 (II) ≈ 0 and PI = PII = 1/2.
1
(i) t >> . In this condition, two lines are observed, provided the absorptions at δ I0 δ I0I
( )
2 π δ I − δ II
0 0
appear as singlet. This is the case of slow rates of exchange between the two different sites. By NMR
T<TC T<TC
δ°1 δ°2 δ1 δ2
(a) (c)
T = Tc
T<TC Δν
δ1 δ2
(b) (d)
T > Tc
Δν
(e)
Fig. 7.28 Changes in the shape of the NMR resonance signal for an increasing exchange rate between two positions, I and II, with equal
populations. Tc is the coalescence temperature.
criterion, under these conditions, the molecules are said to have rigid structures, e.g. cyclotriborazane
(MeNHBH2)3 In this case the lifetime at site I, i.e. t1 is related to the transverse relaxation times as
1 1 1
= + (7.110)
T2 T2 τ1
1
The effective time T 21 is determined from resonance line width Δv, i.e.
1
T 21 = (7.111)
πΔv
T2 is the transverse relaxation time in the absence of exchange. The rate constant k for exchange = 1/t1
which can be obtained from the increased broadening caused by exchange. Similarly, we can determine
t2. The drawback for determining k from line broadening is that t1 must be accurately known.
1
(ii) τ < . Under such a condition, only a single time-averaged line is observed. This is the case
(
2π δ − δ )
of rapid rates of exchange between two sites, i.e. t1 and t2 are small. By NMR criterion, under these condi-
tions the molecules are said to have nonrigid structures, e.g. inversion of NH3 and cyclohexane. In such
molecules often two or more (but not always) nuclear configurations are possible. Single 13C resonance line
is maintained up to 210 K in Fe (13CO)5 while a single 19F doublet (JPF = 916 Hz) is maintained down to 113
K in P 19F 5 Si 19F 5 MeP 19F 4 , etc. Actually in both the cases two peaks are expected in the ratio 3:2.
The position of the observed chemical shift d depends upon the relative populations in the two environ-
ments, i.e.
d = P1 d1 + P2d2 (7.112)
In general, we can say that any change in the system which perturbs the relative populations PI and PII will
affect the observed chemical shift, d. The factors which may alter the populations are temperature, concentration
and nature of donor or acceptor in the case of electron donor-acceptor or H-complexes.
In the limit of rapid exchange, the line width Δν of a single resonance peak at d is
1 1 P P
Δν = such that = I + II (7.113)
πT 2 T 2 T 2 (I ) T 2 (II )
For weak interactions, the exchange rates between different environments at normal experimental temperatures
are comparable to rates of very fast reactions. However, when interactions are of covalent type, the rate of chemical
exchange is of the same order or less than the exchange rate between the magnetic states of the nucleus, for which
line broadening or separate lines will be observed, e.g. a single resonance line due to proton is observed under all
experimental conditions in the H2O−H2O2 system since the exchange rate of proton between the two situations is
large as compared to its lifetime in the two situations. In aliphatic alcohols, the collapse of spin multiplets also
occurs through proton exchange. The spin coupling between the hydroxyl proton and a vicinal CH proton is of
the order of 7 Hz. If the residence time t of hydroxyl proton in any particular molecule is greater than 0.2 second
(2p Jt >10) the s-s splittings will still be resolved in the spectrum but the multiplet lines are broadened. However,
in the presence of an acid or a base, the proton exchange rate increases due to which residence time, t of the
hydroxyl protons in a given molecule reduces to less than 0.01 second. In this situation, the s-s splittings caused
by spin coupling to hydroxyl protons are no longer observable. The collapse of the spin multiplets occurs both in
the OH and CH absorption bands of the molecule, e.g. C2H5OH, the NMR line due to –OH group will appear as a
singlet while due to CH2 group as a quartet. In the case of N, N-dimethylformamide and N N-dimethylacetamide,
three broad NMR signals of interchanging (due to rotation around C-N bond) non-coupled nuclei are observed.
The collapse of spin multiplets in this case is by quadrupole relaxation. Splitting due to s-s interaction between a
proton (or some other nucleus with spin I = 1/2) and nuclei with spin I >1/2 is rendered unobservable by the rapid
quadrupole relaxation of such nuclei. The interactions between the nonspherical charge distribution of nuclei
with I >1/2 and randomly varying in homogeneous molecular electric fields induces thermal transitions between
the magnetic spin states of such nuclei, i.e. spin I >1/2. When the rates of these thermally induced transitions are
large as compared to the spin coupling constant, the spin-coupled proton only senses the vanishing average value
of this interaction over the 2I + 1 spin states of the I >1/2 nucleus. (In case of nitrogen I = 1, so 2I + 1 = 3. Hence,
the three spin states are +1, 0, –1.) At intermediate relaxation times of the I >1/2 nucleus, a spin-coupled proton
experiences a line broadening only. The quadrupole relaxation of the nuclei of halogen atoms Br, C1 and I (except
F, which has spin I = 1/2) are fast enough to eliminate all evidence of proton-halogen spin coupling effects in
proton resonance spectra of these halogenated compounds.
The extended view of the effect of exchange rate between two sites I and II, with equal populations, on the
NMR signal is shown in Fig. 7.28(a) to (e).
7.31.1 Effect of Temperature

Let us now discuss the effect of temperature on the NMR spectra of cyclo-
hexane in relation to their conformations. Cyclohexane and its derivatives
can assume two types of conformations: the chair and boat forms. The
first is more stable (it has a lower energy) and is generally regarded as the 293 K
only form in which cyclohexane actually exists. Monosubstituted chair
cyclohexanes may have two conformations: one with an axial and the
other with an equatorial substituents; the latter has a low energy. The con-
formational equilibrium can be obtained by spectroscopic methods such 303 K
as infrared spectra and combination scattering, in particular by NMR
He Hα
method.
The PMR spectrum of cyclohexane is shown in Fig. (a). At room
temperature, the spectrum cyclohexane contains only one sharp singlet.
This is associated with the very rapid flipping of one chair conformation of
the ring to the other chair from so that the axial and equatorial protons are 173 K
constantly exchanged at lower temperatures (a solution of cyclohexane in Fig. (a) PMR spectra of cyclohexane
carbondisulphide) beginning with 243 K, the signal is broadened and at 173 (operatingfrequence:100MHz).
K, two distinct signals and their fine structure are observed, the fine struc-
ture being due to s-s interaction. The energy barriers for such
conformational changes can be calculated from these studies.
Let us now examine 19F−NMR spectrum of 1,1-difluorocyclo- 303 K
hexane at different temperatures indicated in Fig. (b). At 303 K, F
a single peak appears, which means that both fluorine atoms, the 224 K H F
axial and the equatorial atoms are practically equivalent because H
of the rapid introversion of the two conformations (so that one 213 K H
fluorine atom becomes alternately axial and equatorial). At low H
temperatures the energy barrier between the two conformations
becomes large due to which the lifetime of each conformation 173 K
becomes longer. As the temperature decreases the signal widens Fig. (b) 19F-spectra of 1,1-difluorocyclohexane.
and splits into two maxima (peaks) and finally, at 173 K there appear two signals corresponding to the two dif-
ferent conformation of the molecules. The peaks of the axial fluorine atoms are somewhat widened on the right
due to s-s interaction with the ring protons. From the spectral data we can calculate the lifetime t of an individual
conformer at each temperature by the formula
1/ 2
δτ ⎡ 1 ⎤
= ⎢1 − 2 2 ⎥ (7.108a)
δ ∞ ⎢⎣ 2π τ (δ ∞ ) ⎥⎦
2
where dt is the chemical shift between the signals of fluorine atoms at a given temperature and δ ∞ is the chemical
shift at 173 K (or at lower temperature) when the peaks are distinctly separated.
Thus the lifetime and rate constant (k) of a conformation at different temperatures is as follows:
Temperature (K) t (s) k (s−1)
303 1.0 × 10−6 1×10 + 6
224 6.5 × 10−4 0.15 × 10 + 4

213 1.2 × 10−3 0.83 × 10 + 3
173 2.0 × 10−2 0.50 × 10 + 2
The energy barrier between the conformations can also be determined from this data.
7.32 NMR SPECTROMETERS

Two types of NMR instruments available in the market are wide-line and high-resolution spectrometers. Mag-
nets in wide-line spectrometers have of a few tenths of tesla whereas the strengths of magnets lies in the range
1.4–14 tesla which corresponds to proton frequency of 60–1000 MHz in high-resolution instruments.
CWNMR instruments are equipped with permanent or conventional electromagnets. These days conventional
magnets are seldom used. Permanent magnets with field strengths of 0.7, 1.4 and 2.1 tesla corresponding to
proton frequency of 30, 60 and 90 MHz are used in CWNMR spectrometers. In FTNMR instruments super-
conducting solenoids are the source of magnetic field. Computer is an integral part of FTNMR spectrometers.
The role of a computer is to record spectrum in the time domain, i.e. the radiation emitted by the excited
nuclei in the sample are collected as a function of time in the computer, perform FT on the FID signal stored,
to obtain the frequency domain signal or spectrum. High resolution and sensitivity, both of which depend
upon the strength and quality of magnets, are the two main characteristics of FTNMR instruments. Due to
this, these instruments have made possible, the analysis of analytes containing (i) 13C nuclei, (ii) protons in
microgram quantity, and (iii) 19F, 31P and other such nuclei. These days, all the modern high-resolution NMR
spectrometers are equipped with superconducting magnets. These magnets can attain magnetic field strength
of 21 tesla, corresponding to a proton frequency of 900 MHz. Superconducting magnet system is obtained
by winding superconducting niobium/tin or niobium/titanium wire on a solenoid. The wound solenoid is then
bathed in liquid helium at 4 K. The helium dewar is kept in an outer liquid-nitrogen dewar. Such magnet
systems must be filled with liquid nitrogen every 10 days and with liquid helium every 80–130 days. These
types of superconducting magnets are highly stable, homogeneous and have high strengths as compared to
conventional electromagnets.
7.32.1 Basic Principle of CWNMR Spectrometer

To obtain good NMR spectra, we should have maximum possible spins, i.e. NMR nuclei in the lower state. This
can be achieved according to Boltzmann’s law, i.e. either by lowering the temperature or increasing the field
strength or both. There are, of course, practical limitations to these two variables. The flowsheet diagram of an
NMR spectrometer is shown in Fig. 7.29. The sample tube which consists of a 5 mm ordinary glass tube containing
0.5 ml of liquid (neat or solution with a trace of reference, i.e. 5 per cent TMS; microtubes for smaller volumes are
also available) is placed between the poles of a magnet which produces a uniform constant field of high strength,
say H. A circularly polarised radio frequency field (Hrf) is applied by passing an AC through a coil wound around
the sample tube. Experimentally, it is arranged so that Hrf

AMP
oscillates in a direction (say X) which is perpendicular to
the direction of the applied magnetic field, H (say Z). The O
field Hrf induces transition between the magnetic levels, say
n and m. The spectra can be obtained both by field-sweep M A
method and frequency-sweep method. Frequency sweep R
method is more difficult but has advantages especially for
some double resonance experiments. Both types of spec- G
trometers are available in the market. When H is varied and
Fig. 7.29 Schematic representation of an NMR instrument—A:
v, i.e. frequency of radio frequency field is kept constant, Ampule with sample, M: Magnet, G: Generator, AMP: Ampli-
then at a certain value of H, the sample absorbs energy and fier, O: Oscillograph, R: Recorder.
current flow in the coil falls off. This variation is ampli-
fied and transmitted to the recorder or oscillograph. The spectra of neat liquids or solutions and sometimes of
gases are recorded. However, in case of solids, the absorption bands are blurred, because in crystals the magnetic
moments of nuclei are affected by the spatially fixed magnetic moments of the neighbouring nuclei. The influence
of neighbouring nuclei is averaged out to zero by the random molecular motions in case of liquids. This type of
effect also becomes operative if the applied field is not homogeneous. Under this condition, resonance will occur
at apparently different magnetic fields, say H ± δH1, H± δH2,.., and the broadening will result due to overlapping
of lines. This effect is minimised by spinning the sample tube with a frequency so that all nuclei senses an average
magnetic field. The spinning frequency should be greater than the rate of fluctuation of the magnetic field. Spin-
ning is done by passing a stream of air to drive the plastic turbine that slips over the sample tube at a rate of 20–50
revolutions per second. Sometimes, the magnetic field is modulated at the spinning frequency, consequently side
bands also called spinning bands appear on each side of the absorption lines. Inhomogeneities in the applied field
are also compensated by the skimming process and by locking the magnetic field. In the former case, controlled
currents are passed through the pairs of wire loops called skim coils (not shown in Fig. 7.29) to produce small
magnetic fields in order to compensate the inhomogeneities in the applied field. Skimming is done each time a
new sample is introduced in the NMR spectrometer. The effect of field fluctuations can be compensated by a field
frequency lock system. NMR absorption signal of a reference nucleus, i.e. deuterium in a solvent, is obtained by
continuously irradiating the nuclei at a constant magnetic field of strength H and its behaviour is monitored along
with that of the sample under study. Changes in intensity of the reference signal are given to a feedback network
whose output is fed into the coils in such a way that drift in the magnetic field (positive or negative) is automati-
cally taken care of or corrected. When stable superconducting magnets are used, the spectra can be obtained in
an unlocked mode for a period of 1–20 minutes. Signal integrators (electronic or digital) are the essential compo-
nents of modern spectrometers. They determine the areas under the absorption signals.
The solvents used in PMR spectroscopy should be such liquids which do not give NMR signals (CCl4, CS2) or
absorb in a different regions, e.g. CDCl3, (CD3)2 SO; AsCl3 and C6D6.
The NMR instruments are characterised by three main parameters: the frequency (40, 60, 100 MHz) which
depends upon the field strength, H; sensitivity; and resolving power. The higher the sensitivity of the instrument,
the more intensive the signals, will be observed with the same concentration of protons in the analyte. Current
instruments allow us to distinguish between two lines which are only 0.2–0.5 cps apart from each other. The resolv-
ing power of the most efficient Instruments operating at a frequency of 100 MHz is 0.2 cps/108 cps = 2 × 10−9. Since
revolving power Rp is defined as the ratio of the maximum distance resolvable to the operating frequency of instru-
ments expressed in cps. The Rp grows with growth in the operating frequency of the instrument. The resolution of
spectra, however, depends upon, the resolving power of the instrument, the velocity of the spectrum scanning and
on the properties of the sample, i.e. on the viscosity of the solution.
Thus, depending upon the purpose of NMR studies, we can obtain either low resolution (wideline) or high-
resolution spectra of the analyte. In addition to the blurring of absorption band of a nucleus by the magnetic
moments of the neighbouring nuclei in crystals, line broadening in solids is also caused by chemical shift anisot-
ropy, i.e. changes in the chemical shift with the orientation of the molecule or part of the molecule with respect
to an applied magnetic field cause line broadening. The chemical shift Δd brought about by magnetic anisotropy
is given by the equation
Δd = Δc (3 cos2q–1)/r3 (7.114)
where q is the angle between the double bond and the applied field, Δc is the difference in magnetic suscepti-
bilities between the parallel and perpendicular orientations of bond and r is the distance between the anisotropic
functional group and the nucleus. For q = 54.7o, Δd = 0. Experimentally, line broadening due to chemical shift
anisotropy is annihilated by Magic Angle Spinning (MAS). 19F in K AsF6

The solid samples are rotated rapidly at a frequency greater (a)
than 2 kHz in a special sample holder that is maintained at an
Static
angle of 54.7o with respect to the external field. Under such
experimental conditions the solid behaves like a liquid being
(b)
rotated in the external magnetic field. The 19F spectrum of
static and rotated (at 5.5 kHz) polycrystalline KAsF6 is shown
in Fig. 7.30. The spectrum of static sample shows the ordinary
Rotated at 5.5 kHz
NMR signal, the width of which is governed by d-d interac-
tions among the spins. On the other hand, magic angle rota- −15 −10 −5 0 5 10 15
tion of the sample gives rise to a quartet structure. The struc-
Fig. 7.30 F spectrum of poly crystalline KAsF6: (a) with-
19
ture is because of the electron coupled interaction between out sample rotation, and (b) with magic angle rotation of the
19 ⎛
1⎞ ⎛ 3⎞
F ⎜ I = ⎟ and 75As ⎜ I = ⎟ nuclei. The coupling constant is sample.
⎝ 2⎠ ⎝ 2⎠
J = 905 Hz.
A good NMR spectrum is obtained in about 5 minutes when 20 mg of sample containing 31P, 19F or 1H nuclei
is dissolved in 0.5 ml of the solvent. 10 mg of sample takes about 20 minutes.
Problem 7.18: If the operating frequencies of NMR instruments are 30, 40 and 60 MHz, what will be their
resolving powers if the maximum distance resolvable is 0.3 cps?
Solution We know that resolving power of
Maximum
u distance resolvable in cps
NMR spectrometer (Rp) =
Operating frequency in cps
p
0.3 cps
Thus, Rp (30 MHz) = = 1 10 −8
30 × 106 cps
Similarly, Rp (40 MHz) = 0.75 × 10−8
And Rp (60 MHz) = 0.5× 10−8
7.32.2 Basic Principle of a Fourier Transform NMR (FTNMR) Spectrometer

FTNMR spectrometer has higher sensitivity, and resolution as compared to conventional NMR instrument. A
conceptual block diagram of one-dimensional FTNMR instrument is presented in Fig. 7.31. The sample in a static
magnetic field undergoes bulk magnetisation, M which is the sum of the magnetisation of the individual spins and
is aligned with the Z-axis, i.e. the direction of the external magnetic field, H. The bulk magnetisation vector M is
stimulated with a single p/2 pulse along X-axis, covering a broad range in the radio frequency region. The pulse
is generated by a transmitter/receiver coil (wound around the sample but not shown in Fig. 7.31) whose axis is
aligned, perpendicular to the Z-axis, i.e. along the X-direction. The pulse rotates the vector M by 90o to an orienta-
tion perpendicular to the static field, i.e. Y-axis. The torque due to applied field after the termination of the pulse
causes the bulk magnetisation to relax back to the Z-axis. The response, i.e. the relaxation of the pulsed system
which is picked up by the same coil which now serves as a receiver coil is measured as a function of time using a
digital computer. A separate receiver coil whose axis is aligned with the Y-direction may also be used. The signal
due to relaxation process is called a Free Induction Decay (FID). The FID signal which is normally collected and
analysed is the difference between the NMR signal and the excitation frequency. The FID is then converted to the
normal frequency domain signal by a Fourier transformation with the aid of a digital computer and the signal is
stored in the memory. With FT technique, it is possible to pulse or scan the entire resonance regions of a given
type of nucleus repeatedly. The FIDS derived from each pulse are summed, thus greatly enhancing the signal-to-
noise ratio, thereby making it possible to record the NMR spectrum of small quantities of sample or even contain-
ing a low abundant nucleus such as 13C, 2H, 15N, 17O.
Sample
Z
Magnet
Y
Magnetisation
Perturbation (Pulse) X
Response
Detection
Data
FT
Spectrum
Storage
Fig. 7.31 Block diagram of a Fourier transform NMR spectrometer.
7.33 SIMPLIFICATION OF NMR SPECTRA

The complexity in NMR spectra of complex molecules arises due to the overlapping of several NMR bands with
one another. It is removed by a number of methods such as shift reagents, superconducting magnets and, double
resonance, i.e. spin decoupling, spin tickling, nuclear overhauser effect, internuclear double resonance, etc., so
that the structural pattern in the spectra becomes easily recognisable and analysable.
7.33.1 Shift Reagents

Certain lanthanide chelates that cause a change in the chemical shift of nuclei when added to solutions of organic
compounds that contain a Lewis basic functional group are known as chemical shift reagents. These reagents are
used to spread out a spectrum when several NMR signals overlap with one another. The change in PMR shifts by
such reagents is of the order of 0–200 ppm.
Chelates of the trivalent ions Tb, Dy, Ho and Tm have greater shifting abilities than those of Eu and Pr, still the
latter are preferred firstly because they cause negligible line broadening and secondly they are complementary, i.e.
a dipivalomethane (DPM) complex of Eu (III); (DPM)3 Eu induces a chemical shift to lower field with a certain
substrate whereas Pr(DPM)3 causes a chemical shift to higher field with the same substrate. This is due to differ-
ent reagent–substrate geometry. The parameter (3 cos2q−1) can have either positive or negative values. Here, q
is the angle between the principal symmetry or crystal field axis and the lanthanide proton direction. Lanthanide
complex of 1, 1, 1, 2, 2, 3, 3-hepta fluoro-7, 7-dimethyl-4, 6-octane dione (FOD), i.e. Eu(FOD)3 is preferred over
DPM chelate, i.e. Eu(DPM)3 because of its lower field shifts in the order Er < Eu < Yb < Tm while higher field
shifts are produced by chelates of Dy, Sm, Nd, Pr, Ho in the order Ho > Pr > Nd >Sm > Dy.
The efficiency of a shift reagent depends upon the following:
(i) Ability of the organic compound (analyte) to form a Lewis-acid-base complex. The order of induced shift is
amines, alcohols > carbonyl compounds > ethers, sulphides and alkenes, alkynes and aromatics.
(ii) Distance between the complexing functional group and lanthanide reagent. It follows the 1/r3 law, i.e. as
the distance increases, the induced chemical shift decreases. The line broadening of proton resonances by
lanthanide ion follows approximately the 1/r6law.
(iii) Nature of the lanthanide metal and the ligands attached to it. Eu(DPM)3 produces large shifts than Eu(FOD)3.
HB HA
(a) H1
H2+H3 O
H2
H1+H4 HB′
A2 B2 HA′ H4 H3
8.0 7.0 6.0 5.0

H2+H3
H1+H4 (b)
A2 B2
13.0 12.5 12.0 10.0 9.0 8.0
(c)
H2+H3
B2 A2 H1+H4
5.0 4.0 1.5 1.0 −6.5 −7.0 −7.5

Fig. 7.32 PMR spectrum of 1,4-dihydronapthalene-1,4-oxide on a 60 MHz instrument: (a) 0.5M in CDCl3, (b) 0.5M in CDCl3, with 0.2 equiv-
alent of Eu(FOD)3 added, (c) 0.5M in CDCl3, with 0.2 equivalent of Pr(FOD)3 added; shifts used as d values in ppm from internal TMS.
(iv) Amount of complex added to the sample. The induced shift increases with the increase in the amount of the
complex added.
(v) Type of functional group and structural features present in organic molecules. The induced shift in amines
[in ppm/mole of Eu(DPM)3/ mole of substrate] lies in the range 30–150 whereas in alcohols it is 20–100 ppm.
Shift reagents are also useful with 13C-NMR spectra, e.g. Eu(DPM)3 is used for distinguishing the cis
and trans isomers of 3-methylcyclopentanol and 1,3-dimethylcyclopentanol. Eu(DPM)3 produces down-field
shifts in PMR while it induces up-field shift of 14N resonance lines, the largest shifts observed in alkylamines
and pyridine are around 1500 ppm, the corresponding Yb complex gives down-field shifts of up to 400 ppm.
Chiral shift-reagents, i.e. b-diketoanion of 3-trifluroacetyl-isobornane (C12H15O2F3) or 3-hepta fluoropropyl-
isobornane (C14H15O2F7) are useful for the analysis of mixtures of enantiomers which have identical spectra. In
the presence of a chiral shift reagent, complexes are formed that are diastereomeric and have non-equivalent
NMR spectra in solutions (in chiral solvents) because the optically active solvent solvates the optically active
solute, e.g. a chiral solvent and chiral shift reagent are used to distinguish the meso d and l stereoisomers of
the pesticide dieldrin.
Figure 7.32 illustrates the value of shift reagents, i.e. Eu(FOD)3 and Pr(FOD)3, in the clarification of the spec-
trum of 1,4-dihydronapthalene-1,4-oxide. The sizes of the induced shifts are such that the spectrum can be treated
by a first-order analysis: the A2B2 pattern of the aromatic protons is clearly recognisable and JAB and JAB′ are readily
measured. The induced shifts should be greatest for the protons closet to the coordination site. Indeed, the least
shifted H1 and H4 protons in the uncomplexed spectrum [Fig. 7.32(a)] are strongly shifted in the shifted spectrum
[Figs 7.32(b) and (c)].
7.33.2 Superconducting Magnets

Chemical shift is directly proportional to the strength of applied magnetic field while coupling constant is
independent of it. These days superconducting magnets with field strength of 25 to 30 tesla, i.e. a million
times stronger than the earth’s magnetic field are available. Consequently, it is possible to obtain the proton
spectra in the 60–1000 MHz region. Thus, the overlapping signals can be separated, using magnets of different
field strengths. However, this method is not preferred over shift reagents since it is neither efficient nor
economically viable.
7.33.3 Double Resonance A

JAX
X
JAX
Double resonance implies that the spin system, i.e. sample is subjected nA nX
(a)
to two or more radiofrequency fields simultaneously known as irra- X
diation (secondary) and observation (primary) fields respectively. The
observation field having amplitude H1 and frequency v1 is usually weak
while the irradiation field of amplitude H2 and frequency v2 is strong
enough to cause perturbation of one or more of the irradiated transi- (b)
tions. The effective width, i.e. g H2/2p of the irradiation field, is main- n2
tained around its centre of frequency v2 so that the lines that differ from Fig. 7.33 (a) Highly resolved NMR spectrum of
v2 by few multiples of g H2/2p are not affected by the irradiation field. an AX spin system, and (b) NMR spectrum of AX
Some of the typical techniques based on double resonance used for the spin system using spin decoupling. While record-
simplification of spectra are described here. ing the X nuclei line, irradiation is performed on
the A nuclei resonance.
(a) Spin Decoupling Consider an AX spin system in which each of CH3 CH2 OH CH3
both the nuclei have spin = 1/2. Due to s-s interaction, the first-order
spectra (vdAX >> JAX) of such a system consist of two doublets centred at (a)
resonance frequencies vA and vX respectively and each doublet is sepa- OH 1 ppm

rated by JAX as shown in Fig. 7.33(a).
For an ideal spin decoupling, the widths of the two basic groups CH2
involved are both small as compared with the difference in their
centre frequencies. The s-s splitting of signal of X nuclei by A nuclei
is removed by irradiating the sample with a strong decoupling field of
strength H2 and frequency v2 centred at vA while the signal due to X
nuclei is observed by the usual method. The doublet of X nuclei in the (b)
decoupled spectra appears as a singlet [Fig. 7.33(b)]. The decoupling
of X doublet into a singlet arises because the decoupling frequency, v2
causes such a rapid transitions between the possible spin states of A
that X nuclei are no more able to sense the spin state of A.
Further, if both the nuclei in the system are of same type, i.e.
1
H—1H, 19F—, 19F, etc., the decoupling is termed as homonuclear spin Fig. 7.34 Highly resolved spectrum of ethanol
decoupling and heteronuclear spin decoupling if both the nuclei are of using spin decoupling: (a) while recording the
different kind, i.e. 1H—19F, 1H—31P, 1H—13C, etc. Figure 7.34 displays methyl group line, irradiation is performed on
the methylene resonance, and (b) while record-
a highly resolved proton spectrum of ethanol using spin decoupling.
ing the methylene group line, irradiation is per-
While observing the methyl group signal, irradiation is performed on formed on the methyl group resonance.
the methylene resonance and vice versa.
(b) Spin Tickling Double resonance in spin tickling involves a primary (H1, n1) and a weak secondary (H2, n2)
rather than a strong, radio frequency fields. The effective width of gH2/2p of the secondary field is made compa-
rable to the smallest splitting, Δv observable with the NMR instrument. Under these conditions, a partial decou-
pling of the NMR signal takes place and in some circumstances extra
multiplet splitting may also be observed. Double resonance behaviour
under spin tickling may be stated in the form of three rules: (i) When a
nondegenerate connected ‘a’ transition is selectively irradiated, all con- a b
nected transitions say ‘b’ and ‘c’ split into doublets. (ii) The doublets
corresponding to regressive transitions are well resolved while those of
progressive transitions are broad. (iii) The doublet splitting is not only c
n2
proportional to the square root of the intensity of the irradiated signal
but also to the amplitude H2 of the secondary field.
These rules are valid only for symmetrical doublets when v2, the radio a
b c
frequency of the secondary field, H2 is centred exactly on the connected
‘a’ transition. However, if the weak secondary field is applied in the neigh- b′ c′
bourhood of one single nondegenerate connected ‘a’ transition but slightly
Fig. 7.35 Schematic diagram showing the
off resonance, all connected ‘b’ and ‘c’ transitions split into asymmetrical splittings obtained in the b and c lines when
doublets. The intense line of such a doublet nearly coincides with the the connected ‘a’ transition is perturbed by
position of the corresponding single resonance line but weak satellite is an irradiation field H2 of frequency ν2 .
displaced from that position to the left in the regression doublet and to the right in the progression doublet as shown
in Fig. 7.35. Spin tickling leads to measurements of fundamental NMR constants, particularly the relative signs of
coupling constants.
(c) Broadband Decoupling Broadband decoupling is a type of heteronuclear decoupling in which the
13
C nuclei are decoupled by hydrogen nuclei. The decoupling is carried out by a broadband radio frequency
of the irradiation field H2, that covers the whole range of proton spectral region, while the spectrum of 13C
nuclei is simultaneously monitored. Area under the 13C peaks is enhanced by a factor than that expected from
the collapse of multiplets into singlets. The unexpected increase in intensity of the 13C signals is because of
Nuclear Overhauser Effect (NOE) that arises because of the d-d interaction between a decoupled l3C nuclei
and neighbouring protons, e.g. broad band decoupled, 13C-NMR spectrum of, 13CH3CH2Cl consists of a singlet
which is due to l3C of CH3 group.
(d) Off-spin Resonance Decoupling Despite the fact that the broadband decoupling simplifies the complex
spectra but the disadvantage is that much of the information pertaining to structural elucidation contained in the
uncoupled spectra is lost. So, instead of using broadband radio frequency field, the decoupling field is set at 1–2
kHz above the proton spectral region, which causes a partial decoupling because of which all but the largest s-s
shifts disappear in the off-spin resonance spectra. Under these conditions, CH3 group gives a quartet pattern in the
13
C-NMR spectrum, CH2 group appears as a triplet, CH exhibits a doublet and quarternary carbon yields a singlet.
Figure 7.36 displays the broadband and off-spin proton decoupled, 13C-NMR spectra of p-ethoxybenzaldehyde at
25.2 MHz.
(e) Nuclear Overhauser Effect (NOE) When an electron spin resonance in a metal is irradiated by a strong
microwave frequency field of amplitude H2 and frequency v2, the electron spin resonance undergoes relaxation by
transferring its spin energy to nuclear spin, thereby enhancing the intensity of the simultaneously monitored nucle-
ar spin magnetic resonance signal by the usual method. The efficiency of the relaxation process depends upon the
distance between the nucleus and the unpaired electrons. The relaxation of the unpaired electron spins and of the
nuclei is coupled by s-s/or d-d interactions so that both spins flip simultaneously. For d-d interactions, a downward
C2 C3
2 3
C6 C6 7 8 9
OHC 1 4
OCH2CH3
6 5
C7 C8
C4 C1 C9 TMS
200 ppm 100 0

δc
(a)
200 ppm 100 0

δc
(b)
Fig. 7.36 Comparison of (a) broadband, and (b) off-resonance decoupling in 13C
spectra of p-ethoxybenzaldehyde.
flip of electron spin (↓) is accompanied by an upward flip of nuclear spin (↑) and vice versa. Here, the electron spin
relaxation process leads to enhancement of nuclei population over that predicted from Boltzmann law in the upper
nuclear magnetic levels. Consequently, the NMR signal becomes an emission signal with enhanced intensity. On
the other hand, when s-s interactions dominate over the d-d interactions, the electron spin relaxation process causes
an increase in population of nuclear spins in the lower levels. As a consequence, the NMR signal becomes an ab-
sorption signal with enhanced intensity. The phenomenon of increase in intensity of NMR signal in case of metals
by such relaxation mechanism is known as Nuclear Electron Overhauser Effect (NEOE). An analogous effect has
been observed in nuclear magnetic resonance also and is termed as Nuclear Overhauser Effect (NOE) and is also
sometimes called spin pumping. The effect arises from direct magnetic coupling, between a saturated nucleus and
a neighbouring simultaneously monitored decoupled nucleus that results in an increase in the population of lower
energy level of the monitored nucleus over that predicted from Boltzmann law.
PMR measurements are carried out in solvents without protons such as CC14, CDC13 and also solvents devoid
of oxygen since it is paramagnetic, so that the intermolecular proton–proton and electron–proton spin interac-
tions are minimised and the d-d interactions dominate over them. Thus the main mechanism left for the proton
spin relaxation amongst the solute molecules is d-d interactions between the neighbouring protons on the same
molecule. So when one particular proton is irradiated by a strong radio frequency field (H2, v2) at its resonance
frequency, the relaxation rate of the neighbouring decoupled proton increases. Consequently, the intensity of the
observed NMR signal from the decoupled proton enhances.
The fractional growth f (I1)(I2) of the integrated intensity of spin I1, (when I2 is saturated) compared to its equi-
γ( )
librium intensity, α where γ ( ) and γ ( ) are the gyromagnetic ratios of spins I2, and I1 respectively. For
γ( )
small molecules, where only d-d relaxation mechanism is operative,
γ (I ) γ ( )
(I ) =
f (I ))(I = (7.115)
γ (I ) γ ( )
For protons the maximum increase of intensity is of the order of 50 per cent, 13C-nuclei are relaxed predominantly
by d-d interactions with protons. Irradiation of proton resonances by strong broadband radio frequency field led to
an increase in intensity of l3C-NMR signal by ~2 from NOE but the disadvantage is that the proportionality between
the areas under the peaks and number of 13C nuclei is lost. The g of an electron is larger than that of nucleus, so the
saturation of electron spin can cause an increase in the intensity of NMR signal of the order of 1000.
For large molecules such as proteins and nucleic acids, negative NOE has been observed, i.e. f (I ))(I (I ) is
negative. The negative NOE may be due to (i) the presence of third spin proximate to the spins, (ii) the presence
of nuclei with gyromagnetic ratios of opposite sign, (iii) the presence of scalar coupling modulated by chemical
exchange or internal motion, and (iv) the dependence of NOE on frequency in case of macromolecules.
NOE and consequently the technique based upon it is useful to determine the stereochemical relationships
in NMR active molecular systems that may be difficult by chemical shifts and coupling constants, e.g. the two
methyl resonance signals in N, N-dimethylformamide (DMF) are not clearly distinguished by chemical shifts
or coupling constants. Irradiation of low field (high d value) methyl resonance in 8 per cent solution of DMF,
enhance the intensity of the decoupled formyl proton signal by 18 per cent. On the other hand, saturation of the
high-field methyl signal decreases the intensity of the formyl proton signal by 82 per cent. This led to suggest that
low-field signal is because of the cis-methyl group rather than the trans-methyl group. The PMR of b, b -dimethyl
acrylic acid [(CH3)2C = CHCOOH] exhibits two doublets at d = 1.97 and 1.42 ppm (J = 1.3 Hz) from methyl
groups and a septet because of olefinic proton at higher d value. Saturation of cis-methyl group causes a 17 per
cent increase in the intensity of formyl proton signal while no change in intensity of formyl proton resonance is
observed when trans-methyl group is irradiated. On the other hand when formyl protons are irradiated, the intensi-
ties of resonance signals of either of the methyl groups remain unaffected because the methyl protons relax one
another. The contribution of outside protons is little so that saturation causes no change in the observed methyl
group line intensities.
7.34 TWO-DIMENSIONAL FOURIER TRANSFORM NMR (2DFTNMR)

2DFTNMR is the extension of 1DNMR spectroscopy. The FID signal in 1DFTNMR is recorded as a function
of time t2 for a fixed evolution time t1, and is represented by the function S(t2). However, when a large number
of one-dimensional signals are recorded as a function of time t1, i.e. FID as a function of t2 at various values of
t1, these are represented by the functions S1(t2), S2(t2), S3(t2),…Sn (t2), i.e. Si(t2), where i = 1, 2, 3,…,n, or a two-
dimensional array S(t1, t2). Such an array of one-dimensional signals is collected and is Fourier transformed
twice to obtain a two-dimensional spectrum in terms of two frequency variables v1 and v2 or sometimes the
chemical shift parameters d1 and d2, i.e. f (v1, v2)/f (d1, d2),
S(t1, t2), .... (7.116)
2 × FFT ... f (v1, v2)

where FFT refers to the fast Fourier transform algorithm.
Thus, 2DFTNMR is a technique to obtain two-dimensional NMR spectrum in terms of two frequency variables
v1 and v2 by performing Fourier transformation twice on a large number of one-dimensional FID signals recorded
at various values of evolution time, t1.
The structure of the array S(t1, t2) depends on the chemical shift, coupling constants present at the periods
during which they are expressed, the evolution periods (t1) used and the nature of the applied pulses and the times
during which decoupler is operating. Thus a large number of different types of two-dimensional techniques have
been developed (more than 100) and some selected 2DFTNMR techniques are listed in Table 7.7. In heteronu-
clear J-resolved spectroscopy, the experimental conditions are adjusted such that only one parameter, say spin
coupling, is expressed during the evolution period and the second parameter, say chemical shift, is expressed
during the detection period. The pulse sequence used for a heteronuclear 1H—13C 2D J-resolved NMR spectros-
copy using the 1H-decoupler method is shown in Fig. 7.37. In order to obtain the spectrum, the proton decoupler
is turned off and a p/2 pulse is applied to the system and it is allowed to evolve for time t1. After time t1, the
decoupler is turned on and a p pulse is applied and the system is allowed to evolve for time t1. The carbons, there-
fore, evolve for half of the time as singlets and for the remainder of the time as proton coupled multiplets. After
the equilibrium is re-established, the process is repeated for other values of t1. The FID signals are digitised and
Fourier transformed twice to obtain the 2DNMR spectrum. The 2DNMR spectrum is represented in the form
Table 7.7 Selected 2DFTNMR techniques.

Pulse Sequence Experiment f1 f2 Information
π Heteronuclear Heteronuclear
−τ π − τ − t2 JAX dX
2 J-resolved coupling constants
π Homonuclear J
−τ π − τ − t2 Homonuclear 2DJ JAA dA
2 and d
CHESY or HETCOR Correlation of dX

dA dX
or HETEROCOSY and dA
Identification of
π π 2DCOSY or
− τ − − t2 dA dA all scalar coupling
2 2 (HOMCOR) (2-D)
interactions
Spatial proximity
π π π NOESY dH dH
− τ − − tm − − t2 and three-dimensional
2 2 2 EXSY JHH JHH
molecular structure
Homonuclear
π π 2D-INADE-QUATE dA + dX dX connectivities
− t D − π − t D − − t 2 (1 ) (Double quantum
2 2 (coupling)
Coherence experiment)
H1−H1, 13C−C13
π π
− tD − π − tD −
2 2
π
−τ − − t 2 (13 )
2
t—incremented delay;
tm—mixing periods help in causing magnetisation transfers between the spins in the network,
HETCOR—Heteronuclear chemical shift correlation or HETEROCOSY—Heteronuclear correlated spectroscopy,
CHESY—Chemical shift heteronuclear spectroscopy,
COSY—Correlation spectroscopy or HOMCOR—Homonuclear correlated,
NOESY—Nuclear Overhauser effect spectroscopy, EXSY—Exchange spectroscopy,
2D-INADEQUATE—Incredible natural abundance double quantum transfer experiment.
of a map which contains two frequency axes, f1 and f2. One of π/2 π
the two frequency axes, f2 always represents the parameter,
e.g. chemical shift, of the observed (acquired) nucleus and the 13C τ τ
other f1, axis may represent information, e.g, chemical shift,
spin coupling or both, that depends on the pulse sequence and
the evolution time, i.e. the behaviour of the spins during the 1H Decoupler
evolution period appears along f1-dimension, e.g. in the 13C-
HET2DJ spectrum, the f2-axis gives the chemical shift of the
observed 13C nucleus and the f1-axis gives coupling constant AQ
for l3C—1H. The position of the signal on the map tells us
about both the chemical shift of that signal, and another piece Fig. 7.37 Pulse sequence for 1H⎯13C two-dimensional
of information which depends on the nature of f1. J-resolved Spectroscopy using the gated decoupler method.
The display mode of 2DNMR is either the contour plots or
stack plots. Contour plots enable to see all the responses at
once whereas the stack plots usually have most interesting peaks hidden behind another, less important signal.
Contour plots are the excellent way of abolishing the noise in 2DNMR spectrum Stack plots give a more realistic
picture of signal-to-noise ratio, and line shape but are otherwise less convenient. The users of 2DNMR spectrum
who do not themselves decide on the contour levels to be drawn and the readers of NMR literature should be care-
ful about the contour plots they see before them. The plots might be disguising more than they reveal.
The power of 2DNMR technique to unravel the complex spectra has been demonstrated with the following
examples.
1
H2DCOSY Spectrum for m-dinitrobenzene The process of correlating scalar J couplings is called
COSY. The basic COSY pulse is: p/2 − t − p/2 − t2. The p/2 preparation pulse is applied to the sample.
It excites the coherences (transitions) for an evolution time t1. During this period, the coherences are
frequency labelled. i.e. these precess during t1. Subsequently, a p/2 mixing pulse is applied which
performs a controlled transfer of coherence to different nuclear spin transitions where the coherences
continue to precess with a new frequency during time t2. After the equilibrium is re-established, this
process is repeated for other values of t1. The FID signals
are digitised and Fourier transformed twice to obtain 2D
spectrum correlating the frequencies during the evolu- O2N
2
NO2
tion with the frequencies during the detection. The set
of spectra will have normal shifts and coupling in f2, but 6 4
each peak will contain coded information about the fre- 5
quencies of all the other peaks to which it is coupled. (b)
The second Fourier transformation then gives a spec- H5
H2 H4,6
trum with 1H chemical shifts in each axis. The normal
spectrum will be found along the diagonal because the
intensity of each transition oscillates at v1 during t1 and
because its detected frequency in t2 is also v1. The off-
diagonal cross-peaks show which chemical shifts are
spin coupled to one another, because the intensity of v1
transition is modulated during t1 by all the transitions to (a)
which it is attached.
The basic 2DCOSY spectrum of m-dinitrobenzene is com-
pared with its 1DNMR spectrum in Fig. 7.38. The three multiplets
of the one-dimensional spectrum are along the diagonal line.
The off-diagonal cross peaks are observed at the intersection of
the chemical shifts of H4 6 with H2 and of H4 6 with H5. H5 is a
triplet because it is coupled equally to H4 and H6. H2 is also a
triplet because of its coupling with H4 and H6. However, in the
COSY spectrum, the central line expected in both the triplets is
completely missing from the off-diagonal cross-peaks because
the intensity distribution arising from population transfer from
H4 to H6 is l : 0 : −1. Similarly, the central lines of quintets are
Fig. 7.38 (a) 250 MHz 1H COSY-p/2 spectrum of
often missing. m-dinitrobenzene, and (b) the one-dimensional spectrum.
Although 1HCOSY experiments are most common, 1 2 3 4

CH3CHOHCH2CH2OH
the technique may be applied to observe nuclei having
diverse natural abundance, i.e. 11B, l9F, 31P and 51V and
to materials that are enriched in nuclei such as 13C
or 29Si. Let us now examine the spectrum in which a 20
proton decoupled spectrum appears in f2-dimension 1
and proton coupled spectrum is obtained in f1-dimen-
sion. The 2DNMR spectrum of l,3-butanediol is shown
in Fig. 7.39 along with its 1DNMR spectrum. The 2D
spectrum consists of a quartet (—l3CH3), two triplets δc
(— CH2, — CH2OH) and a doublet (— CHOH).
l3 l3 13
3 40
These multiplets appear due to the interaction of pro-
tons (with l3C) attached to each of the four l3C nuclei
in the molecule. On the other hand, this information
is missing in 1DNMR spectrum. In conclusion, the
2DNMR spectrum can be used to identify l3C-peaks in 4
the 1DNMR spectrum. 60
3DNMR spectroscopy is an extension of 2DNMR 2
spectroscopy and can be treated as a combination
of two 2D experiments. Instead of a single mixing (a)
process which relates two frequency variables, two 4 3
sequential mixing processes relate three frequencies: 2 2 1 1
the original (first evolution) frequency, v1; the second 4 4 3 3
1
1
evolution frequency, v2; and the detection frequency,
v3. Likewise, higher dimensional experiments can be
conceived and generated in a similar manner. The sche- 60 40 20 (ppm)
matic diagram for one-, two- and three-dimensional δc
NMR pulse sequences is shown in Fig. 7.40. The time Fig. 7.39 Illustration of the use of the two-dimensional stack
axis is segmented into various periods as indicated. spectrum (a) to identify the 13C peaks in a one-dimensional
The preparation period is used to prepare the spectrum of 1,3-butanediol (b).
nuclear-spin system in a suitable equilibrium or a non-
equilibrium state depending upon the purpose of the experiment. The mixing periods following the evolution
periods help in causing magnetisation transfer between nuclei spins in the network which are frequency labelled
during evolution and thus the detected signals carry information about the history of spin system evolutions. The
data are actually collected only during the detection period.
(a) 1D Pulse Sequence

Preparation Detection(t2)
(b) 2D Pulse Sequence
Preparation Evolution (t1) Mixing Detection(t2)
(c) 3D Pulse Sequence
Evolution Mixing Evolution Mixing Detection(t3)

Preparation
(t1) tm(1) (t2) t m(2)
Fig. 7.40 Schematic diagram of (a) one-, (b) two- and (c) three-dimensional NMR pulse sequences.
7.35 13
C-NMR SPECTROSCOPY
13
C-NMR was first studied in 1957. With the advent of superconducting magnets and Fourier transfer techniques,
it has become a versatile tool for the structural elucidation of molecules.
13
C is the naturally occurring stable isotope of carbon and, therefore, is present in all carbon compounds. The
l3
C nucleus resonates at 15.09 MHz in a magnetic field of 14092 G. Owing to low magnetic moment and low
natural abundance (1.1 per cent), the sensitivity of 13C resonance signal in natural abundance samples is only
3
⎛ γ 13C ⎞ 1 1
⎜⎝ γ ⎟⎠ × 100 = 6300 of that of equal number of proton nuclei. In order to overcome this problem, large sample
H
tubes and strong radio frequency field Hrf are used to obtain l3C resonance signals. Moreover, the s-l relaxation
time (T1) of l3C nucleus is large as compared to that of proton and varies from few seconds to several minutes, e.g.
relaxation time of 13C nucleus in small highly symmetrical molecules lies in the range of 10–15 seconds at room
temperature, in larger molecules with slower reorientation time in solution phase such as cholesteryl chloride,
many of the l3C nuclei have relaxation time less than 1 second and only side chain carbons linked to protons have
relaxation times in the range of 2–8 seconds. Consequently, unless the scan of 13C nuclear resonance is very fast,
the l3C multiplets show distorted peak intensities because of magnetisation.
The intensity of l3C resonance signal depends upon the relaxation time and roughly it is inversely proportional
to the relaxation time of the nucleus under consideration, e.g. the intensity of secondary carbon nuclei (two pro-
tons attached; 2–4 seconds) > primary carbon nuclei (three protons attached; 10–20 seconds) > quaternary carbon
nuclei (no protons attached; 30 seconds). The low intensity of 13C peak of CH3 group is used to distinguish it from
peaks of CH2 or CH groups. Enhancement of 13C-resonance signal can be obtained in the presence of free radicals
(Overhauser effect). The l3C spectrum usually exhibits normal intensity pattern after the addition of a solution of
paramagnetic ion such as iron or chromium to the sample. Free electrons in such reagents are efficient relaxers
and 13C spectrum usually exhibits normal intensity pattern after the addition of such reagents. In case the reagent
complexes with the sample, the relaxation becomes too efficient that the NMR signal becomes too broad to be
observed.
Positive enhancement of 13C signal is observed in compounds such as CH3C1, CH2C12, CH2Br2, CH2I2 and
CC14. Here, the lone pairs of electrons of halogen atoms participate in the transfer of electron density from the
radical to the l3C nucleus. This process facilitates scalar coupling which leads to positive increase in signal inten-
sity. Negative enhancement has been observed for aromatic compounds, alcohols, ketones and CS2. On the other
hand, positive Overhauser signal enhancements have been observed in aromatic polyfluorocarbons.
Intensities of l3C resonances cannot be simply correlated with the number of atoms contributing to each
resonance.
I3
C-NMR has several advantages over PMR regarding the structure elucidation of molecules. It provides informa-
tion about the backbone of molecules rather than about the periphery. Diamagnetic effect dominates in shielding
the carbon nucleus whereas paramagnetic effect predominates in shielding the hydrogen nucleus. Long-range
shielding effects are less important in 13C-NMR, consequently 13C chemical shifts do not parallel proton chemical
shifts. l3C-chemical shifts/coupling constants to protons, lie in the range 0–200 ppm/0–300 Hz in contrast to 0–20
ppm/0–20 Hz range for protons. Because of this large difference of chemical shifts, the overlapping of signals in 13C
spectra is less as compared to PMR spectra. The l3C chemical shifts of some typical systems are listed in Table 7.8.
Substituents affect the 13C chemical shifts and shifts of polar molecules are solvent dependent. The effect of sub-
stituents on 13C shifts is similar to that of 1H chemical shifts. Heteronuclear s-s coupling between 13C and 12C does
Table 7.8 The approximate chemical shift ranges of some 13C

resonances relative to TMS.
System Chemical Shift (in ppm)
CH3 C 9−29
CH C 31−59
C C 28−52
—CH3—Hal 3−25
—CH2—Hal 4−40
CH2 Hal 30−66
≡ C – Hal 33−76
CH3 N 12−40
CH2 N 42−58
(Continued)
Table 7.8 The approximate chemical shift ranges of some 13C

resonances relative to TMS—Cont'd.
CH N 53−68
C N 59−74
CH3—O 49−60
—CH2—O 42−69
CH O 64−77
≡C-O 69−83
—C ≡ C— 72−96
C = C (alkene) 105−145
C = C (aromatic) 105−160
—C ≡ N 116−127
C = O (ester, amide) 164−175
C = O (acid) 168−179
C = O (ketone) 196−213
C = O (aldehyde) 193−205
C = O (quinone) 179−190
—O—C ≡ N 103−120
—S—C ≡ N 97−118
—N = C = S 117−141
not take place since I for 12C = 0. Owing to low natural abundance of 13C, the probability of occurring of 13C atom
adjacent to other 13C atom in the molecule is remote, and consequently, the homonuclear s-s interaction between l3C
nuclei in molecules is not observed. Thus, the 13C-NMR spectrum yields peak for individual carbon atoms. How-
ever, 13C nuclei bound directly together in a molecule or separated from each other by two or three bonds give rise
to the AX type spectrum. Protons attached directly to 13C nuclei also exhibit AX type spectrum, e.g. primary carbon
nucleus yields a quartet with intensity ratio 1 : 3 : 3 : 1, secondary carbon nucleus shows a triplet having intensity
ratio 1 : 2 : 1, tertiary carbon yields a 1 : 1 doublet and quarternary carbon exhibits a singlet only. The coupling
constants of primary, double bonded and triple bonded carbon to proton follows the order.
J C H = 125 Hz < J C C H = 170 Hz
< J C C H = 250 Hz
Assigment b
a
a 31.1 O CH3
b 36.2
c 162.4 Cc N
b a
H CH3
c
Solvent
190 180 170 160 150 140 130 120 110 100 90 80 70 60 50 40 30 20 10 0
dc (in ppm)
Fig. 7.41 C-NMR spectrum of N, N-dimethylformamide.
13
b a
Assignment O
a 28.8 t
b 40.4 d Cc b a
c 173.6 HO CHCH2Br
c
Br
Solvent
190 180 170 160 150 140 130 120 110 100 90 80 70 60 50 40 30 20 10 0
δc (in ppm)
Fig. 7.42 13
C-NMR spectrum of 2,3-dibromopropionic acid.
Assignments e
d
a 15.3 a a
b 18.1 Br H Br CH3
c 104.7 C f d e
C C C C
d 108.9 b
e 129.4 H CH3 H H
f 132.7 trans cis b
f c
Solvent
190 180 170 160 150 140 130 120 110 100 90 80 70 60 50 40 30 20 10 0
δc (in ppm)
Fig. 7.43 13
C-NMR spectrum of 1-bromo-2-methylene.
The large difference in chemical shifts of 13C nuclei as well as in the JCH coupling constants is helpful in assign-
ing resonances to a particular 13C nuclei in the molecule. However, it spreads out the intensity of the already weak
13
C resonance, signal. This is overcome by simplifying the 13C spectrum by double resonance technique called the
broad band decoupling. 13C-NMR spectra of some typical molecules are shown in Figs 7.41 to 7.43.
13
C shifts are additive and constitutive. The chemical shifts of simple unsubstituted hydrocarbons may be
predicted using additivity relations, although there are deviations from the constitutive additivity rule when
the hydrocarbons are substituted by polar groupings. The additivity rule for polar and branched chain alkanes
is given by
= −2 5 + ∑ nA (7.117)
Here, d is the predicted chemical shift for a carbon atom, A is the additive shift parameter whose values for 13C
atoms in linear and branched chain hydrocarbons have been reported in Table 7.9 and n is the number of carbon
atoms for each shift parameter (−2.5 is the shift for the l3C of methane). The effect of substituents on linear and
branched alkanes is listed in Table 7.10.
Table 7.9 13C shift parameters in some

linear and branched chain hydrocarbons.
l3
C atoms Shift (ppm) (A)
a + 9.1
b + 9.4
g -2.5
d + 0.3
e + 0.1
0
1 (3°) a
−1.1
l°(4°) a
−3.4
2°(3°) a
−2.5
2°(4°) −7.2
3°(2°) −3.7
3°(3°) −9.5
4°(1°) −1.5
4°(2°) −8.4
1°(sp3C,—CH3), 2°(sp2C,—CH2), 30(spC, —CH), 4°(quarternary C). a— The notations 10(3°) and 1°(4°)
denote a CH3 group bound to a R2CH and to a R3C group, respectively. The notation 2°(3°) denotes a RCH2
group bound to a R2CH group and so on. Carbons a, b, g, d and e are one bond, two bonds, three bonds, four
bonds and five bonds apart from the carbon atom (C*) where d is to be estimated, i.e.
* α β γ δ ε β α * α β
C C C C C C C C C C C
α α β
C C C
β α α β α β γ
C C C* C C C C* C C C
α
C
β γ
β α
C C C C
β α α γ δ
C* C C C C* C *
α C C C C C
α β
α
βC C
The deviation, DX, from the constitutive rule, in ppm, for a given X in the series alkyl-X is given by
DX n ( 50 − d x ) .6 (7.118)
where n is the number of carbon atoms attached to the monosubstituted a-carbon atom and dx is the C—X bond
distance. With the aid of this relationship, the chemical shifts of the a -carbon atoms in C2H5X, i-C3H7X, and
t—C4H9X can be estimated to an accuracy of ±3 ppm for X = F, Cl, Br, I, OH, NO2 and COOH. The C—F, C—Cl,
C—Br and C—I bond lengths are 1.38, 1.77, 1.94 and 2.12 Å respectively.
Empirical relations for the prediction of l3C—1H coupling constants are also available in the literature. The
J 13C 1H of monosubstituted methanes is given by
J 13C H
d χ 94
94.9 (7.119)
Here, c is the substituent Mulliken electronegativity.

For polysubstituted methanes,
J 13C 1
H = c[3.25 nd + 4.6(n − l)] − 68.74 + 163.6 (7.120)
Table 7.10 Incremental substituent effects (ppm) on replacement of H by Y in alkanes. Y is terminal or internal
( + down-field, − up-field).
Y
γ α γ γ
α
Y
β β β
Terminal Internal
a b g
Y Terminal Internal Terminal Internal
CH3 +9 +6 + 10 +8 −2
CH = CH2 + 20 +6 −0.5
C CH + 4.5 + 5.5 −3.5
COOH + 21 + 16 +3 +2 −2
− + 25 + 20 +5 +3 −2
COO
COOR + 20 + 17 +3 +2 −2
COOl + 33 + 28 +2
CONH2 + 22 + 2.5 −0.5
COR + 30 + 24 +1 +1 −2
CHO + 31 0 −2
Phenyl + 23 + 17 +9 +7 −2
OH + 48 + 41 + 10 +8 −5
OR + 58 + 51 +8 +5 −4
OCOR + 51 + 45 +6 +5 −3
NH2 + 29 + 24 + 11 + 10 −5
NH 3+ + 26 + 24 +8 +6 −5
NHR + 37 + 31 +8 +6 −4
NR2 + 42 +6 −3
+
NR 3 + 31 +5 −7
NO2 + 63 + 57 +4 +4
CN +4 +1 +3 +3 −3
SH + 11 + 11 + 12 + 11 −4
SR + 20 +7 −3
F + 68 + 63 +9 +6 −4
Cl + 31 + 32 + 11 + 10 −4
Br + 20 + 25 + 11 + 10 −3
I −6 +4 + 11 + 12 −1
Add these increments to the shift values of the appropriate carbon atom of the alkanes calculated from Table 7.10.
Here, n is the number of substituents attached to the same carbon atom.
Problem 7.19: Calculate the chemical shifts (in ppm) of individual carbon atoms in (a) hexane, (b) neopen-
tane (c) 2,2-dimethylbutane, (d) 2,2,3-trimethylbutane, (e) 3-chloropentane and,(f) 3-iodopentane.
Solution
(a) The structural formula of hexane is
1 2 3 4 5 6
CH3 CH2 CH2 CH2 CH2 CH3
Carbon atom 1 has la-, 1b-,1g-, 1d-and le-carbon atoms.So

d1 = −2.5 + la + 1b + 1g + 1d + 1e
= −2.5 + (1 × 9.1) + (1 × 9.4) + (1 × −2.5) + (1 × 0.3)+ (1 × 0.l)
= 13.9 (obs = 14.1)
Carbon atom 2 has 2a-, 1b-, lg-, and 1d-, carbon atoms.
d2 = −2.5 + (2 × 9.1) + 9.4 − 2.5 + 0.3
= 22.9 (obs = 23.1)

The carbon atom 3 has 2a-, 2b-, 1d-, carbon atoms.
d3 = − 2.5 + (2 × 9.1) + (2 × 9.4) + (1 ×−2.5)
= 32.0 (obs = 32.2)
(b) The structural formula of neopentane is
4
CH3
1 2 3
CH3 C CH3
5
CH3
Carbon atom 1 has la- and 3b-carbon atoms. It is a 1° atom with one 4° atom attached.
[1°(4°) = −3.4]
d1 = −2.5 + (1 × 9.1) + (3 × 9.4) − (1× 3.4)
= 31.4 (obs = 31.7)

The carbon atom 2 has 4a carbon atoms. Carbon 2 is a 4° carbon with four 1° atoms attached
[4°(1°) = −1.5]. So
d2 = −2.5 + (4 × 9.1) + (4 × −1.5)
= 27.9 (obs = 28.1)

(c) The structural formula of 2,2-dimethylbutane is
5
CH3
1 3 4
CH3 C CH2 CH3
6
CH3
The carbon atom 1 has la-, 3b- and lg-carbon atoms attached and is a 1° carbon with one 4° carbon
attached. [1°(4°) = −3.4]
d1 = −2.5 + (1 × 9.1) + (3 × 9.4) − 2.5 − 3.4
= 28.9 (obs = 29.1)

The carbon atom 2 has 4a-, and 1b-carbon atoms attached. Carbon atom 2 is a 4° carbon with three 1°
carbons and one 2° carbon attached. [4°(1°) = −1.5, 4°(2°) = −8.4]
d2 = −2.5 + (4 × 9.1) + 9.4 + (3 × −1.5) − 8.4
= 30.4 (obs = 30.6)

The carbon atom 3 has 2a-, and 3b-carbon atoms attached. Carbon atom 3 is a 2° with a 4° carbon attached
[2°(4°) = −7.2]
d3 = −2.5 + (2 × 9.1) + (3 × 9.4) − 7.2
= 36.7 (obs = 36.9)

The carbon atom 4 has la-, 1b- and 3g-carbon atoms attached.
d4 = −2.5 + 9.1 + 9.4 + (3× −2.5) = 8.5 (obs = 8.9).
(d) The structural formula of 2,2,3-trimethylbutane is
5 7
CH3 CH3
1 3 4
CH3 C CH CH3
6
CH3
d1 = −2.5 + la + 3b + 2g + 1°(4°)
= −2.5 + 9.1 + (3 × 9.4) + (2 × −2.5) − 3.4
= 26.4 (obs = 27.4)
d2 = −2.5 + 4a + lb + 3 × 40(1°)
= −2.5 + (4 × 9.1) + 9.4 + (3 × −1.5)
= 38.8 (obs = 38.3)
d3 = −2.5 + 3a + 1b
= −2.5 + (3 × 9.1) + 9.4 = 34.2 (obs = 33.1)
d4 = −2.5 + la + 2b + 3g + 1° (3°)
= −2.5 + 9.1 + (9.4 × 2) + (3 × −2.5) − 1.1
= 16.8 (obs = 16.1)

(e) The structural formula of 3-chloropentane is
1 2 3 4 5
H3C CH2 CH CH2 CH3
Cl
The calculated values of chemical shifts for individual carbon atoms are
d1 = 13.8 (obs = 13.9)
d2 = 22.6 (obs = 22.8)

d3 = 34.5 (obs = 34.7)
The contribution of Cl atom at position ‘3’ is + 31 (Table 7.10). So d3 in 3-chloropentane = 34.5 + 31 = 65.5
(f) Proceeding as above, d3 in 3-idopentane = 34.5 − 6 = 28.5.
7.36 19
F-NMR SPECTROSCOPY
F possesses a spin quantum number of 1/2 and thus exhibits characteristic NMR absorptions. The sensitivity of
l9
F(mI = 2.6273) is close to that of 1H (mI = 2.79270). The energy level diagram for l9F nucleus in terms of MI is
19
as follows.
1
− −
2
1
+ −
2
Thus, a l9F nucleus attached to another 19F nucleus senses two different fields, all equally probable. Consequently,
the 19FNMR spectrum of a 19F nucleus attached to another 19F nucleus will be a 1 : 1 doublet. The I9F-NMR
spectrum of HF involves coupling of the fluorine nucleus with a proton. Since the 1H nucleus has a spin of 1/2, the
l9
F nucleus can be in either of the two net fields ( + 1/2, −1/2) of protons. The 1H spin may be parallel or antipar-
allel to the external field. Thus, l9F-NMR signal of HF will be a symmetrical doublet. Similarly, the PMR spectrum
of proton attached to l9F will consist of a symmetrical doublet.
The Larmor frequency for l9F appears at lower energies than that of the proton by a ratio of 1/1.063. Thus, on a
30 MHz proton spectrometer, 19F is observed at 28.22 MHz, on a 60 MHz at 56.44 MHz, on a 100 MHz at 94.07
MHz and on a 200 MHz at 188.14 MHz.
The fluorine chemical shifts (dF) and coupling constants are much larger (10 fold more) than those of protons.
Fluorine chemical shifts are sensitive to chemical environments and spread over a range of 300 ppm. J(19F − 1H)
lies in the range of 0–90 Hz. 19F chemical shifts of some selected compounds (in ppm) relative to CFCl3 are CH3F
(278), HF(203), F−(aq) (120), PF5 (96, 62), CF3H (88), (CF3)2 CO (82), CF3COOH (77), CF4 (70), PF3 (68, 12),
(CF3)4 C (61.5), CF3Cl(32), CF2C12(9), CF3I (4), CFC13 (0), IF5 (−4), SOF2 (−70), ClF3 (−80), ClF3 (−116), BrF3
(−132, −269), CF3OF (−142), OF2 (−250), F2 (−422). The larger positive numbers show higher fields.
The difference between fluorine chemical shift in F2 and in F− (aq) is 542 ppm compared to the range of about
12 ppm for proton chemical shifts.
Because of high sensitivity, large chemical shift and small size of ,19F nucleus, it is used as a site specific probe
for biochemical systems. 19F−NMR has been successfully used to differentiate between amino acids (see biological
applications).
19 −
F NMR may also be used for structural elucidation of fluorine containing molecules, e.g. 19F−NMR spectrum
of PF5 shows one resonance peak instead of two in the ratio 3 : 2, i.e. ratio of axial to equatorial sites. ClF3 has two
long Cl—F bonds and one short Cl—F bond giving rise to non-equivalent fluorines. 40 MHz 19F−NMR spectrum
of ClF3 consists of a triplet and a doublet and the intensity of triplet is half that of doublet as expected. The spectrum
is in accord with the interaction between non-equivalent fluorines.
7.37 31
P-NMR SPECTROSCOPY
The spin quantum number of 31P nucleus is 1/2 (I = 1/2) and its sensitivity is ≈1/15 of that of proton nucleus. In
a homogeneous magnetic field, a given nucleus can assume any one of the 2I + 1 possible orientations relative to
the applied field. Thus, the energy level diagram for a 31P nucleus in terms of MI is as follows.
1
2
1
+ −
2
Thus, a proton attached to 31P nucleus experiences two different fields of 31P nucleus and the probability of
coupling of proton nucleus with the 31P nucleus fields will be the same; consequently, the PMR spectrum of proton
coupled to 31P nucleus will be a 1 : 1 doublet. We would expect one 31P nucleus coupled to another 31P nucleus to
exhibit a 1 : 1 doublet assuming that the two nuclei are under identical chemical environments so that overlap-
ping of peaks due to two 31P nuclei takes place. 31P-NMR spectrum of P4O6 conforms the expected behaviour of
one 31P nucleus towards another 31P nucleus, it exhibits only a single peak at −113 ppm relative to 85 per cent
H3PO4. On the other hand, if the phosphorous nuclei in a system are in different chemical environments, the sig-
nals corresponding to these nuclei will be observed at different positions in the 31P− NMR spectrum. 31P− NMR
spectrum of adenosine-5′-triphosphate (ATP) exhibits three signals corresponding to a a, b and g phosphorous
atoms as shown in Fig. 7.44. The peak (triplet) due to the b−31P nucleus in ATP is more isolated as compared to
peaks (doublets) of a and g −31P nuclei because its environment is unique, it alone is flanked by phosphate groups.
Chemical shifts due to g, a and b phosphorous atoms of ATP in its FT31P,-NMR spectrum in aqueous environment
are observed at -11, −14 and −22.5 ppm respectively.
NH2
6
C N
N
ATP 8 H
2 4 N9
OH OH OH H 3N
γ β α 5′
HO P O P O P O CH2 O
4′
1′
O O O H H
H H
3′ 2′
HO OH
β
20 Hz
α
Fig. 7.44 31
P-FT NMR spectrum of adenosine-5′-triphosphate.
Table 7.11 31
P Chemical shifts and coupling constants of some phosphines.
Phosphine dP, ppm JPH, Hz

PCl3 −215 JPH(1) = (150−250)
PCl2C6H5 −164 JPH(2) = (−5− + 27)
C
P(OCH3) −141 JP C H(2) = (46)
P(OC2H5) −140 JPCCH(3) = (10−20)

P(OC6H5) −128 JPC = CH(3) trans = (5−41) cis = (6−20)
PF3 −97 JPCCCH(4) = (0−5)
P(C6H5)3 + 6.6 Range of coupling constants in
P(C2H5)3 + 19.1 Phosphines: −5− + 250
P(CH3)3 + 62 Phosphine oxides: 2−750
PH3 + 240 Phosphonium salts: 2−900
Phosphites: 5−20
[CH3PH] + + 2.8
Phosphates: 0−30
The numbers between brackets attached to J’s correspond to the number of intervening bonds.
The Larmor frequency for 31P nucleus appears at lower energies than that for proton nucleus by a ratio of
17.235/42.577 = 1/2.470. Thus, on a 60 MHz proton spectrometer, 31P is observed at 24.29 MHz, on a 100 MHz
proton spectrometer at 40.48 MHz and on a 200 MHz proton spectrometer at 80.98 MHz.
The drawback of 31P−NMR owing to low sensitivity of 31P as compared to 1H and 19F nuclei is off set because
like 1H and 19F; 31P has also 100 per cent natural abundance, due to which no labelling technique, i.e. isotopic
enrichment is required as in case of 13C and 15N nuclei. Moreover 31P nuclei have large chemical shifts as compared
to 1H and spread over a range of 0–700 ppm due to which it is easy to handle the 31P−NMR spectrum. The 31P
chemical shifts of some phosphines in ppm (dP), and coupling constants in Hz (JPH) of some typical systems with
respect to 85 per cent H3PO4 are given in Table 7.11.
The coupling constants J(31P—O—31P) in phosphites and phosphates lie in the range 0−20 and 0−30 Hz
respectively.
7.38 NMR SPECTRA OF NUCLEI WITH I > 1/2

The NMR spectra of nuclei having I >1/2 are usually intrinsically weak either due to low magnetic moment or
low natural abundance or because of both, and are difficult to observe. However, the most common nuclei having
I >1/2 attached to hydrogen nuclei are l4N, 35Cl, 79Br and 127I; they may affect the PMR spectra of molecules
through s-s coupling. The spins of these nuclei are 1, 3/2, 3/2 and 5/2 respectively. The energy-level diagrams for
these nuclei in terms of MI are as follows.
Nuclei 14
N 35
Cl, 79Br 127
I
I 1 3 5
2 2
3 1 1 3 −5 −3 −1 +1 +3 +5
MI −1, 0, + 1 − ,− ,+ , , , , , ,
2 2 2 2 2 2 2 2 2 2
Thus, a proton coupled to a I4N nucleus ‘sees’ three different fields, all equally probable. All transitions are
degenerate and the NMR spectrum of proton coupled to a 14N nucleus will be a triplet with all the peaks of equal
intensity. This happens only when nitrogen quadrupole effects do not operate as in symmetrical molecules, N−CH
coupling may be observed, J values being commonly less than 1 Hz. The 31.65 MHz PMR spectrum of tetram-
ethylammonium bromide (CH3)4N + Br− in H2O (Fig. 7.45) exhibits a triplet (1 : 1 : 1) due to (CH3)4 N + ion, J is
∼0.55 Hz. On the other hand, PMR spectrum of CH3NH2 in presence of mineral acid exhibits a quartet (1 : 3 :
3 : 1) for the protons of CH3 group and a triplet with very broad resonance lines (1:1:1) due to protons of NH 3+
group. The quartet due to protons of CH3 group arises because of the interaction of CH3 group protons with pro-
tons of NH 3+ group. The sharp nature of the quartet indicates that the exchange rate between hydrogens attached
to nitrogen is slowed down in the presence of the acid. The broadening of triplet is because of quadrupole relax-
ation which causes a rapid transition between the spin states.
PMR spectra of protons attached to 35Cl/79Br and l27I should consist of a quartet (1:3:3:1) and a sextet (1 : 5 :
10 : 10 : 5 : 1), respectively. However, in practice these multiplets are not observed because the nuclear quadrupole
relaxation causes line broadening.
The Larmor frequency for l4N, 35C1, 79Br, and 127l will 1 Hz
appear at lower energies than that for 1H by a ratio of 1/13.841,
1/10.205, 1/3.991 and 1/4.998 respectively. Thus, on a 60 MHz
proton spectrometer, l4N is observed at 4.3 MHz, on a 100 MHz
proton spectrometer, l4N is observed at 7.2 MHz and on a 200
MHz proton spectrometer, l4N is observed at 14.4 MHz.
The primary standards used are nitromethane and the nitrate
+
ion for organic and aqueous solutions, respectively and tetrani- Fig. 7.45 31.65 MHz PMR signal of (CH3)4N ion in water.
tromethane and dimethylformamide are used in cases where
there are complications because of very small dN values (Table 7.12).
14
N chemical shifts of aliphatic nitro compounds, for aromatic nitro groups (from nitrate resonance); amides
(from nitrate resonance) and thioamides (from nitrate resonance) lie in the range −2.5−30 ppm, 6−20 ppm, 235−283
ppm and 210–239 ppm, respectively. In NH3, JN(14)) −H(1) = 43.6 Hz and JH−H = 10.35 Hz. In NH4NO3 in HNO3 on
successive deuteration, i.e. NH 4+ , NH3D + , NH 2 D 2+ , NHD
H 3+ ; JN(14)— H(1) = 52.7 Hz. Thus deuterium substitution
has little effect on 14N—1H coupling constant. However, a down-field shift of the proton resonance on successive
deuteration in ammonium nitrate–nitric acid system is observed and is attributed to electrostatic effects arising
Table 7.12 Internal standards for 14N measurements.
Standard dN (ppm) Application

MeNO2 0 Primary: For general use with organic solvents and concentrated mineral acids
−
3 0 Primary: For general use in aqueous solutions
C(NO2)4 + 48 Secondary: For nitro compounds
Me2NCHO + 276 Secondary: For aromatic nitro compounds and nitrogen heterocycles.
from hydrogen bonding with solvent molecules. The N15 nucleus (I = 1/2) resonates at 4.315 MHz in a magnetic
field of 10 kG. Since this isotope has a low magnetic moment (mI; = − 0.28304) and is present in only 0.365 per
cent of natural nitrogen, the sensitivity of 15N resonance signal in natural abundance samples is only about 1/2630
× 104 which in case of 14N nucleus is ≈1/2650 of that of an equal number of proton nuclei. However, 15N has no
quadrupole moment and s-s coupling is easily detected under suitable conditions. The l5N chemical shifts (in
ppm from nitrobenzene−15N) of aromatic nitro groups of nitrobenzene derivatives lie in the −3.85−4.38 region.
The observation of 15N coupling can be used for studying isomeric and tautomeric phenomenon which are very
common in organic nitrogen-containing compounds. The nitrogen proton coupling constants for the separately
observable cis and trans isomers of 15N-n-butylformamide and 15N−4−propylacetamide are shown in Fig(A).
1 2 1 2
H H H (CH2)3CH3 H3C H
15 15 15
N N N
O (CH2)3CH3 O H O (CH2)2CH3
J15N − H1 = 15.0 Hz J15N − H1 = 14.3 Hz J15N − H2 = 92.0 Hz
J15N − H2 = 92.2 Hz J15N − H2 = 89.8 Hz
The reaction of phenylisocyanate and acylmethylenetriphenylphosphoranes gives a product whose structure

may be represented by the tautomeric equilibrium. The observed 15N−H coupling constant J15 NH = 85−88 Hz
indicates that tautomeric form B predominates over A irrespective of solvent or temperature.
C6H5 15N C O + (C6H5)3P − CHCOR
H O
H O
C6H5 15N C R C6H5 15N C R
C C
C C
P+(C6H5)3 P+(C6H5)3
A B
7.39 APPLICATIONS OF NMR SPECTROSCOPY

The use of NMR spectroscopy makes it possible to investigate the structure of molecules, their conformation,
electron density delocalisation, weak intermolecular interactions, i.e. complex formation, solvation, hydrogen
bonding, hindered internal rotation, tautomeric equilibria, composition of multicomponent systems and some-
times also the kinetics of reactions. NMR technology is also widely recognised in the fields other than chemistry
such as biology, agriculture, engineering, medicine and industrial quality control. The NMR parameters which are
generally exploited to carry out these type of studies are chemical shift, integrated intensity or height of signal or
peak areas when the signals have significant width, coupling constant and bandwidth.
7.39.1 Mole Fraction of a Non-ideal Solution from NMR Spectrum

The nature of NMR is such that for a homogeneous mixture, one obtains a spectrum whose integration peaks are
proportional to the number of protons present at a given field producing that signal. So, if a particular proton in a mol-
ecule can be associated qualitatively with a given peak and then the area of that peak for each molecule can be related
by an exact ratio to each other, the mole ratios as well as mole fractions of the components may be determined.
Let A and B be the two components of a homogeneous solution. To determine the mole fraction of A and B from
a spectrum, the areas of specific protons of A and B are measured. These areas are then normalised by the number
of protons resonating in each material. Thus,
Ia
Ia = (7.121)
na
Ib
Ib = (7.122)
nb
where na and nb are the number of protons associated with the signals having areas Ia and Ib respectively. Barred
quantities indicate the areas per proton. The mole fractions of A and B are then given by
Ia (7.123)
Xa =
Ia Ib
Ib
Xb = =1− X a (7.124)
Ib Ib
e.g. in case of chloroform–acetone system, the normalised areas of the peaks corresponding to protons of chloro-
form and acetone are given by
I ch
I ch =
1
I ac
I ac =
6
Here, the subscripts ch and ac stand for chloroform and acetone respectively.
This method is equally applicable to a three-component system, i.e. chloroform–acetic acid–H2O. The inte-
gration peaks used are the methyl group for acetic acid (3 protons), the single hydrogen from chloroform and
the hydrogens of water minus the one hydrogen from the O—H group of acetic acid which falls at the same
chemical shift as H2O, since the protons are in rapid exchange between the two molecules. Consequently, the
normalised areas of the signals corresponding to protons of chloroform (ch) acetic acid (HOAC) and H2O are
given by
I ch
I ch =
1
I
I HOAC = HOAC
I H2 O (2 water protons + HOAC acidic proton) I HOAC

I H2O =
2 prottons/molecule of water
7.39.2 Analysis of Mixtures

The NMR spectra may be used for the quantitative analysis of mixtures provided the applied magnetic field is
homogeneous. The residual hydrogen in D2O may be determined as follows: standard samples containing known
amounts of H2O and D2O are prepared. A calibration curve between intensity of proton signal and the amount of
water in the standard sample is plotted. The intensity of the proton signal in the unknown sample of (H2O + D2O)
is observed. The amount of water corresponding to this intensity can then be directly read from the calibrated
graph.
The NMR spectra is also used for the analysis of multicomponent mixtures, e.g. a mixture of cyclohexane
(C6H12), dimethylacetone ((CH3)2CO), 1,1,1-trichloroethane (CH3CC13), dioxane (C4H8O2), dichloromethane
(CH2C12) and chloroform (CHC13) has been analysed by this technique. The amount of each component in the
mixture is proportional to the number of equivalent protons contained in their molecules. So from the ratios of
the normalised integrated intensities of all the signals, the concentration of each component can be determined.
The NMR spectrum of the above-said mixture is presented in Fig. 7.46. Since the intensities of all the signals are
identical, the concentrations of the components in the mixture are, therefore, inversely proportional to the number
of equivalent protons contained in the molecules, i.e.
C6H12: C3H6O: CH3CCl3: C4H8O2: CH2Cl2: CHCl3:: 2 : 4 : 8 : 3 : 12 : 24,
Thus, the amounts of various components in the mixture are C6H12 = 3.77; C3H6O = 7.54; CH3CC13 = 15.09;
C4H8O2 = 5.66; CH2C12 = 22.64 and CHC13 = 45.28 per cent.
(CH3)2CO
(CH3)4Si
CH3OCl3
C4H8O2
CH2Cl2
CHCl3
(CH2)6
8 7 6 5 4 3 2 1 0
δ (in ppm)
Fig. 7.46 1
H-NMR spectrum of a mixture of compounds.
7.39.3 Study of Tautomeric Equilibria

The amount of tautomeric forms in any molecular system can be determined from the normalised integrated
intensities of the NMR active nuclei responsible for tautomerism. The amount of keto and enol forms in a system
exhibiting tautomeric equilibrium can be determined from the intensities of the olefinic proton (—CH) of enol
form and of methylenic protons (—CH2) of keto form. If ICH is the intensity of —CH proton and I CH2 that of
—CH2 protons, then
I CH
Percentage of enol form = × 100 (7.125)
I CH + I CH 2
I CH2
Percentage of keto form = × 100 (7.126)
I CH + I CH 2
I CH I
Here, I CH = and I CH2 = CH2 are the normalised integrated intensities.
1 2
Problem 7.20: Determine the amount of keto and enol forms in acetylacetone if the integrated intensities of
olefinic and methylenic protons are 5.2 and 2 units respectively.
Solution The tautomeric equilibrium in acetylacetone is
O O OH O
CH3 C CH2 C CH3 CH3 C CH C CH3
We know what, I CH2 = 2 units and ICH = 5.2 units.

Using Eqs (7.125) and (7.126) we obtain
52
Percentage of enol form = 1 × 100 = 83.3
52 2
+
1 2
2
Percentage of keto form = 2 = 16.2
52 2
+
1 2
Keep in mind that this type of equilibrium is a function of temperature and solvent.
7.39.4 Determination of Hydration Number

None of the methods based on conductance, electromotive force or transference number provides a direct measure
of the coordination number of an ion. Protons in water molecules occupying the first hydration sphere exhibit a
proton magnetic resonance signal discernible from bulk water signal because of their different magnetic environ-
ments. At low temperatures (generally ≤ −20°C), the exchange of both protons and H2O molecules is slowed to
the extent that these two peaks may be resolved. The ratio of their areas then gives a direct measure of the primary
hydration number.
Hydration number =
Area of the coordinated peak
×
[ H 2 O] (7.127)
Area of coordinated peak + area
e of bulk peak Moles of salt
Here, [H2O] = 55.5 moles.

Incomplete resolution owing to overlap between bulk and complexed water peaks may pose a problem for
accurate determination of areas. Complete resolution of peaks is hindered because of freezing, precipitation as
well as increase in viscosity of solutions at low temperatures. Freezing may be prevented by using solvent systems
containing low-freezing-point organic solvents or by use of solutions of high ionic strength. Solvent exchange in
non-aqueous solutions is also slow enough to permit the observation of signals resulting from the protons of bulk
molecules and molecules in the first coordination sphere of anions. All these factors cause large line broadening
of water signals. The much greater line width of the signal of complexed water molecules may be because of s-s
or quadrupole interaction with the metal ion.
The dependence of NMR signal shape on rate processes occurring in solutions may be demonstrated by spectra
recorded at different temperatures. At room temperature when only one water proton signal is noticed, the lifetime
1
of proton at a particular site is given by τ = . When the separate signals are observed (at low temperature),
2πΔv
the proton (not water) exchange has been slowed down to such an extant that t increases by 103.
Problem 7.21: The PMR spectrum of 1.75 M AlCl3 plus 1.65 M Ca (NO3)2 at −38°C on a 60 MHz instrument
exhibits two bands corresponding to coordinated and bulk water molecules.The coordinated water peak is to the
left (upfield = 124 Hz) of the bulk water peak. If the areas of the coordinated and bulk water peaks are 84 and 600
units respectively, determine the coordination number of Al(III).
Solution Concentration of A1C13 = 1.75 M

[H2O] = 55.55 moles
Area of coordinated water peak = 84 units
Area of the uncomplexed water peak = 600 units
Substituting these values into Eq. (7.127),
84 55.55
Hydration number = × = 3 88
600 + 84 1 75
The low value of hydration number of Al (III) in the presence of CI− ions may be taken as evidence for contact
ion pairing.
7.39.5 Determination of Optical Purity

A mixture of equal quantities of two enantiomers E + and E_ called racemic mixture does not rotate the plane of
polarised light. However, another mixture of enantiomers will exhibit optical rotation. The percent optical purity
(P) of a mixture of enantiomers is defined as
P (%) =
[α ] ×100 (7.128)
[ ]
where [a] is the specific rotation of a mixture of enantiomers and [A], is the specific rotation of the pure enantiomer
and is also called the absolute rotation. The value of the per cent optical purity is equal to the per cent enantio-
metric purity; which is a measure of the excess of one enantiomer over another,
P (percent) = enantiometric purity (percent)
E+ − E _ (7 129)
= × 100
E+ + E _
= (2E + − 1) (7.130)
where, E+ and E_ are the mole fractions of the enantiomers. E+ is assumed to be the predominant isomer. The
optical purity may be determined by polarimetry provided the absolute rotation of the pure enantiomer is known,
which is always not possible.
In order to resolve the racemic mixture, it is treated with a second chiral substance Z + , to produce a mixture of
diastereomers Y + Z + and Y_Z + which have different physical properties, e.g. melting point, solubility, infrared and
NMR spectra. The NMR method of optical purity is based on the difference in properties of diastereomers. The
two enantiomers Y + and Y_ are completely converted into Y + Z + and Y_Z + by reaction with an excess of Z + . Under
this condition, the ratio of mole fractions of enantiomers reacted equals the ratio of diastereomers formed, i.e
E + EY + Z +
= =R (7.131)
E _ EY _ Z +
R −1
P( )= × 100 (7.132)
R +1
Thus, if R (the ratio of enantiomers) is known, the optical purity of the original mixture Y + and Y_ can be calculated,
R is taken as the ratio of intensities of particular NMR signals corresponding to diastereomers formed from enantiom-
ers, e.g. the optical purity of a partially resolved sample of a-phenylethylamine (A + B) is determined by converting,
H H
C6H5 C NH2 H2N C C6H5
CH3 CH3
(A) (B)
into a mixture of diastereomeric amides (A′+ B′) by reaction of optically pure o-methyl-mandelyl chloride.
H O H CH3 O H
C6H5 C NH C C C6H5 C6H5 C NH C C C6H5
CH3 OCH3 H OCH3

A⬘ B⬘
The two C-methyl groups (as well as the other corresponding groups) in A and B though have same chemical
shifts because of enantiomeric environments but have different chemical shifts in A′ and B′ due to diastereomeric
environments. Thus from the ratio of integrated intensities (R) of any of the sets of signals in the PMR spectrum
of a mixture of diastereomers, the optical purity of enantiomers can be determined from Eq. (7.132).
The NMR method for the determination of optical activity is not only limited to amines but is also applicable
to alcohols, sulphinates, phosphinates and sulphinamides.
7.39.6 Determination of Molecular Mass

Molecular mass can be determined from the integration of proton signals. A certain integral is equated with the amount
of protons by adding a known weight of standard in the NMR sample of the unknown. If Istd = integral of the standard,
Wstd = mass of standard, Mstd = molecular mass of standard, and nstd number of protons in the standard peak, then
W st dn
nstd
I std = protons (7.133)
M std
W std nstd
Therefore, I= .I (7.134)
M std I std
where, I is the integral of unknown signal from say n protons.
W std nstd
The ⋅ I proton signal is because of W of unknown. Therefore, n protons correspond with
M std I std
nW M std I std
= molecular mass of unknown (7.135)
W std nstd
The standard used should not overlap any signals of the unknown and hexmethylcyclotrisiloxane [(CH3)2SiO]3
is recommended. It is a very soluble inert solid with a high field chemical shift a few cycles down-field from
TMS. This method does not appear to have received much application in practice. Molecular mass of poly-
mers may be determined by PMR spectroscopy. The OH resonance of the polyethylene glycol carbowax 600
[HO(CH2CH2O)nH] may be integrated separately from the —OCH2CH2—O signal and the number average
molecular mass (NAMM) is calculated as follows.
Let I(OH) is the OH integral and I(OCH2CH2O) is the OCH2CH2O integral.
I(OH) integral ≡ 2 protons
2I ( )
Therefore, I( )≡ protons
I( )
Since I(OCH2CH2O) ≡4 protons ≡ mass 44 (CH2CH2O).
2I ( ) mass 22I (OCH 2 CH 2 O)

Therefore, protons =
I( ) I (OH)
2I ( )
i.e. NAMM ≡ + 18
I( )
The PMR value of 422 is much lower than the cryoscopic value of 598. This method is not very reliable since
it depends on the accurate integration of a very small area I(OH) relative to a large area I(OCH2CH2O) and upon
the absence of water which may contribute to the OH signal. So it is better to exploit the terminal methylene
signal (HO—CH2). This is not resolvable in the CDC13 spectrum. The PMR spectrum is recorded in pyridine—
CF3COOH containing a trace of hydrogen chloride to decouple HO from the adjacent CH2 signal on a 60 MHz
instrument. Polyethylene glycol 400 is HOCH2 (CH2OCH2)nCH2OH.
Let I(HOCH2) is the signal of CH2 protons of HOCH2 · I(HOCH2) ≡ 4 protons
If I (CH2OCH2) is integral of CH2OCH2 protons.
4I (CH 2 OCH 2 )
Then, I (CH 2 OCH )2 ≡ protons.
I (OHCH 2 )
Now I (CH 2 OCH 2 ) 4 protons ≡ mass of 44 (CH 2 OCH 2 )
Therefore,
4 (CH 2 OC ) protons ≡ mass 44I (CH 2OCH 2 )
2
I ( HOCH 2 ) I ( HOCH 2 )
44I (CH 2 OCH 2 )

i.e. NAMM ≡ + 62
I ( HOCH 2 )
where 2 HOC 2 62
The molecular mass determined by this method is 592.
7.39.7 Degree of Unsaturation in Natural Fats

A general formula for natural fats may be written as follows.
CH2OCO(CH2)a (CH = CH)x Me
CH OCO(CH2)b (CH = CH)y Me
CH2OCO(CH2)c (CH = CH)z Me
The molecular mass corresponding to this formula is

MM = 173 + 45 + 14(a + b + c) + 26(x + y + z) (7.136)
where, 173 is the formula mass of the C6H5O6 glyceryl triester radical, while 45 is the mass of three terminal
methyl groups; 14 is the mass of CH2 group and 26 is the mass of CH = CH group. The total number of protons
(T) is
T = 5 + 9 + 2(a + b + c) + 2(x + y + z) (7.137)
The number of vinylic protons is
V = 2(x + y + z) (7.138)
Lumping together Eqs (7.137) and (7.138), we obtain
( )
( ) = (7.139)
2
Thus, Eq. (7.136) may be expressed in terms of T and V as,
MM = 218 + 7 (T − V − 14) + 13V (7.140)
= 120 + 7T + 6 V
Knowing T and V from integral data, the molecular mass of the natural fats can be determined from Eq. (7.140).
Once MM is known, the iodine value may be calculated from the expression
126.9
Iodine value = × number of vinylic protons × 100
MM
12690V
= (7.141)
MM
Using this method, the iodine number, the molecular mass and the number of vinylic protons in a sample of
pure trilinolein have been found to be 173.3, 879 and 6.00 respectively.
7.39.8 Study of Conformational Equilibria

The conformational studies by NMR technique are carried out using 1H, l3C and other nuclei. Consider a system
exhibiting conformational equilibria, i.e.
K
Aa Ae (7.142a)
K
Ae Aa (7.142b)
where suffixes a and e stand for axial and equatorial, respectively.
K=
[A e ] (7.143)
[A a ]
By making use of Eq. (7.112), i.e
δ δ + P2δ 2 ,
1 1 (7.144)
we write δ e δ e + Paδ a
Here, da and de are the chemical shifts of the NMR active nuclei responsible for the conformational equilibria.
d is the observed chemical shift of the identical NMR active nuclei rapidly equilibriating in the parent structure
of the system exhibiting conformational equilibria.
Since, Pa + Pe = 1,
δa − δ
∴ Pe = ; (7.145)
δa − δe
δ − δe
Pa = (7.146)
δa − δe
Pe δ a − δ
Ka e = = (7.147)
Pa δ − δ e
Pa δ − δ e
Ke a = = (7.148)
Pe δ a − δ
Thus, by monitoring the chemical shift of the system exhibiting conformational equilibria as a function of
temperature, the value of K and hence ΔG at a particular temperature can be evaluated. The best approach to deter-
mine the conformational equilibrium constants is the examination of an equilibriating system at sufficiently low
temperatures so that the rate of conformational inversion is slow on the NMR time scale. Consequently, resonance
lines for each individual conformer will appear whose relative intensities will provide a direct measure of K at that
very temperature. The results of conformational equilibria based on PMR spectroscopy may be supplemented by
13
C-NMR technique. However, the determination of K from l3C-chemical shift data using Eqs (7.147) and (7.148)
depends upon the availability of reliable values of da and de. In case it is difficult to obtain accurate values of da
and de, the latter approach based on intensities of the conformers is appreciated.
Problem 7.22: The NMR data on the conformational behaviour of 3- and 4-t-butyl-substituted cyclohexanes
are listed in the tabular form as under. Determine the equilibrium constants of the systems.
C1-proton signals in 3- and 4-, t-butyl substituted cyclohexanes.
C1 substituted −vCHX
a
H (in cps)
X Axial H Equatorial H
trans−4 cis−3 cis−4 trans−3
OH 202.0 206.0 236.0 244.0
F 281.1≠ 282.0≠ 304.6≠ 312.5≠
Cl 221.5 225.5 262.5 269.8
J = 49.4 cps in all cases, a-down field from TMS in CC14, ≠ lower field signal of fluorine split doublet
Solution The general conformational equilibrium for these systems is
H X
X Ka − e
H
(CH3)3C (CH3)3C
νe νa
By Eq. (7.148),
v ve
Ke a = (7.149)
va v
Here, v is the observed chemical shift for —CHX proton in the rapidly equilibrating cyclohexyl system For
hexanol system, v = –211 cps is the shift of—CHX proton in cyclohexyl chloride. The observed shift v = –291.6
cps to be used in fluoride system is the lower field component of the CHF doublet. In case of chloride system,
v = –234.4 cps is the chemical shift of —CHX proton in the cyclohexyl chloride. Thus,
−211 + 236 25
K OH ( t) = = = 2.77
−202 + 211 9
−211 + 244 33
K OH ( t) = = = 6.6
−206 + 211 5
For other systems, the calculated values of K’s are
KF (4–t) = 1.25; KCl (4–t) = 2.18
KF (3–t) = 2.18; KCl (3–t) = 3.98
7.39.9 Study of H-bonded Equilibria

Proton-exchange processes are affected by temperature and are accelerated in the presence of strong acids and
bases. The time of spin orientation in the magnetic field or the time required for the resonance measurement
(≈ 10 −22 − 10 3 second) is long compared to the lifetimes of H-bonded moieties (10 −12 − 10 −13 second). This is
because of very rapid formation and breaking of hydrogen bonds in solution. Thus, the H of an AH group will
participate in many H-bonds with the neighbouring molecules in a time of the order of 10 −3 second. Hence,
the H of an AH will experience at each concentration an averaged out environment which will depend on the
number and kind of proton donor species present. This fact decreases the usefulness of PMR as a tool for
studying the H-bonded dimers and n-mers and their equilibria in the case of self-associated H-bonded sys-
tems. However, the PMR technique proved to be very useful tool for the investigation of H-bonded 1: 1 com-
plex because there is only one type of proton donor H-bonded to acceptor molecules in such systems. In view
of it, PMR differs from infrared spectroscopy in that only one PMR signal is observed for protons involving
H-bonding. This signal is a weighted average of all protons in their environments, in the free as well as in the
associated forms.
The measured chemical shift of such a nucleus is,
δ = ∑ iδi
where, Pi is the population of the nucleus in the ith environment relative to the total population of the nucleus,
i.e. ∑ i = 1, and di is the chemical shift of the nucleus purely in the ith environment. In the simple case of 1 : 1
complexes formed between donor and acceptor in which the rates of the forward and backward reactions are very
fast, the measured chemical shift of nucleus in the acceptor or donor moiety will be
δ A δ A + PDA δ DA (in acceptor moiety) (7.150)
δ D δ D + PDA δ DA (in donor moiety)
Here, PA /PD is the fraction of uncomplexed acceptor/donor molecules and PDA is the fraction of the acceptor/
donor molecules complexed. Since PA + PDA or PD + PDA = 1, the expression (7.150) may be written as
δ DA (δ DA − δ A ) δA (7.151)
δ DA (δ DA − δ D ) δD
This equation can be used to evaluate the association constants of various types of complexes, viz., H-bonded
and charge transfer.
(a) Case(i): Extent of Molecular Association due to H-Bonding Let us consider a system in which there
is a rapid exchange of proton between free and H-bonded protons. The observed chemical shift d of proton in
such a system is
(δ1 1 δ 2C 2 + 3δ 3 3 δ nC n )
δ= (7.152)
(C1 + 2C 2 + 3C 3 + + nC n )
Here, C1 C2, C3, ..., Cn are the concentrations of monomer, dimer, trimer, ..., n-mer moieties respectively. If
K1 K2, K3 ,..., Kn are the formation constants of the H-bonded species respectively, then we may write
Cn
Kn = (7.153)
C1n
δ=
(δ δ + + δ ) (7.154)
( + + + )
The direct solution of Eq. (7.154) from experimental measurements of d at several concentrations is not pos-
sible in H-bonded systems involving several equilibria. Therefore, in order to determine the extent of association,
assume that only monomer and n-mer are present in the system. Under this assumption, Eq. (7.154) reduces to
δ=
(δ δ ) (7.155)
( + )
Let the stoichiometric (total) concentration, C (C + nK C ). n
n
Then δ=
(δ δ ) (7.156)
C
An experimental curve is drawn between d and C. A comparison is then made between the observed curve and
the trial and error computed curves until the best fit is found. In case of o-cresol it is found that the theoretical
curve (S-shaped) for n = 3 matches with the observed curve between d of hydroxyl proton and log C (S-shaped).
Thus in o-cresol, the major H-bonded moiety present is trimer.
(b) Case (ii): Formation Constant of 1 : 1 H-bonded/charge Transfer Complex Let us consider the
equilibria
K
A−H+B A − H --- B
(7.157)
K
D A D---A
or K=
[ − ]= (C C )
[ − ][ ] C AH ⎡⎣C B0 − (C C )⎤⎦
(a) If C B0 C A0 H
then
K=
(C C ) (7.158)
[C ] ⎡⎣C B0 ⎤⎦
0
Here, C AH and C B0 are the initial concentrations of A—H and the acceptor B respectively, and CAH is the
equilibrium concentration of free proton donor AH. The weighted average observed chemical shift d in such
a case is
⎛ C AH ⎞ ⎛ C0 −C ⎞
δ δ1 + ⎜ AH 0 AH ⎟ δ 2 (7.159)
⎝ C AH ⎠
0
⎝ C AH ⎠
Here d1 is the chemical shift of the monomer proton donor A–H and d2 is the chemical shift of the A–H
proton in the H-bonded complex A–H …B.
From Eqs (7.158) and (7.159), the following relationship is obtained:

δ δ1
= δ 2 δ = (δ 2 δ1 ) − (δ δ1 )
KC B0
⎡ 1 ⎤
(δ δ1 ) ⎢ 0
+ 1⎥ = δ 2 δ1
⎣ KC B ⎦
1 1 1
= + (7.160)
δ δ1 KC (δ 2
0
B δ1 ) δ 2 δ1
1 1 1
= + (7.161)
Δ KC B Δ 0 Δ 0
0
where, Δ = δ − δ1 and Δ 0 = δ 2 − δ1 .
The value of d1 is obtained by extrapolation of experimental results to infinite dilution.
0
(b) In case C AH C B0 , Eq. (7.161) will become
1 1 1
= + (7.162)
Δ Δ 0 KC AH
0
Δ0
Reference Signal
δA
Δ
δ
ΔAD
0
δAD
Fig. 7.47 The relationship between Δ , Δ 0A D , δ , δ A and dAD in a donor (D)-acceptor (A) system.
The relationship between various forms of d’s in donor (D)-acceptor (A) system is shown in Fig. 7.47.
1 1 1
Thus, a plot between and 0 will be a straight line with slope m = 1 / K Δ 0 and intercept = 1/Δ0. From
Δ C AH C B0
the slope and intercept, both K and Δ can be determined. The behaviour of K as a function of temperature may
be employed to evaluate the thermodynamic parameters of the system. The values of association constant (K) for
some charge-transfer complexes from PMR and 19F chemical shifts measurements determined under the condi-
tions [ D ]0 [A ]0 and [D0 ] A 0 (all at 33.5°C) are listed in Table 7.13 while in Table 7.14 are the values of K
of some 1:1 H-bonded complexes.
(c) Case (iii): Formation Constant of 1:2 H-bonded Complex Let us consider the equilibria of the type
K
HAH − .... − B + B B − .... − HAH − .... − B (7.163)
K=
[ ] (7.164)
[ ][ ]
Here, all HAH is H-bonded, HAH-----B is the 1:1 H-bonded complex formed between proton donor HAH and
acceptor B, B-----HAH-----B is the 1:2 H-bonded complex formed from 1:1 H-bonded complex and acceptor B.
Table 7.13 Values of association constant of some charge-transfer complexes determined by NMR technique
under different experimental conditions (at 33.5°C).
System Relative Nuclei Measured K(kg mol−1 )

concentration
1,3,5−trinitrobenzene−mesitylene in [A]0>>[D]0 H2,4,6 in D 0.2
1,2–dichloroethane [A]0<<[D]0 H2,4,6 in A 0.2
3,5−dinitrobenzene−trifloride− H1,6 1.73
[A]0>>[D]0
hexamethylbenzene in CCl4 H4 in A 1.76
19
F 1.19
1,3,5−trinitrobenzene−indole in [D]0 >>[A0] H3 in D 0.9
CH2ClCH2Cl H*aromatic 1.0
4−nitrobenzaldehyde−hexamethyl [A]0>>[D]0 H3,5 in A 0.5
benzene in CCl4 H2,6 in A 0.7
HCHO 0.9
N,N−dimethylanaline− [A]0<<[D]0 0.4†
[ ]0 [ ]0 ⎫ 1.0†
⎪⎪
1,3,5−trinitrobenzene in CHCl3 [ ]0 [ ]0 ⎬ optical 0.6†
⎪ 0.4†
[ ]0 [ ]0 ⎪⎭
* Band of maximum intensity in the aromatic multiplet Component containing measured nuclei, i.e. D or A.
† in lit mole−1,
Table 7.14 Values of K (at 33°C) for some 1:1 H-bonded complexes.
System K(mole fraction)−1

Pyrrole−dimethyl sulphoxide(N−H…..O) 8.69
Pyrrole−pyridine(N−H…N) 23.1
F13C6−CF2−H−(CH3)2CO (C−H…O)* 2.0
F13C6−CF2−H−(C2H5)3N (C−H…N)* 1.5
F13C6−CF2−H−(C2H5)2 O (C−H…O)* 0.6
*At 30°C
Let d be the chemical shift of HAH proton in the system in a certain range of mole fractions nHAH and B of mole
fraction nB in some suitable solvent. Let the value of d extrapolated to zero HAH donor concentration be d0. If d11
⎛ n ⎞ ⎛ n ⎞
and d12 are the proton chemical shifts in 1:1 and 1:2 H-bonded complexes respectively and ⎜ 11 ⎟ and ⎜ 12 ⎟
⎝ n HAH ⎠ 0 ⎝ n HAH ⎠ 0
are the fractions of HAH present as the 1:1 and 1:2 in the limit of zero HAH donor concentrations respectively
then from Eq. (7.112), i.e. d = P1d1 + P2d2, we write
⎛ n11 ⎞ ⎛ n ⎞
δ0 δ11 + ⎜ 12 ⎟ δ12 (7.165)
⎝ n HAH ⎠ 0 ⎝ n HAH ⎠ 0
⎛ n11 ⎞ ⎛ n12 ⎞
Also, we know that ⎜⎝ n ⎟⎠ + ⎜⎝ n ⎟⎠ = 1 (7.166)
HAH 0 HAH 0

1 ⎛n ⎞ ⎛ 1 1⎞
= ⎜ HAH ⎟ ⎜ − ⎟ (7.167)
δ0 δ11 ⎝ n12 ⎠ 0 ⎝ δ12 δ11 ⎠
Eq. (7.164) may be expressed in terms of concentrations as

⎛ n12 ⎞
⎜⎝ n ⎟⎠
HAH 0
K= (7.168)
⎛ n11 ⎞ ⎡ ⎛ n12 ⎞ ⎤
⎜⎝ n ⎟⎠ ⎢ n B − ⎜⎝ n ⎟⎠ ⎥
HAH 0 ⎣ HAH 0 ⎦
⎛ n12 ⎞
⎜⎝ n ⎟⎠
HAH 0
K= (7.169)
⎛ n12 ⎞ ⎡ ⎛ n12 ⎞ ⎤
⎜⎝1 − n ⎟⎠ ⎢ n B − ⎜⎝ n ⎟⎠ ⎥
HAH 0 ⎣ HAH 0 ⎦
2
⎛ n ⎞
The square term ⎜ 12 ⎟ in Eq. (7.169) being small, as compared to the other terms in the equation can be
⎝ n HAH ⎠ 0
ignored. Hence Eq. (7.169) becomes
⎛ n12 ⎞
⎜⎝ n ⎟⎠
HAH 0
K= (7.170)
⎡ ⎛ n12 ⎞ ⎛ n ⎞ ⎤
⎢ n B − ⎜⎝ ⎟ − n B ⎜ 12 ⎟ ⎥
⎣ n HAH ⎠ 0 ⎝ n HAH ⎠ 0 ⎦
Lumping together Eqs (7.167) and (7.170), we obtain
1 1 1
= + (7.171)
(δ 0 δ11 ) Kn (δ12 δ11 ) ( 12 δ11 )
δ
1
Let (δ12 δ11 ) be independent of temperature. Trial values of d11 are selected and then plots of against
1
(δ 0 δ11 )
will be a straight line with slope m = 1/K( δ12 δ11 ) and intercept = l/( δ12 δ11 ). From the slope and intercept
nB
of various lines, different values of K and intercept corresponding to different trial values of d11 will be obtained.
The set of trial d11 values which gives almost identical values of K is taken as the true value of d11. Based on this
true value of d11 then K and d12 are evaluated from the curve. Thermodynamic functions may be determined from
the effect of temperature on equilibrium constant.
7.39.10 Dissociation Constant of an Acid

The dissociation constant of an acid may be determined from proton chemical shifts as follows.
Let us consider that an acid HA dissociates into H + and A− ions, i.e.
K
HA H + + A− (7.172)
Here, the dissociation constant,

[A − ]
[K] = [H + ] (1.173)
[HA]
Let dobs, d HA and d A− are the observed chemical shift at a given pH, the chemical shift of the protonated species,
and the chemical shift of the unprotonated moiety, respectively.
By Eq. (7.112),
d = P1 dHA + P2 dA
Since, P1 + P2 = 1
δ − δ HA
Therefore, P2 = (7.174)
δ A − − δ HA
Comparing Eqs (7.173) and (7.174), we obtain

⎡ δ − δ HA ⎤
K [H + ] ⎢ ⎥
⎣ δ A − δ HHA ⎦
δ δ HA
or log = pH − pK a (7.175)
δ A δ HA
Trial values of δ A − are selected and then from the plot of log ⎡⎣(δ δ ) / (δ δ )⎤⎦ against pH, the value of
pKa can be estimated. pKa can also be obtained from the plot of δ (in Hz) versus pH. The shape of the curve is
analogous to the potentiometric curve. This method can be extended for the determination of microscopic ionisa-
tion constants of various systems, e.g. amino acids and peptides. 13C and 31P chemical shifts are also employed
to determine pKa of nicotinamideadeninedinucleotide (NAD) and NAD phosphate (NADP). pKa of imadizole
obtained from chemical shifts of 13C-2 and l3C-4 have been found to be of the order of 7.2 and 7.15 respectively.
7.39.11 Structural Elucidation

The NMR spectroscopy alone or in conjunction with IR technique proved to be a very powerful tool for establish-
ing the structure of molecules. The magnitude of J and pattern of s-s interaction are useful parameters for geo-
metrical identification as is evident from the following examples. 31P-NMR spectrum of P4S3 consists of two peaks
(separation between two peaks is 185 ppm. J = 70 Hz) with intensity ratio 3 : 1 the more intense peak appears as
a doublet (1 : 1) while the less intense as a quadruplet (1 : 3 : 3 : 1) to the right of the doublet. Since I = 0 for 32S,
both the s-s splitting and relative intensities of the signal reveal that there are three equivalent phosphorous atoms
and one unique phosphorus atom in P4S3. Consequently, the following structure is proposed for P4S3.
P
S S
S
P P
P
P-NMR spectra of phosphorous, HPO(OH)2(I) and hypophosphorous, H2 PO(OH)(II) acids exhibita doublet
31
(1 : 1) and a triplet (1 : 2 : 1) respectively. The s-s splitting pattern indicates that the phosphorous atom interacts
with protons in both the acids. Consequently, the following structures of the acids conform to the 31P-NMR spec-
tral data.
O O
P P
H OH H OH
OH H
(i) (ii)
The simple nature of the spectra of both the acids is due to the following reasons: (i) the interaction between
OH protons and phosphorous is too weak to be resolved since OH protons are at a distant apart from 31P, and (ii)
the OH protons do not couple with 3IP nucleus because of fast exchange of proton with the other acid molecule
through H-bonding.
PMR spectrum of 1,3,5-trimethylcyclotriborazane (MeNHBH2)3 exhibits two types of doublets due to methyl
protons, i.e. a single doublet and two such double doublets with relative intensity ratio 1 : 2 and J = 5 Hz. The
doublet pattern may arise from the interaction of NH protons with the protons of the methyl group. The nature of
two types of doublets suggests that the compound exists in two isomeric forms. By analogy with cyclohexanes,
the most stable isomer should be a chair with all the methyl groups in the equatorial position (I) whereas the other
less stable isomer should have one axial and two equatorial methyl groups (II) as shown here.
CH3 BH2 CH3 CH3 BH2 CH3
N N N N
H H H H
H2B BH2 H 2B BH2
N N
H H
CH3 CH3
(I) (II)
The barrier to internal conversion is large and both the structures are rigid (by NMR criterion) as compared to
cyclohexanes.
The structure of two products of photochemical reaction given below is proposed simply from JFF.
Br Br F
hn
CFBr3 + N2F4 C N (JFF = 17.6 Hz) + C N (JFF = 218 Hz)
F F F
These are rigid structures by NMR criterion.

Let us now examine how NMR in conjunction with infrared spectra could be useful in structural elucidation
of molecules.
Molecular formula: C3H7O2N.
NMR-triplet (d = 1.03 ppm; 1 : 2 : 1), triplet (d = 4.38 ppm; 1 : 2 : 1),
sextet (d = 2.07 ppm; 1 : 5 : 10 : 10 : 5 : 1)
−1
IR: 1560 (s), 1350 cm −1 (s).
The chemical shifts and splitting pattern of the NMR signals tentatively conforms to the skeleton structure
H3C⎯CH2⎯CH2⎯.
a b c
By empirical rules regarding the multiplicity of the NMR signal, the signal due to ‘b’ protons should appear
as a twelve (3 + 1)(2 + 1) = 12 lines multiplet. However, the sextet nature of the ‘b’ protons arises because the
s-s coupling between protons at ‘a’ and ‘b’ and ‘b’ and ‘c’ centres is almost the same, i.e. Jab Jbc. Actually the
PMR spectrum of the compound when monitored on a moderate resolution NMR instrument shows a sextet pat-
tern for the ‘b’ protons. Further, the two strong infrared bands at 1560 and 1350 cm−1.correspond to the asym-
metric and symmetric stretching vibrations of NO2 group. Thus the structure of the compound having molecular
formula C3H7O2N is CH3—CH2—CH2—NO2.
The structure of complex molecular systems can be established from infrared group frequencies and from
chemical shifts and coupling constants by comparison methods. The IR group frequencies, chemical shifts and
coupling constants of the system under study are matched with their respective literature values assigned to the
standard structures.
NMR Evidence for Alkali Metal Anions The possibility that an alkali metal can exist as an anion has been
explored as early as 1864 by W Weyl in Germany. The first-ever known anion of sodium, i.e. sodium ion Na − in ( )
solid state was synthesised in 1974. The ‘cryptand’ 2, 2, 2 crypt [4, 7, 13, 16, 21, 25-hexaoxa-l, 10, diazabicyclic
(8, 8, 8) hexacosane], was dissolved in ethylamine and allowed to react with metallic sodium. The blue colour
formed was rapidly evaporated to leave behind a gold-coloured compound at −80°C. The experiment was car-
ried out in moisture-free atmosphere and at a pressure of 10−5 Torr. The composition of the compound so formed
is [(Na + C)(Na−)] where C is the cryptand. The nature of both Na + and Na− in the solutions has been established
by 23Na-NMR technique. The crowned Na + shows a resonance at 10.4 ppm with a large line width while Na −
resonance occurs at 62.8 ppm with a narrow line width. The presence of well-separated signals shows that there is
no exchange between Na + and Na − and the solvent is excluded from their immedidate spin-paired vicinity. 23Na-
NMR results point out to the fact that the added electron in Na − resides in 3s orbital Spin pared.
The molecular structures determined by NMR are essentially motionally averaged. Therefore, it is important
to note the motional properties of the molecule. The range of correlation times (τ c ) that can be measured by
NMR is from picoseconds to greater than a second. While real-time monitoring after the pulse is possible for
τ c > 1 second, T1 can be measured for cases where tc is between 30 ps and 1ms. The 2D-EXSY methods measure
tc values between 10 ms – 1s.
7.39.12 Kinetic Studies

If two groups of chemically equivalent nuclei, say A and B, are exchanged by an intramolecular process, the NMR
spectrum is a function of the difference in their resonance frequencies, v A v B = Δv, and of the rate of exchange,
k (a typical value of Δv is about 10 cps). The effects of exchange at several temperatures on the width of NMR
signals at vA and vB are exhibited in Figs 7.48(a) to (d). At low temperatures, the exchange is slow and k << Δv. The
spectrum thus consists of two sharp singlets at vA and vB [Fig. 7.48(a)]. At high temperatures, the exchange is fast
and k >> Δv and a single sharp peak is observed [Fig. 7.48(d)]. There is also an intermediate temperature range
Δν0 = νA − νB
(a) Slow exchange

(ΔνA)01/2 (ΔνB)01/2
(b) Intermediate Δνe
(ΔνA)e1/2
Imin Imax
(c) Coalescence
(d) Fast exchange
(Δν1/2)e
νA + νB
2
Fig. 7.48 Effect of exchange of chemically equivalent nuclei on NMR line shapes.
over which the spectrum consists of two significantly broadened overlapping lines [Fig. 7.48(b)].
We know that usually the s-s and s-l relaxations determine the width of an NMR absorption line. However,
the exchange of two groups of chemically equivalent nuclei also causes additional line broadening of the NMR
absorption line.
According to Heisenberg uncertainty principle,
ΔE ΔE Δt 1
Δv 1/ 2 = = = (7.176)
h h Δt 2πΔt
where Δv 1/ 2 is the Heisenberg absorption linewidth for the transition, Δt is lifetime of the excited state.
The exact shape of the NMR signal in the case of two equivalent exchanging groups with no coupling is
expressed by the function
K n (v A − v B )
2
I (v ) = (7.177)
⎡1 ⎤
⎢⎣ 2 (v A v B ) − v ⎥⎦ + 4π τ (v A − v ) (v B − v )
2 2 2 2
Here, I(v) is the intensity at the frequency ‘v’, Kn is a normalisation constant for the exchange and t = 1/2 k,
where k is the rate constant of the exchange; t, vA and vB are functions of temperature and cannot be determined
separately. These days computer programs are available in which estimated values for t, vA and vB are used to
determine I(v) which is then compared to the observed spectrum, i.e. hit and trial values of t, vA and vB are selected
such that the difference between the theoretical and observed NMR peak is minimised.
Direct calculation of lifetime of a specific spin state from Eq. (7.177) can be made over a limited temperature
range. Beyond a certain temperature, the rate of exchange is so fast that the magnetic environment of the two sets of
nuclei are identical and hence cannot be discerned. Thus, we have only one set of spins with a lifetime determined
by s-s and s-l relaxation processes.
In order to study the exchange rate over a wide range of temperature, a series of approximations are made in
Eq. (7.177), though the approximate methods may give less accurate results as compared to Eq. (7.177).
(a) Slow and Intermediate Exchange When the exchange rate is very slow, the spectrum consists of two
lines. In this region of temperature, τ >> (v A − v B ) , consequently Eq. (7.177) becomes
−1
K nT
I (v )A = I (v )B = A′
(7.178)
1+T 2
2A (v A − v )2
Here, T2A′ is the s-s relaxation time. Comparing Eqs (7.177) and (7.178), we obtain the linewidth of line vA
1⎛ 1 1⎞
( Δv A )1/ 2 = + ⎟ (7.179)
π ⎝ T2A ′ τ A ⎠
In the absence of exchange, the linewidth is
1
( Δv 0 )1/ 2 = (7.180)
πT 2 A ′
The line broadening due to exchange is equal to1/ πτ A . A value of k ( = 1 / 2τ ) is determined by comparing line-
widths of exchange peaks to those of peaks observed at temperatures where the rate of exchange is very small.
k π [( v e ) − ( Δv ) ] (7.181)
For slow exchange, the rate is related to peak separation by the expression
π 1
k
2
( v − ve ) 2
(7.182)
where Δvi is the peak separation in cps (i = e or 0). Equation (7.182) is obeyed over a limited temperature range
where there is extensive overlapping between the two separate peaks but not too close to coalescence.
A third method based on the intensity measurements called ratio method may also be applied in the slow
exchange region. k is determined from the ratio of the intensities of the peaks, Imax to the intensity midway between
the peaks, Imin;
I max
r=
I min
−1
π Δv 0
and k= [r ( r 2 − r )1/ 2 ] 2 (7.183)
2
(b) Coalescence Temperature (Tc) The temperature at which the appearance of the spectrum changes from
that of two separate peaks to that of a single, flat-topped peak is called coalescence temperature [Fig. 7.48(c)]. At
this temperature,
π Δv 0
k= (7.184)
2
(c) Fast Exchange NMR may be used to study fast exchange rates where the mean lifetime is in the region of
one second to one millisecond. Above the coalescence temperature, the spectrum consists of a single peak. In this
region, τ << ( − ) −1 and Eq. (7.177) reduces to
K nT 2′
I( ) = (Eq. 7.178a)
1 + π T 2′ ( A B )2
If the signal is not completely collapsed, i.e. the process is slow enough to contribute to its width but is still well
beyond the rate corresponding to separate signals, then the following expression results.
π Δv 02
k= [( Δv e )1 2 ( Δv 0 )1 2 ]−1 (7. 185)
2
Problem 7.23: The PMR spectra of N, N-dimethylacetamide; N, N DMA (in CC14 containing 2 per cent
TMS) in the temperature range 279–368 K on 50 MHz instrument is presented in Fig 7.49. Calculate the rota-
tional barrier from the PMR spectral data.
Solution The study of barrier to internal rotation is an example of dynamic nuclear magnetic resonance
spectroscopy. Reaction rates and rotational barriers in the range 10−1−10−5 s−1 and 3–20 kcal mole−1 respectively
can be studied by this method, e.g. N, N DMA (16.8 kcal mole−1), nitrosoamine (23 kcal mole−1).
cis-trans isomerisation of N, N DMA in which reactants and products are chemically and hence magnetically
identical is shown as
T = 368 K
T = 348 K
T = 334 K
T = 331 K
T = 308 K
n, Hz (Relative to TMS)
Fig. 7.49 Effect of temperature on the shapes in the NMR spectrum of
N,N-dimethylacetamide.
O (CH3)A O (CH3)B
C N C N
CH3 (CH3)B CH3 (CH3)A
In order to determine the barrier to internal rotation, the value of rate constants at different temperatures need
be estimated from the PMR spectra at different temperatures (Fig. 7.49). The data deduced from the spectra are
T, K (Δne)1/2
368 2.8193
348 3.9427
334 9.1337
331 10.1778
308 r = 8.46
Δv0 = 10.844 (Δv0)1/2 = 2.5103
Using this data and employing the Eqs (7.181−7.185), the values of k, the rate constant, are
T, K 368 348 334 331 308

−1
k, s 598.0 129.0 27.8 24.1 5.95
From the plot of ln k versus 1/T, the value of barrier to internal rotation about C—N bond has been found to
be around 17 kcal mole−1.
Problem 7.24: The behaviour of PMR spectra of 1-ethylaziridine (C4H9N) and l-ethyl-2-methyleneaziridine
(C5H9N) has been observed as a function of temperature on a 40 MHz instrument. The coalescence peak due to
ring protons for the former has been observed at 108 ± 5°C whereas for the latter at −65 ± 10°C. If the bandwidths
of the respective peaks are 29.9 and 26.4 cps, determine the rate of inversion of nitrogen in the two systems.
Solution We know that sufficiently rapid exchange of nuclei between two individual chemical environments,
is capable of collapsing a complex NMR pattern into a simpler one. So the time of inversion of nitrogen is related
to the bandwidth of the coalescence peak by the expression
2
τ= (7.186)
π ( Δv 0 )
Since the rate constant k = 1/t

π ( Δv 0 )
k=
2
For 1-ethylaziridine,
22
k= 66.45 s −1
× 29.9 = 66
7 2
For 1-ethy1-2-methyleneazirdine,
22
k= 58.67 s −1
× 26.4 = 58
7 2
Problem 7.25: The hydration of pyruvic acid (C3H4O3) in presence of HCl was carried out by NMR technique.
The spectrum of pure pyruvic acid consists of two signals at 10.70 and 2.60 ppm relative to an internal reference
TMS. The integrated intensities of these bands are in the ratio 1 : 3. The NMR spectrum of 4M solution of pyruvic
acid in heavy water at 34.5°C exhibits three bands at 5.48, 2.60 and 1.75 ppm respectively. Assign the bands to
the different functional groups present in the system. Calculate the equilibrium constant of hydration process of
pyruvic acid if the intensities ratio of methyl protons of pyruvic acid and hydrate of pyruvic acid is 1:1.4.
Solution On the basis of intensities ratio of the signals of pyruvic acid, the band at 2.60 ppm may be assigned
to methyl protons while at 10.70 ppm to carboxyl proton. The hydration of pyruvic acid in the presence of HCl
takes place according to the scheme
K
CH3COCOOH + H2O CH3C(OH)2COOH
Pyruvic acid 4M HCl 2,2-dihydroxy
propionic acid
The proton exchange process in presence of 4M HCl becomes so fast that the carboxyl, hydroxyl and water
protons may exhibit only one signal. Thus, the signal at 5.48 ppm may be assigned to these protons. The signals
at 2.60 and 1.75 ppm may be assigned to the methyl protons of pyruvic acid and 2,2-dihydroxy propionic acid
respectively. We know that
Intensity of hydrated methyl resonance
K eq =
Intensity of pyrru
uvic acid methyl resonance
1
= = 0 71
1.4
Problem 7.26: The kinetics of hydration of pyruvic acid
k
CH3COCOOH + H2O CH3C(OH)2COOH
in presence of HCl at 34.5°C to form 2,2-dihydroxy propionic acid has been investigated by NMR technique. As
the exchange is increased by the addition of HCl, the methyl proton resonances of pyruvic acid and its hydrate
are observed to broaden. With further increase in HCl, the methyl proton resonance coalesces and finally a single
narrow line is observed at very fast exchange. The exchange broadened resonance bandwidth of the methyl proton
resonances of pyruvic acid and its hydrate at different concentrations of HCl are given in Table 7.15. Calculate the
hydrogen-ion catalytic rate constant.

1 1
πΔv 1/ 2 = +
T2 τ
Table 7.15 Methyl proton resonances of pyruvic acid and its hydrate.
Bandwidths (cps)
HCl Molar concentration Pyruvic Acid Hydrate of Pyruvic Acid
0.250 0.318 0.374
0.450 0.949 1.413
0.900 1.817 2.193
1.00 1.590 2.446
(Continued)
Table 7.15 Methyl proton resonances of pyruvic acid and its hydrate. —Cont'd.
1.50 2.446 3.975
1.80 3.180 4.542
2.25 3.975 6.360
3.00 5.300 −
4.00 7.950 −
T2 is mainly responsible for the natural width of the resonance of a nonviscous liquid, t the mean lifetime, a
proton spends at a particular site in the exchange process contributes to the natural width of the signal. Thus, by
measuring the bandwidth in the absence of exchange, the transverse relaxation time may be found. The mean
lifetime may now be found by measuring the bandwidth of the exchange broadened resonance, i.e.
1
τ=
πΔv 1/ 2
The specific rate constant, k of the exchange process is given by
k = kH+ [H + ]
where, kH+ is the hydrogen ion catalytic rate constant. We know that k = 1/t. Consequently, 1/ τ = k H+ [H + ].
The reciprocals of the mean lifetimes of pyruvic acid and its hydrate are plotted against the hydrochloric acid
concentration to obtain k H+ . The value of k H + is equal to the slope of the lines for pyruvic and its hydrate, and are
found to be 5.55 and 16.32 s−1 respectively.
(d) Proton Exchange Rate Constant of N, N-Dimethyl Urea (DMU) The rates of exchange of protons in proteins
have been postulated to reflect the structure and conformation of these molecules. The proton-exchange rate of model
amides has been observed to be both acid and base catalysed by H + and OH− ions. The acid-catalysed exchange occurs
through a mechanism involving protonation of the C = O groups with subsequent release of the N—H proton to the sol-
vent. The proton-exchange rate of DMU can also be determined from the effect of pH on the width of the NMR signal.
Several mechanisms of the general acid catalysis type may be proposed by which an N—H proton can be
exchanged with the solvent proton, i.e.
O OH
k1 δ+
CH3 N C N CH3 + H3 O+ CH3 N C N CH3 + H2O
k2
H H H H
kH
OH
CH3 N C N CH3 + H3O+ (A)
O O
kOH −
CH3 N C N CH3 + OH− CH3 N C N CH3 + H2O (B)
k3
H H H
O O
*
CH3 N C N CH3 + (HOH)n + CH3 N C N CH3
H H* H H
k0
O O
*
CH3 N C N CH3 + (HOH)n + CH3 N C N CH3 (C)
H H H* H
Scheme (A) describes the acid-catalysed exchange in which hydronium ion protonates the oxygenatom caus-
ing the N—H proton to become more acidic. The N—H proton is then abstracted by the solvent molecule (H2O)
affecting an observable exchange. The reproduction of N site is considered diffusion controlled and is not impor-
tant in describing an effective NMR proton exchange.
The mechanism (B) indicates the direct abstraction of proton by a base which in H2O solution is the hydroxide
ion. There also exists a self-catalysed exchange through water bridge as shown in scheme (C). For DMU, this
contribution to the overall exchange rate is too small to be measured by high-resolution NMR line-broadening
techniques.
In presence of acetic acid (HA), chloroacetic acid (CA), trifluoroacetic acid (TFA) and phosphoric acid (H3PO4),
two additional mechanisms, i.e. (D) and (E) are possible:
O HO HA
k4 δ+
.
CH3 N C N CH3 + HA CH3 N C N CH3
k5
H H H H
OH
k6
CH3 N C N CH3 + HA (D)
H
O O
kb −
CH3 N C N CH3 + A− CH3 N C N CH3 + HA (E)
H H
H
Scheme (D) gives the contribution due to HA, the unionised acid in solution whereas scheme (E) gives the
contribution from the conjugate base.
In the absence of added acetic acid, substituted acetic acids or phosphoric acid, the net equations describing
proton exchange in the acid and base catalysed regions as obtained from schemes (A) and (B) are
1
= k 1k ⎡⎣ H + ⎤⎦ / ( k 2 + k H ) (acid catalysed) (7.187)
τ
1
and = k OH ⎡⎣OH − ⎤⎦ (base catalysed) (7.188)
τ
In Eq. (7.187), k2 may be assumed to be diffusion controlled and the assumption k 2 k H is made,
Consequently,
1
= k 1k ⎡⎣ H + ⎤⎦ / k 2 (7.189)
τ
k 1k H
The terms k 1 k and k 2 may be combined to yield k ex = , producing the acid catalysed rate equation,
k2
1
= k ex ⎡⎣ H + ⎤⎦ (7.190)
τ
The reverse reaction in the kH step is not considered since a net observable NMR exchange occurs with
completion of kH.
Considering mechanisms (D) and (E) for the effect of acetic acid, substituted acids or phosphoric acid, we
obtain
1
= ka [ ] + k b ⎡⎣ A − ⎤⎦ + k ex ⎡⎣ H + ⎤⎦ + k OH ⎡⎣OH − ⎤⎦ (7.191)
τ
The effect of addition of acetate ions is due to the second term i.e. k b ⎡⎣ A − ⎤⎦ of Eq. (7.191) since in all cases
except acetic acid, the concentration of undissociated acid HA is negligible. For phosphoric acid, the pH range of
interest is 6-8, thus the important species are H 2 PO 4− and HPO 2−

4 . In
1.5
such a case, Eq. (7.191) in the acid catalysed region reduces to
1 1.0
log 1
τ
= k b ⎡⎣ A − ⎤⎦ + k ex ⎡⎣ H + ⎤⎦ (7.192) (a) (b) (c)
(d)
τ
In order to obtain proton-exchange rate constant under differ- 0.5
ent conditions, T2 at different pH values are obtained from the line

width of the CH3 doublet under conditions of slow exchange and are 5 6 7 8
limited by field homogeneity. Thus, by making use of Eqs (7.190) pH
and (7.188), to acid and base catalysed regions, respectively, one Fig. 7.50 Log of the exchange rate (1/t) of N.N-
can determine kex and kOH. The plot between log (1/t) and pH for dimethyl urea against pH for (a) 1M dimethylurea,
(b) 1M dimethyl urea + 1M trichloroacetic acid,
DMU in aqueous solution and in the presence of acetic acid, mono- (c) 1M dimethyl urea + 1M monochloroacetic
chloroacetic acid and trichloroacetic acid has been shown in Fig. acid, (d) 1M dimethyl urea + 1M acetic acid.
7.50. In order to obtain kb, use is made of Eq. (7.192). From the
slope of the curve describing 1/t as a function of ⎡⎣ A − ⎤⎦ at constant
pH, kb can be calculated.
The proton-exchange rate constants for N,N-DMU determined by NMR technique are listed in Table 7.16.
Table 7.16 Kinetic results for N,N-dimethylurea (Temp = 39°C, Operating frequency = 56.4 MHz).
System kex (mole−1 s−1) kOH (mole−1 s−1) kb (mole−1 s−1)

1 M DMU 2.5 × 106 (x) 1.0 × 106 NA
DMU + 1 M(HA) 1.48 × 10 (x) 8
1.1 × 10 6
73.0
DMU + 1 M(MCA) 4.2 ×10 (x)
7
1.0 × 10 6
15.0
DMU + 1 M(DCA) 2.09 × 107 1.0 × 106 11.6
DMU + 1 M(TCA) 2.4 × 10 (x)
7
1.3 × 10 6
6.5
DMU + 1 M(TFA) 2.5 × 10 7
0.8 × 10 6
6.5
DMU + 1 M(NaCl) 1.1 × 10 6
1.6 × 10 6
NA
DMU + H3PO4 NA NA 205(pH 7.0)
(x)-values for the observed rate constant obtained from the intercepts in Fig. 7.50.

NMR technique being rapid and non-invasive is widely used in studies related to biology, e.g. human,
animal and plant subjects. It has been extensively used in studies related to agricultural plants and their
products. The principle of FID of low-field NMR is used for rapid and nondestructure determination of oil
and moisture in oil seeds. Both s-l and s-s relaxation times T1, and T2, have been exploited to obtain degree
of unsaturation of oil in oilseeds and dry rubber content in natural rubber latex. 31P-NMR is used to eluci-
date the mechanism of uptake of phosphorous by plants. The fatty acid composition of oil in a single intact
seed can be obtained using 31C-NMR. Analysis of mixtures of amino acids and of peptides can be carried
out using 19F-NMR.
7.40.1 Determination of Oil Contents in Oilseeds

The oil contents in oilseeds, i.e. groundnut, mustard, til, cotton, sunflower, soyabean, safflower, etc., can be
determined by the liquid FID signal and solid-liquid FID signal methods.
(a) Liquid FID Signal Method Hydrogen nuclei which show strong NMR peaks in oilseeds exist mainly
in four forms: oil, moisture, carbohydrates and proteins. The s-s relaxation times for protons of oils, moisture
and carbohydrates—proteins are of the order of 10−1, 10−3 and 10−6 second respectively. Thus, the decay of
NMR signal in oilseeds will follow the order: [Solid (carbohydrates—proteins)] (fast) > [moisture and oil]
(slow). Based on the order of decay of NMR signal, the oil contents in oil seeds can be determined as follows.
The dried and weighed seeds are subjected to a p/2 pulse and FID corresponding to the oil content in seeds is
monitored. The oil content is then noted from the standard linear graph between the FID NMR signal and the
amount of pure oil.
The major factors which affect the efficiency of this method are s-l relaxation time T1 and field inhomogene-
ity. The effect of s-l relaxation time on the NMR signal is eliminated by selecting the pulse repetition time long
enough to attain the full equilibrium magnetisation. T1 for most of the oilseeds is < 200 milliseconds, so pulse
repetition time of the order of one second is sufficient, i.e. 5 × T1, Magnetic field inhomogeneity is eliminated
by proper selection of delay times tD’s so as to measure T2 rather than T2* which is governed by magnetic field
inhomogeneity of the spectrometer.
(b) Solid-liquid (S-L) FID and Spin Echo Signal Method The liquid-signal method is slow and not truly
nondestructive since the seeds are to be weighed and dried. In the S-L method, the system is subjected to a p/2−tD
−p−AQ sequence and the response is monitored. The FID and spin echo signals of protons corresponding to the
solid and liquid constituents measured by the p/2−tD−p−AQ pulse are displayed in Fig. 7.51.
π/2
Dead time of the receiver
π Pulse
S
Spin-Echo
Intensity
Signal
S+L Oil percent = [L/(L + FS)] ×100
t1(μs) t2(μs)
Fig. 7.51 FID signal of dried seeds.
The signal intensity at t1 μ s (immediately after the dead time of the receiver) will have full contribution from
the liquid phase signal (L) and partly decayed solid phase signal (S) in the seeds. The spin echo signal measured
after the application of p-pulse, at t2 μs represents nearly full liquid phase signal only as solid-phase signal has
almost completely decayed by then. The oil content is then determined by the formula.
L
Seed oil content = × 100 (7.193)
L + FS
This formula holds good for samples having similar moisture and solid phase, T2. The correction factor,
F accounts for the decay of solid signal during the dead time of the receiver, and proton density difference in solid
and liquid phases of oilseeds and the residual moisture signal at t2 m s. Using Eq. (7.193), mean values of F are
estimated from some dried samples of seeds of known oil contents. Once F is established, Eq. (7.193) can be readily
used for the estimation of oil in oilseeds.
7.40.2 Dry Rubber Content in Natural Rubber Latex

s-s relaxation rate constants of solid and aqueous phases of latex are markedly different. The signals due to rubber
particles (mostly solid) and the aqueous phase of latex are resolved by spin echo technique. The signals cannot be
resolved by FID method where both the phases decay with T 2* .
7.40.3 Study of Fate of Inorganic Phosphates (Inorganic phosphate)

Pi
(Pi) in Developing Seeds Phytate
31
P-NMR spectra of perchloric acid extract of seeds collected at differ- 2
3
4
1
ent periods of flowering showed a decrease in intensity of the signal due
(a)
to Pi accompanied by an increase in intensity of signals corresponding
to phytates (Fig. 7.52). A similar behaviour of Pi and phytate signals
have been observed in the 31P-NMR spectra of intact seeds (in vivo) at
similar period of flowering. Thus, 31P-NMR study of seeds supplements
the already established fact that phytate acts as a store of phosphorus (b)
which may be utilised during the seed development.
7.40.4 Analysis of Mixtures of Amino Acids

and of Peptides
The mixture to be analysed is treated with trifluoroacetic anhydride and the
19
F-NMR spectrum of the resultant N-trifluoroacetylated (N-TFA) prod-
uct is recorded with acetone as solvent. The fluorine chemical shift is very 3 2 1 ppm
sensitive to small differences in shielding, hence, most N-TFA derivatives (c)
of amino acids have distinctive F chemical shifts as shown in Fig. 7.53.

19 Fig. 7.52 31
P-NMR spectrum of perchloric
extract of oilseeds after different periods: (a) <
The relative proportions of components of the mixture of amino acids (b) < (c) of flowering. The numbered peaks
may be judged from the peak heights of each COCF3 signal. Dipeptides correspond to phytate and the signal labelled Pi
may also be differentiated in this manner even when two peptides differ corresponds to inorganic phosphate.
only in their second amino acid units, e.g. the 19F resonances of methyl
esters of N-TFA-glycyl-L-Leucine and N-TFA-glycyl-L-alanine are resolved at
84.6 MHz. Alanine
7.40.5 Optical Purity of Dipeptides and Leucine
Configuration of Diketopiperazines Isolated

Glycine
from Natural Sources
Dipeptides may be converted to corresponding cycloforms (diketopiperazines)
without change of configuration (See A) and the cyclodiastereoisomers also differ
in their PMR characteristics especially if aromatic groups are present. Thus the
methyl resonances of DL (trans) and LL (cis) cyclo-alanyl phenylalanine differ by
34 Hz. The higher field position of the cis signal is attributed to aromatic shield-
ing which arises as a result of preferred conformation in which benzene ring faces
a piperazine ring (see C), equatorial methyl as in (B) is too far removed to be
affected in this way. These PMR differences serve as a further check of the optical
purity of dipeptides and also are of potential value in establishing the configura- 0.3 0.5 0.7
tion of diketopiperazines isolated from natural sources. ppm from CF3COOCH3
In addition to various parts of plants, NMR technique is also used to investi- Fig. 7.53 F-84.6 MHz NMR spec-
19
gate small animals or parts of large animals, e.g. a human leg or arm still attached trum of a mixture of N-TFA-L-leucine,
to its living host. The concentration of intracellular adenosine-5′-triphosphate N-TFA-L-alanine and N-TFA-glycine
(ATP) in some cases approaches 1 mM or more, so 31P-NMR method has been in acetone.
exploited to study the behaviour of phosphate compounds within cells. 3!P-NMR
spectrum of the contents of mouse tumor cells displayed in Fig. 7.54 reveals that inorganic phosphate (Pi) com-
bines with adenosine diphosphate (ADP) to form ATP, a molecule in which living cells store chemical energy. The
interconversion of ATP into ADP and Pi in the cell is coupled to passage of protons through the membrane of the
cell. The reaction that generates chemical energy by splitting ATP into ADP and Pi serves to expel protons from
the cell, thereby creating a proton gradient that stores both electrical and chemical energy. Conversely, the re-entry
of protons into the cell provides energy required by the reaction that combines Pi and ADP into ATP. A change in
the concentration of protons inside the cell constitutes a change in the intracellular pH. At an intracellular pH of
6.7 the populations of the singly protonated Pi form and the doubly protonated Pi form are equal. At higher pHs
(a more basic environment), the singly protonated form dominates and at lower pHs (a more acidic environment),
Me CH2Ph
t-BuOCONHCHCONHCHCO2Me
L L
Boil HCOOH
O
Me
NH
HN
CH2Ph
O
LL(cis)
(A)
O
O H
H N C
N
CH2Ph H N
Me H
N O
H Me
O
(B) (C)
Trans Cis
νMe86.5 Hz νMe52.5 Hz from TMS,
60MHzinTFA
the doubly protonated form dominates. Each form has a distinctive chemical shift, but the inorganic phosphate
peak in a NMR spectrum of cells will be found in some intermediate position depending on the average degree
of protonation of all the Pi in the cells. Thus the position of Pi peaks yields an accurate measure of the degree
of protonation of the Pi, which in turn yields an accurate measure of the intracellular pH. 31P-NMR spectrum of
living muscle from the frog exhibits phosphorous peaks from Pi, from ATP and from phosphocreatine (Pcr), a
molecule that muscle cells draw on to convert ADP to ATP. When external energy is not available to the muscle,
the intensity of the signal corresponding to Pcr decreases whereas the intensity of the ATP signals is maintained
Inorganic phosphate Adenosine diphosphate (ADP) Adenosine Triphosphate

(Pi) (ATP)
NH2 NH2
C N C N
N C N C
CH CH
O O O HHC N C N O O O H
HC
N
C
N
−O
β α −O
O P O + P O P O C H P O P O P O C H
γ β α
−O −O −O O −O −O −O O
H H H H
H H H H
OH OH OH OH
Absorption
−5 0 5 10 15 20
Fig. 7.54 31
P-NMR spectrum of the contents of mouse tumor cells. Change in magnetic field strength (ppm of base value).
Table 7.17 15
N shifts (in ppm from H 15NO3 ) of nitrogen atoms in ATP and in equimolar metal ion-ATP solutions.
NH2
N
N 6 J15N–7–H–8 = 10 Hz
5 7
1 H
2 8
3 4 9 J15N–1–H–2 = 16 Hz
H N N
N-1 N-3 N-7 N-9 NH2-6

ATP 144.7 135.6 129.5 191.6 282.8
2+
Mg -ATP 144.9 135.0 129.5 191.6 282.3
2+
Zn -ATP 144.9 135.0 132.5 190.1 279.7
constant by the reservoir of Pcr. When Pi is completely utilised, the intensity of ATP signal decreases leaving ADP
and Pi. The position of ATP signal indicates that all the ATP in the cells are complexed with magnesium ions.
FT 31P-NMR spectra of ATP in presence of Mg2 + in an aqueous medium reveal that phosphorous chemical shifts
corresponding to α β , and γ phosphorous atoms move down-field as the Mg2 + concentration is increased. For a
1 : 1 composition of Mg2 + and ATP, β − 31P shifts from −22.5 to −20.5 ppm, α − 31P moves slightly, i.e. from −14
to ≈ − 13.7 ppm and γ − 31P moves from −11 to ≈ − 10 ppm.
The metabolism throughout the animal’s body can be investigated with NMR spectra. 13C-NMR spectra of rats
fed on glucose labelled with l3C revealed that the intensity of glucose peaks decreases and ultimately disappears
from the stomach of the living animal. After some hours, the level of glycogen in the animal’s liver increases, the
glucose is stored as glycogen there. Finally, the glycogen signal disappears as the labelled glycogen is consumed.
The nitrogen-15 resonance spectra have been used to ascertain the extent and type of complexing of alkaline earth
metals (Mg2 + and Zn2 + ) to adenosine triphosphate (ATP). The I5N chemical shift values given in Table 7.17 indi-
cate that although Mg2 + does not bind to the adenosine ring nitrogens in the ATP-Mg2 + complexes, the shifts for
the Zn2 + case are consistent with a model in which Zn2 + interacts at both the 6-amino and the N-7 positions.
7.40.6 Protein–Drug Interactions

NMR is used in pharmaceutical industrial research including screening studies when a particularly important
protein is being considered as a target for new drugs. The protein is exposed to a variety of small molecules; the
potential candidates for a new drug. When a particular small molecule interacts with the macromolecule, the
NMR spectrum of the macromolecule will change. If the resonances of the macromolecule have been assigned, it
is possible to determine which residues are involved in the interaction, and the potential use of the small molecule
in further drug development can be assessed.
5
4
2
Phenyl H⬘s H⬘s of C10 + C14H2O C5H C18H⬘s H⬘S of C7 + C8
1 6
2 4 5
H
Fig. 7.55 Proton magnetic resonance spectrum of penicillin G (a) D2O and (b) in D2O in the presence of 10% W/V bovine serum albumin.
Table 7.18 Effect of bovine serum albumin on the NMR of penicillin G.
1
−1
Peak T 2* (s ) Stabilisation ratio
1 (pheynl protons) 3400 5200

4 (C5 proton) 1430 1700
5 (C18 proton) 5975 2060
In general, it has been found that PMR spectra of proteins consist of a few very broad peaks which form a
rather uninformative background on which the high resolution spectrum of small molecules is superimposed. Dif-
ferential broadnening of one peak as compared to others gives some indication that the portion of small molecule
from which that peak originates is more affected by the interaction with the protein than the remainder of the
molecule. For example, the peak arising from the benzyl methylene group of penicillin G broadens much more
than the other peaks when antibiotic is bound to bovine serum albumin (Fig. 7.55). This effect is clearly specific,
in that it is not observed in the presence of other proteins, e.g. ribonuclease.
The s-l relaxation time T1 can be determined by saturating PMR signal then measuring its intensity (At) as a
function of time (tD). T1 can then be calculated from the equation.
At A 0 ⎡⎣1 − exp ( −t D / t1 )⎤⎦ (Eq. 7.31)
17 8
O H H H
9 CH3
H H
S
C
H CH2 C N C C CH3
18 16 15 14 10 7
H H
N C
5
COO–
C
O 12 4
11 H
13
*
by plotting ln [( A 0 A t ) / A 0 ] versus tD. The s-s relaxation time T 2 can be obtained from the line width of high
resolution spectrum using the relationship.
1
Δv 1 = (Eq. 7.27)
2 πT 2∗
where Δv 1 is the line width at half-height.
2
The relaxation rates for three of the peaks of bound penicillin G are listed in Table 7.18.
While the methylene peak has the fastest relaxation rate, it can be seen that the phenyl group has the highest
stabilisation ratio R, where
T 2* for free penicillin G
R= * (7.194)
T 2 for bound penicillin G
Since the stabilisation ratio is a more significant indicator of an effect on molecular motion, it would appear
that the binding of penicillin G to bovine serum albumin involves mainly hydrophobic interactions between the
phenyl group of the antibiotic and the protein binding site.
7.40.7 Nuclear Magnetic Resonance Imaging (NMRI)

NMRI is one of the most important and promising applications of FTNMR. In NMRI, the map of density of the
NMR nuclei, say 1H, 2H, 13C, 31P, 23Na, in various regions of the solid and semisolid objects is produced rather than
the NMR spectrum. Basically, the FIDs are Fourier transformed and converted to three-dimensional images of the
interior of the objects. NMRI enjoys advantage over x-ray CAT scans or other methods in that it is non-invasive, i.e.
there is little or no potential danger of radiation injury or other damage to human, or animal or plant subjects.
In 1DFTNMR, the magnetic resonance signal of a sample placed in an uniform magnetic field (1 part in 108)
of strength H is given by
γ
v H (7.195)
2π
Being a waveform, the FID corresponding to each resonating species is a composite of three parameters, viz.,
the signal amplitude, frequency (v) and phase (j). Amplitude is a measure of the number of resonating nuclei
(proton density), v and j may be used together to code the resonance signal at each time-domain point.
The fundamental principle of NMRI is that the signal frequency is made to vary deliberately in a linear manner
with the spatial location of the group of nuclei producing the signal by applying moderate (up to 20 mT m−1) linear
magnetic field gradients. Thus, for imaging experiments, Eq. (7.195) modifies to
γ
vi
2π
(
H + G ri ) (7.196)
where, vi is the frequency of the proton at position ri; and G is the vector representing the magnetic gradient ampli-
tude and direction, i.e. intensity at different spatial locations of the proton.
⎛ ∂H ⎞ ⎛ ∂H ⎞ ⎛ ∂H ⎞
G i⎜ + j⎜ +k⎜ (7.197)
⎝ ∂x ⎟⎠ y z
⎟
⎝ ∂y ⎠ z , x ⎝ ∂z ⎟⎠ x y
Thus, in order to obtain ID image projection, the signal is obtained in the simultaneous presence of field gradient Gz
and radio frequency pulse and the intensity is plotted against the frequency, with the frequency axis being coded.
Consider, for example, the situation as shown in Fig. 7.56 where the field gradient in H, Gz is created by auxiliary
coils in the magnetic bore that is under the control of the computer of the magnetic resonance instrument. Protons
in different positions in the sample experience different magnetic fields H1 and H2 for two of the regions shown in
γ H1
Fig. 7.56 and will have different frequencies ν1 = and v 2 γ H 2 / 2π . The frequency n2 will be higher than the
2π
frequency v1, and the frequency difference Δ v = v2 – v1. Thus by changing the centre frequency of the NMR probe
pulse in increments of Δv, it is possible to probe successive ri positions in the direction of the applied magnetic field
gradient. Each consecutive radio frequency pulse produces an FID signal that encodes the concentration of protons
at each ri position along the direction of the field gradient. By performing the Fourier transformation on the FIDs,
concentration information can be produced as indicated by the heights of the peaks. Images in three directions can
be made by an extension of this coding concept using additional magnetic field gradients along the Y and X axes.
The precessing spin will experience a gradient pulse of duration Δt along either of these axes and will accumulate
an additional phase, e.g. jy = g Gy y Δt. Thus phase and frequency encoding of the three spatial dimensions can be
achieved by linear magnetic field gradient applied successively along three orthogonal directions for the duration t1
t2 and t3 respectively in a pulsed FT experiment. The recorded FID signal is Fourier transformed to produce S(v1, v2,
v3) which is equivalent to three-dimensional spatial image when spatial information is decoded using the relations;
x = v1/Gx; y = v2 /Gy and z = v3/Gz with three field gradients Gx, Gy and Gz, i.e. the concentration of protons in a volume
element or voxel corresponding to each set of coordinates x, y and z is contained in S(v1, v2, v3). The ability to gener-
ate high-resolution (millimetre) images from ‘three-dimensional
data is a unique strength of the NMRI technique. Note that three
orthogonal gradients are produced by employing three sets of H H2 Field Gradient
H1
orthogonal gradient coils. Further, the direction of gradient may
be changed at will by suitable linear combination of currents in
the three sets of coils. dc Offset
Abnormalities due to different diseases in different parts of
Field Strength
humans or animals or plant subjects are imaged routinely these
days. Tumours in the human brain, spine, etc., and perfusion
(a microcirculation process which measures nutrient delivery to
tissues) are being imaged routinely. In food industry, NMRI has
been used for the determination of fat and muscle distribution in
γ H1 γ H2
meats, the determination of water and lipid contents in sunflower ν1= ν2=
2π 2π
oil, composition of foods such as biscuits during processing,
investigation of hydration of foods, to measure water movement
in heterogeneous systems, i.e. plants, etc., the determination of
NMR relaxation parameters in cured pork.
Problem 7.27: A biological sample is placed along the bore Frequency

(or Z-axis) of a NMRI magnet. What would be the difference Fig. 7.56 Fundamental concept of nuclear magnetic
between the resonance frequencies of the two successive protons resonance imaging.
1 mm apart in the sample

p if a magnetic
g field gradient of 10−4 T(mm)−1 is applied along the bore of the NMRI
(
magnet? γ H = × )
Solution According to Eq. (7.196), the frequency of proton in the sample in a magnetic field gradient Gz is
given by
γ
vi (H + G z ri ) (7.198)
2π
Let v1 and v2 be the frequencies of the protons at positions r1 and r2 respectively in the sample. Accordingly,
we write
γ
v1 (H + G z r1 ) (7.199)
2π
γ
v2 (H + G z r2 ) (7.200)
2π
Thus, the difference between the two frequencies is
γ
Δv = v 2 − v 1 = (r2 − r1 )G z
2π
γ
or Δv = ΔrG z (7.201)
2π
Here Δr = r2 – r1, is the distance between two successive protons.
Substituting the values of the respective parameters, we obtain
Δv =
68 × 108
2.68 ( ) ⋅1( ) ⋅100 −5
( ( ) )
2π ( radians )
= 426 Hz
Problem 7.28: A biological sample having 31P nuclei is placed in the bore of a NMRI magnet. A magnetic
field gradient of strength 10−5 T cm−1 is applied along the bore of the magnet and the frequency difference between
the two consecutive phosphorous nuclei has been found to be 10 Hz. Determine the distance between the two
consecutive 31P nuclei. g (31P) = 1.08 × 108 radians s−1 T−1.
Solution Here, Δv = 10 Hz, g = 1.08 × 108 radians s−1 T−1, Gz = 10−5 T cm−1,
2π Δv
Δ =
γ ⋅Gz
Substituting the values of the respective parameters into the above expression, we obtain
2π ( radians ) ⋅ 10(
0 Hz )
Δ =
1 08 × 10 8
( )10 (−5
)
2 22 10
= cm = 0.058 cm = 0.58 mm.
7 1.08 × 108 × 10 5
The distance between the two consecutive P31 nuclei in the biological sample is 0.58 mm.
7.40.8 Liquid Crystals in High Resolution NMR Spectroscopy

The NMR spectra of organic molecules in isotropic liquids are not caused by direct ‘through space’ interactions of
the individual dipoles because this interaction, i.e. d-d interaction averages out to zero through rapid reorientation
of internuclear axes. However, there remains a small isotropic interaction between nuclear spins mediated by
the electron-nuclear and electron-electron interactions in a molecule. Thus, the high-resolution NMR spectra
of organic molecules in isotropic liquids are determined (i) by isotropic chemical shifts (dAB), and (ii) by the
isotropic indirect s-s coupling (JAB ) acting between nuclei (A, B) of a molecule. As stated above the third basic
magnetic interaction, the direct dipole-dipole interaction between the nuclear magnetic moments is completely
averaged out. Moreover, the solute tumbling motion averages out the anisotropic contributions to JAB and dAB. The
anisotropic magnetic interactions affect the NMR spectra in isotropic media only through their influence of the
relaxation times (or the line widths). The isotropic molecular tumbling in ordinary liquids leads to a loss of most
valuable information. However, NMR spectra obtained in liquid crystals are primarily determined by the direct
intramolecular spin-spin coupling whereas isotropic and anisotropic contributions to chemical shift and the indi-
rect nuclear spin coupling respectively, appear as second-order effects.
The isotropic part of internuclear interaction is identical to indirect coupling constant. The anisotropic nuclear
( )
coupling DAB contains both the direct d-d interaction D AdirB and the anisotropic part of the indirect nuclear cou-
pling, D AindB i.e. D A B = D AdirB + D AindB D AindB has the some orientational dependence with respect to H as D AdirB and is
therefore, called pseudodipolar coupling. D AindB D AdirB at least for H H and C H interactions. D AindB amounts
to about 10 percent of the corresponding D AdirB in C6F6. D AindB can be estimated from the deviations between the
observed DAB values and the values of D AdirB calculated from the known geometry of the molecule.
D AdirB is a function (i) of the time average over the anisotropic solute tumbling motion, and (ii) of the inter-
nuclear distances. The NMR spectroscopy in liquid crystals may, therefore, provide information on both the aver-
age solute orientation and molecular geometry. Moreover, information on the absolute sings of indirect coupling
constants and on the anisotropic contribution to chemical shift and indirect coupling constant respectively may be
obtained from NMR spectra of partially oriented molecules.
The interpretation of NMR spectra in isotropic media is simplified by the rule that the indirect s-s coupling
between the so-called equivalent nuclei (nuclei with identical chemical shifts) does not lead to observable line
splitting. However, this rule does not hold good for direct nuclear coupling. The proton resonance spectrum of
acetylene is split up in doublet by the dipolar coupling between two equivalent nuclei. The spectrum of partially
aligned benzene exhibits about fifty lines. The spectrum of n interacting nuclei yields n(n − l)/2 interaction
parameters DAB, from which the geometry of the molecules can be established The direct d-d coupling (derivable
from the observed NMR spectra) between two nuclei A and B of molecules performing a rapid tumbling motion
(isotropic or anisotropic) is given by
3 2
θA B − 1
D AdirB K AB ; (7.202)
rA3B
γ Aγ Bh
where, K AB = (7.203)
4π 2
θ A B is the angle between the vector connecting the coupling nuclei and the applied field Hz and rAB is the inter-
nuclear distance. The angle brackets < > denote the time average both over the molecular tumbling motion and
0
over the internuclear motions (vibrations, internal conversions). When rAB is expressed in A and D AdirB in MHz,
then for protons
γ H2 h
K AB = = 120067 Hz Å3. (7.204)
4π 2
For a cylindrical molecule (such as acetylene) whose long axis Z coincides with the internuclear vector r A B, D AdirB
is given by
1
D AdirB 2K A B 3 S zz (7.205)
rA B
→
provided the magnetic field is parallel to the liquid crystal optical axis L S zzis the average orientation of the long
→
molecular axis (of acetylene) with respect to L .
For simple molecules, the line positions and the intensities can be expressed in terms of chemical shifts (iso-
tropic and anisotropic) and coupling constants DAB and JAB. A system of equations may be derived which allows
the determination of the coupling constants (DAB and JAB) from the experimental line positions. However, for
complex molecules, the spectra are only analysed by computer techniques either by a simulation or an iterative
method. In the simulation method, the input parameters used are molecular structural parameters and motional
constants. The theoretical spectra predicted from the input parameters (bond distances and motional constants)
are compared with experimental spectra. The trial and error procedure is carried out till both the spectra overlap.
Thus, the geometry of the molecular system under study is obtained.
The nematic phase has proved most useful as a solvent for NMR studies. Two kinds of nematogens are used as
NMR solvents. Organic, thermotropic nematogens are most widely applicable solvents (See Table 1.5). Lyotro-
pic liquid crystals may also be used as NMR solvents and have important property of dissolving both ionic and
neutral species. A mixture of caesium decyl-sulphate, n-decanol and water will accept a wide range of solutes to
form nematic phases. The applicability of nematogens as anisotropic solvents in NMR spectroscopy is limited
to relatively simple and symmetric molecules which do not have more than eight protons since overlapping of
lines occurs. Note that pure liquid crystals are formed from molecules having typically at least 12 protons, and
the number and magnitude of interproton dipolar coupling results in unresolved proton spectra covering about
30 kHz, e.g. 220 MHz proton spectrum of tropone (C7H6O) dissolved in nematogen Merck Phase IV exhibits
sharp intense lines from the six protons whereas the protons of the liquid crystal exhibit broad bands. The NMR
intensities are much lower in nematic solvents than in isotropic liquids, due to the large number of lines caused
by partially aligned molecules. So, concentration of the solute is kept in the order of 20 mole per cent. The quite
complicated and large molecules could, in principle, be studied and analysed with this technique if species that
are deuterated at specific sites could be prepared.
The ratios of nuclear distances can be obtained from NMR spectra of molecules in an isotropic solvent. Abso-
lute values of order parameters can only be determined for molecules with known internuclear distances. The
order parameter S = 1 / 2 < 3 c 2 θ − 1 > is a measure of the efficiency of the molecular orientation along the optic
→
axis L . Here, q is the angle between the long molecular axis and optic axis. For q = 0, S = 1, i.e. all molecular
axes are parallel. Such a nematic order is possible near the absolute zero point of temperature provided the mate-
rial would not freeze. For isotropic liquids S = 0. Depending on temperature, S = 0.4−0.7 for real nematic liquid
crystals. S is not substantially changed by applied magnetic or electric fields.
(a) Geometry of Molecules The relative bond lengths of molecules can be obtained from the NMR spectra in
anisotropic solvents. However, the geometry can be obtained in anisotropic media if the internuclear distances from
other techniques, namely x-ray diffraction, microwave spectroscopy are available, e.g. the proton resonance spectra of
X—CH3 (X = CN, I, OH, etc.) is characterised by one direct H—H coupling constant DHH. The spectrum consists of a
triplet with relative intensity ratios 1 : 2 : 1. The splitting between two adjacent lines is 3DHH. For molecules containing
1
13
C-isotope (I, the spin quantum number = ) in CH3 group, each of the triplet lines splits in doublets of separa-
2
tion DCH. The orientation of these molecules is described by only one order parameter (if X is an atom or a cylin-
derical symmetric group). Thus, we may write
D HH γ H rCH
3
= ⋅ 3 (7.206)
D CH γ C rHH
The angle H—C—H can directly be determined from the expression (7.206).
The internuclear distances determined from D AdirB are averages over the vibrational motions of the molecules. In
order to get accurate geometry of the molecules, the influence of molecular vibrations on D AdirB be estimated. Due
to this reason, the structural parameters determined from NMR differs from microwave and electron diffraction
methods, e.g. HCH bond angle in CH3OH determined by liquid crystal NMR technique is 110°3′ ± 8′ as compared
to 109°2′ from microwave spectroscopy. In case of CH3I, < HCH = 111° 42′ ± 2′ from liquid crystal NMR and
111°25′ from microwave spectroscopy.
(b) Shape of Molecules In favourable circumstances, we can predict theoretically the ratios of the dipolar cou-
plings to be expected for regular hexagon, pentagon and square and then by comparison with the experimentally
determined ratios, we can determine whether the rings really are regu lar in shape, e.g. data in Table 7.19 indicate
that benzene and the cyclopentadiene ring in the nickel compound have regular polygon shapes, whereas the ratios
of DAB values for other two compounds indicate a lower symmetry. However, it is not possible to determine the nature
of the distortion. Note that the internal rotation about the metal-ring bonds is giving an averaged ratio of the dipolar
coupling.
(c) Optical Active Molecules The order parameters are very sensitive to small changes in molecular geometry.
This makes the difference between the NMR spectra of d- and l-isomers of optically active molecules, e.g. the two
isomers of racemic mixture of trichloropropyleneoxide (C3H3C13O) yield different NMR spectra in a compensated
Table 7.19 Comparison of ratios of experimental interproton dipolar couplings for various molecules, with theoretical
values for regular polygons.
Molecule Experiment Theory
1.000 : 0.192 : 0.125 1.000 : 0.192 : 0.125(0.125)

Benzene
Mn(CO)3
Cyclopentadiene 4.11 ± 0.01 4.216(4.236)

manganese
tricarboxyl
Ni-NO
4.211 ± 0.003 4.216(4.236)
Cyclopentadienyl
nickel nitrosyl
Fe(CO)3
Cyclobutadiene iron 2.67 ± 0.04 2.768(2.828)
tricarboxyl
The quantities within brackets are without averaging.
nematic mixture of cholesteryl derivatives, i.e. cholesteryl chloride or cholesteryl bromide with cholesteryl esters in ap-
propriate amounts. Such a compensated mixture provides an environment which exhibits both anisotropy and helicity.
The two isomers have somewhat different order parameters in a solvent with a local screw sense and correspondingly
exhibit different values of the direct coupling constants. However, in an ordinary nematic solvent such as n-hexyloxya-
zoxybenzene, the two isomers give identical spectra.
(d) Barrier to Internal Rotation: The sensitivity of D AdirB to small changes in molecular geometry has been
exploited for determining barrier to internal rotation or conversion, e.g. 3,5-dichloroanisole, 3,3-bithienyl, tetra-
fluoroanisole, etc., o-chlorotoluene assumes a staggered conformation with a barrier height of 1.2 kcal mole−1.
The magnitude of the barrier in this case has been estimated by comparing the motional constants and the
relative molecular geometry obtained from the NMR spectrum with the corresponding values calculated by
assuming different models for the form of the rotational barrier.
7.40.9 RECENT DEVELOPMENTS

The 2002 Nobel prize for Chemistry was shared among John B Fenn, of Virginia Commonwealth University,
USA; K Tanaka of Nishinkokyo Kuwabaracho Nakagyou Ku, Kyoto, Japan; and K Wuthrich of Swiss Federal
Institute of Technology, Zurich, Switzerland; in the areas of mass spectrometry and NMR as applied to biological
macromolecules. These laureates have contributed in different ways to the development of these methods to
embrace biological macromolecules.
Fenn and Tanaka shared the Nobel prize for developing soft desorption ionisation methods for mass spectro-
metric analysis of biological macromolecules in relation to their identification and structure. The method due to
Fenn is called hovering through spraying, whereas the one due to Tanaka is called hovering through blasting. In
the former method, the sample is sprayed using a strong electric field to produce small, charged, free hovering ions
whereas in the latter the free ions are produced using an intense laser pulse.
Wuthrieh, on the other hand, shared the prize for his development of NMR spectroscopy for determining
the three-dimensional structure of biological macromolecules in solution. He showed that NMR is possible for
proteins and has invented a systematic method of pairing each NMR signal with the right hydrogen nucleus
(proton) in the macromolecule. This method is called sequential assignment and is today a cornerstone of all
NMR structural investigations. He also showed how it is subsequently possible to determine pairwise distance
between a large number of nuclei and used this information with a mathematical method based on distance geom-
etry to calculate three-dimensional structure for the molecule.
Presently, scientists can see the proteins and understand how they function in the cell. The 3He NMR has been
proved to be a powerful new tool for following fullerene chemistry. The soccer-ball shaped, sixty-carbon mol-
ecule, Buckminister fullerene and its larger relatives (the fullerenes) all have the intriguing feature of possessing
cavities surrounded by cages of carbon atoms. The rare isotope helium−3 can be imprisoned within the hollow
cavities in quantities sufficient for its observation by NMR. The electrons on the fullerene cage partially shield the
magnetic field, and so the resonance line of trapped 3He appears at a higher field (lower frequency) than for 3He
outside the fullerene. Each type of fullerene and each derivative of fullerene has a different shielding and hence
a different resolvable, NMR line, i.e: the sharp helium NMR peaks in different chemically substituted fullerenes
cover a wide range of NMR shifts; they have given us valuable information about the aromatic character of fuller-
enes and the isomers of higher fullerenes. They are also useful in following the many addition reactions which
attach groups to the outside of the fullerene and in following processes which open windows in the cage to release
atoms, e.g. release of noble gas atoms from inside fullerenes.
PROBLEMS
1. Calculate the gyromagnetic ratio and Lande g′ factor 8. FT 13C-NMR signals for some of the molecular systems
for 17O, 25Mg, 39K, 43Ca, 63Cu, 65Cu, 67Zn, 107Ag and 109Ag are listed below. Assign the 13C-resonances to the different
nuclei if their NMR frequencies are observed at 5.772, carbon atoms present in the molecules.
2.606, 1.987, 2.865, 11.285, 12.090, 2.635, 1.722 and
1.981 MHz respectively in a magnetic field of strength
10 kG. mI′s for 17O, 43Ca, 107Ag and 109Ag nuclei are
Molecular
negative. System dc (in ppm)
Formula
[Ans: γ (17O) = 3.628 x 103 G−1 s−1; gN(39K) = 0.2605]
2. A high resolution 100 MHz NMR instrument is capable C5 H8O2 [H2C 18.3, 51.5,
Methyl metha-
of differentiating between frequency differences of 0.01 = C(CH3) 124.7, 136.9,
acrylate
ppm. What will be the resolving power of the instrument? CO2CH3] 167.3
[Ans: 1 × 10−8] 1, 1, 2-tribromo-
C2H3Br3 38.7t, 40.3d
3. Predict the spin decoupled PMR spectra of ethane
CH3—CH2—CH2—CI 118.6, 143.2,
Thiazole C3H3NS
(a) (b) (c) 152.7
if the secondary field is centred on the resonances of 1, 2-dibromo- 24.1q, 37.6t,
(a), (b) and (c) protons respectively. Given Jab = Jbc. C3H6Br2
propane 45.7d
[Ans: (i) a—triplet; b—triplet; c—triplet (ii) a—singlet;
1, 2-dibromo- C4H8Br2 31.8, 44.6
c—singlet; b—twelve line multiplet (sextet under
moderate resolution, Jab ≈ Jbc); (iii) a—triplet; 2-methyl propane
b—quartet; c—triplet] tert-butylalcohol C4H10O 31.2, 68.9
4. (a) Predict the number of signals and their multiplicities
in the FT 13C−NMR spectra of CH3*COOH and *CH3CH2- p-quinone C6H4O2 136.4, 187.1
COOH. [Ans: quartet in both the systems]
122.4, 126.7,
(b) Predict the high-resolution PMR spectrum of Bromobenzene C6H5Br
129.8, 131.4
(i) acetic acid (ii) 1, 2-dimethoxyethane, (iii) ethyl ben-
zene (iv) methyl formate (v) methyl ethyl ketone (vi) form-
aldehyde (vii) thiopropionic acid (viii) acetaldehyde. 9. CW−I3C-NMR spectrum of crotononitrile (C4H5N) exhibits
(c) Predict the high-resolution 1H, 13C, 19F and 31P-NMR a set of four lines at (17.3, 18.8), (100.9, 101.2), (116.0,
spectra of (i) l3CH4, (ii) 13CF4, (iii) 31PH3. 117.6), and (150.2, 151.6) ppm. How do you interpret
5. How many major sets of peaks would you expect in the this spectrum? Also assign these bands to the carbon
PMR spectra of (a) isopropyl alcohol in presence of acid atoms involved in the spectrum.
or base, (b) toluene, (c) propionamide, (d) n-propyl alco- Hint: cis and trans.
hol, (e) biphenyl (f) methanol? Indicate the relative areas
under the major sets of peaks for each compound. What 10. The operating frequency of a NMR spectrometer is
will be the multiplicity of the signals in (c), (e) and (f)? 60 MHz; what magnetic fields are required to bring
3I
P and l9F nuclei to resonance at this operating fre-
6. The resonance frequencies of 13C, 1H, 19F and 31P nuclei quency if the resonance frequencies of 19F and
for 10 kG field are observed at 10.705, 42.577, 40.055 31
P nuclei are observed at 40.055 and 17.235 MHz
and 17.235 MHz respectively. What will be their resonance respectively in an external magnetic field of 10 kG
frequencies for a magnetic field of 2.3487 T? strength?
[Ans: 13C = 25.142 MHz, 1H = 100.000 MHz; 19F = [Ans: 19F = 1.4979 T; 31P = 3.4812 T]
94.077 MHz; 31P = 40.479 MHz]
11. Sketch the PMR spectrum of 1,1,2-trichloroethane
7. A special sample of petrol prepared from tetralin,
given the chemical shifts of—CH2 and —CH protons
naphthalene and n-hexane, exhibits three sets of PMR
as 3.95 and 5.77 ppm (on d scale) respectively and the
signals in the regions of aliphatic, alycyclic and aromatic
spin coupling constant between the —CH2 and —CH
protons respectively. If the areas of the respective sets
protons as 6 Hz. Indicate the relative intensity of the
are of the order of 6.6, 1.6 and 3.6 units, determine the
lines in the first-order approximation.
composition of the petrol sample
Hint: Let x, y and z be the number of moles of tetralin, 12. Predict the structure (s) consistent with each of the
naphthalene and n-hexane respectively. Then the areas following sets of PMR spectral data:
under the bands will be 4x = 1.6; 4x + 8y = 3.6 and 4x + (a) C3H5C13: singlet, d (2.20 ppm), 3H; singlet, d (4.02
14z = 6.6. ppm), 2H
(b) C10H13Cl: singlet, d (1.57 ppm), 6H; singlet, d (3.07 HOH- - -OCMe2 + OCMe2
K
Me2CO - - - HOH - - - OCMe2
ppm), 2H; singlet, d (7.27 ppm), 5H.
(c) C10H14: singlet, d (1.30 ppm), 9H; singlet (7.28 ppm), Table 7.19
5H
v0(Hz relative to TMS)
(d) C9H11Br: quintet, d (2.15 ppm), 2H; triplet, 8 (2.75 nA
ppm), 2H; triplet, d (3.38 ppm), 2H; singlet, d (7.22 36°C 19°C 2°C −14°C
ppm), 5H.
0.99 163.0 169.7 178.0 184.3
(e) C4H9NO: Two peaks at 2.85 and 3.0 ppm and one
peak at 2 ppm. C4H6D3NO: Two peaks at 2.95 and 0.85 157.0 164.4 173.0 179.7
3.1 ppm.
0.68 150.2 157.8 165.3 174.5
Intensities of peaks in both the cases are equal. NMR
data is in 15 per cent CCl4 (2 per cent TMS) IR: 1655 0.59 147.5 163.5 163.5 174.7
cm−1
(f) C7H8O: d (2.43 ppm) 1H; d (4.58 ppm) 2H, d (7.28 (a) Calculate K, ν11 and ν12.
ppm) 5H.
(b) Calculate Δ G0 at 19°C, Δ H0 and Δ S0.
13. The protons of 2,3-dibromothiophene are consistant
[Ans: (a) ν11 = 100, ν12 = 230; K = 0.9 at 36°C; 1.2 at 19°C;
with an AB spin system. The PMR spectrum of the com-
1.5 at 2°C, and 2.0 at −14°C, (b) Δ H0 = −2.4 kcal mole−1,
pound is shown in the following illustration.
Δ S0 = −6.4 e.u; Δ G°292 = + 0.061 kcal mole−1.]
d2 d3
16. How many times is the sensitivity of 2H relative to 1H if g
H
2 = 4.106 × 10−3 G−1 s−1; g H1 = 26.7519 G−1 s−1? Relative
natural abundance: H1 = 100, 2H = 0.0156.
[Ans: 5.6 x 10−7]
d1 d4
17. Calculate the inherent bandwidth of a spectral line in
a liquid and a solid provided the relaxation time of a
ν = 397.63 nucleus in the former is one second and in the latter is
10−4 second.
ν = (in Hz) 5.77
[Ans: For liquid: = 0.1 Hz; For solid = 100 Hz].
19.78
18. Calculate the chemical shifts of individual 13C atoms in
19.75
(a) propane, (b) decane, (c) isobutane, (d) isohexane,
(a) Calculate the frequencies of the lines. (e) 3-methyl pentane, (f) 2,3-dimethyl butane, and
(b) Compute the values of JAB, νA and νB. (g) 2,3-dimethyl pentane.
(c) Calculate the intensity ratio of the inner to outer [Ans: (a) obs: d1 (15.8), d2 (16.3), d3 (15.8)
lines in the spectrum. (b) obs: d1 (14.2), d2 (23.2), d3 (32.6), d4 (31.1), d5 (30.5)
(d) Determine the δ values of A and B. (c) obs: d1 (24.5), d2 (25.4)
(e) Calculate the positions of the lines that will be (d) obs: d1 (22.7), d2 (28.0), d3 (42.0), d4 (20.9), d5 (14.3)
expected if the spectrum is recorded on a 100 MHz (e) obs: d1 (11.5), d2 (29.5), d3 (36.9), d4 (18.8, 3-CH3)
instrument. calc: d1 (11.3), d2 (29.5), d3 (36.2), d4 (19.3, 3-CH3);
[Ans: (a) 410.39, 404.65, 390.64 and 384.87 Hz. (f) obs: d1 (19.5), d2 (34.3);
(b) JAB = 5.70 Hz] (g) obs: d1 (7.0), d2 (25.3), d3 (36.3), d4 (14.6, 3-CH3).]
14. (a) The two signals due to complexed and uncomplexed 19. Calculate the s-l relaxation time if the time taken for
water in the PMR spectrum of 2M AlCl3, solution in a 10 : some hypothetical NMR peak to decay to zero intensity
1 mole ratio mixture of water to DMSO (at −20°C) are 252 is ≈ 7 seconds.
cps apart (operating frequency of instrument = 60 MHz). [Ans: ≈10 s]
The areas of the complexed and bulk water peaks are 20. Systox is a mixture of (EtO)2 PSO—C2H4 SEt, A (thiono);
52.5 and 390 units respectively, Calculate the hydration
(EtO)2 POS—C2H4 SET and (EtO)2 PO—PO (OEt)2.
number of Al (III) and comment on the value so obtained.
[Ans: 3.27] B (thiol) C (Sulfotepp)
(b) The PMR spectrum of 1.75M AlCl3 in H20 at −49°C 60 MHz PMR spectrum of 20 per cent systox sample in
on a 60 MHz instrument exhibits a coordinated water CCl4 exhibits three distinct multiplets, the low field signal
peak to the left (up-field ≈ 124 Hz) of the bulk water due to the OCH2 protons (d = 3.8–4.3 ppm) the mid-field to
peaks. If the areas of the coordinated and uncom- SCH2 (2.3−3.1 ppm) and the high field to C—Me (1.1−1.5
plexed water peaks are 24 and 480 units respectively ppm) and the areas of each are measured by integration.
determine the hydration number of Al (III).
If the areas of the respective peaks are IOCH2 ,ISCH
SCH2 and ICH3
[Ans: 1.5]
and nA, nB and nC are the relative mole fractions of com-
15. The hydrogen bonding of water to acetone, tetra-
pounds A, B and C respectively, show that
hydrofuran and N,N-dimethyl acetamide, present in
large excess in cyclohexane was investigated by PMR 3 1 1
nA = IOCH2 + ISCH2 − IMe
spectroscopy. The chemical shift of the water proton 8 8 4
signal ν, in ternary systems containing water in the
range of mole fractions nw = 0.003 to 0.02 and acetone 1 1
nB = IMe − IOCH2
of mole fraction nA in cyclohexane, were determined at 6 4
several temperatures. The values of ν extrapolated to
zero water concentration, ν0, are given in Table 7.19. 5 3 3
nC = IMe − ISCH2 − IOCH2
Assume that all the water is H-bonded and the process 48 32 32
under consideration is
Hint: Component A has 23. Predict the spectrum of a two-spin system for ν = 60
3 × OCH3 (6 protons) MHz and ν = 270 MHz provided σ A σ B = 24 × 10−8
3 × SCH2 (4 protons) and J = 8.0 Hz. Sketch the spectrum in each case and
comment. [Ans:
3 × C−Me (9 protons)
Component B has 2 × OCH2 (4)
v 1→2 = 60 MHz ∓ 12 2 Hz
3 × SCH2 (6) ( 3→ 4 )
3 × C—Me (9)
Component C has 4 × OCH2 (8) = 270 MHz ∓ 36.6 Hz
4 × CMe (12)
v 1→3 = 60 MHz ± 4.2 Hz
Thus, IOCH2 = 6nA + 4nB + 8nc. (3→ 4)
21. Predict the number of signals and their multiplicities and

assign these to the protons causing them in (a) n-propy- = 270 MHz ± 28.6 Hz
lalcohol in presence of acid or base, (b) isopropyl alcohol
(pure), and (c) tert-butyl alcohol (pure). Relative intensities: 0.35 to 1 at 60 MHz and 0.78 to 1 at
[Ans: (a) 4; triplet, 12 line multiplet, triplet, singlet, (b) 3, 270 MHz. The 270 MHz spectrum looks like a first-order
doublet; 14 line multiplet, doublet, (c) 2; singlet spectrum with J = 8.0 Hz. While the 60 MHz spectrum
(strong), singlet (weak)] looks like a second-order spectrum. The (n + 1) rule
22. Compute the reduced coupling constants for CI and Br holds good for the 270 MHz spectrum but not for the
in PCl3 and PBr3 60 Hz spectrum.]
24. Which of the following groups of elements are most suit-
⎛ J 31P35Cl = 120⎞ ⎛ J31P79Br = 350 ± 17⎞
⎜ ⎟ and PBr3 ⎜ ⎟ respectively. able, depending on their atomic number, mass number
⎝ J 31P37Cl = 100⎠ ⎝ J31P81Br = 380 ± 19⎠ and nuclear spin, for studies by NMR method?
[Ans: 252 × 1020 and 288 ± 14 × 1020 cm−3] [Ans: (c), (e)]
(a) (b) (c) (d) (e) ( f) (g)
Mass number 2n 2n 2n + 1 2n 2n + 1 2n + 1 2n
Atomic number 2n 2n + 1 2n 2n + 1 2n + 1 2n + 1 2n + 1
1 1 1
Spin 0 An integer (2n − 1) (2n + 1) (2n − 1) 0 0
2 2 2
CHAPTER 8 NUCLEAR QUADRUPOLE
“Only God knows the profound and original things. Among men, only those he loves.”
—Plato Timaeus
8.1 INTRODUCTION
Nuclear Quadrupole Resonance (NQR) spectroscopy is a branch of NMR spectroscopy and deals with the absorption
of radio waves by molecules containing quadrupole nuclei, i.e. 2H and 14N(I = 1); 75As, 35Cl, 37CI, 79Br, 81Br, 33S,
63
Cu, 65Cu and 11B (3/2); ,10B (3); 17O and I127 (5/2); 59Co (7/2), etc., in zero magnetic field. Other isotopes like 57Fe
1
are inaccessible in their nuclear ground states since I = , but have a nuclear quadrupole moment in one or more
2
of their nuclear excited states and so can be studied by Mossbauer Spectroscopy (MB) (Chapter 10).
Before describing the basic principles and applications of NQR spectroscopy, the origin of two main terms,
namely nuclear quadrupole moment and electric field gradient involved in it needs to be discussed.
8.2 NUCLEAR ELECTRIC QUADRUPOLE MOMENT

The quadrupole moment ‘Q’ or in multiple of e, i.e. ‘eQ’ is a measure of deviation from spherical symmetry of the
rotating nuclear charge and is generally expressed as
eQ = ∫rnr2(3cos2q − 1) dt (8.1)
where, e is the charge on the electron (= 4.8 × 10 esu), rn is the nuclear charge density in the volume element
−10
dtn (i.e. dx dy dz) at a distance ‘r ’from the centre of the nucleus and at an angle ‘q’ to the nuclear spin axis. Q is
expressed in barns (1 barn = 10−28 m2) and can be either positive or negative.
r 2 cos 2θ
For a sphere, this integral will be zero since the average value of r2cos2q equals but for a prolate
3
spheroid the integral is finite and positive, and negative for an oblate spheroid, because of the presence of the fun-
damental unit of charge on both sides of Eq. (8. l), itself has the dimensions of r2, viz. cm2. Thus, for an elongated
(i.e. prolate rotational ellipsoid) nucleus along the axis of the nuclear spin, the sign of ‘eQ’ is positive [Fig. 8.1(a)]
whereas negative for a flattened (i.e. oblate rotational ellippsoid) nucleus along the axis of the spin [Fig. 8.1(b)].
Nuclear Spin Axis Zn
Z
q
+ q
−
τ +
d n
− − + O r Yn
−
+
Xn
(a)
dτn
(b)
Fig. 8.1 The origin of nuclear electric quadrupole moments through deformation of a (rotating)
charged spherical nucleus (a) elongated nucleus with positive ‘eQ’, and (b) flattened nucleus with
negative ‘eQ’.
Nuclear Quadrupole Resonance Spectroscopy 611
Further, atomic beam as well as proton scattering experiments show the existence of electric hexadipole moment
also. However, the magnitude of a hexadipole moment is 10−8 of the electric quadrupole moment.
8.3 ELECTRIC FIELD GRADIENT (EFG)

A point charge ‘q’ at a distance r = (x2 + y2 + z2)1/2 (in the arbitrarily chosen X, Y and Z axes which are different
q
from the coordinate axes for Q; Fig 8.2) from the nucleus gives rise to a potential U (r) = at the nucleus. The
electric field gradient at the nucleus is given by r
U xx U xy U xz
EFG = ∇E = ∇ U = − U yx U yy U yz
2
(8.2)
U zx U zy U zz
where E is the electric field gradient at the nucleus (E = ∇U), ∇ is the Laplacian operator
∂ 2U
U ij = (i j = x y , z ) (8.3)
∂i∂j
Since electric field gradient is symmetrical, so
Uxy = Uyx; Uxz = Uzx; Uyz = Uzy (8.4)
The principal axis of MB nucleus is selected in such a way that the off-diagonal elements are zero i.e.,
Uxy = Uyx = Uzx = 0
and the diagonal elements are ordered as
|Uzz| ≥ |Uyy| ≥ |Uxx| (8.5)
With respect to principal axis, EFG is described by two independent parameters choosen as Uzz = eq and the
symmetry parameter h, defined by
( − )
η=
yy
(8.6)
U zz
and is a measure of the departure from the axial symmetry. ‘eq’ can theoretically be obtained from the equation
eq = ∫ ϕ * ⎡⎣(3 2
− ) / r 3 ⎤⎦ ϕd τ
where j is the appropriate wave function.
With the above ordering of the diagonal elements, the asymmetry parameter is restricted to 0 ≤ h ≤ 1. For
a four-fold or three-fold axis (120° rotation) passing through the nucleus (i.e. choosing these axes as Z-axis)
Uxx = Uyy and therefore h = 0 and the field is called axially symmetric. Moreover, it can be shown that two mutually
perpendicular axes of three-fold or higher symmetry give rise to a vanishing EFG. In general, the following three
factors contribute to the total EFG: (i) noncubic charges on electrons belonging to the MB atom, i.e. noncubic
electron distribution in partially filled valence orbitals of the atom, e.g. in transition metal ions, the inner d or f
shells are not completely filled and produce EFG at the nucleus. The effective contribution due to this phenomenon
is given by (Uzz)val (1 − R) where R is called the Sternheimer shielding factor and has been estimated to be 0.2 − 0.3
Z ′ (Zn)
Z Symmetry Axis
H
G
EF
X ′ (Xn)
j
Y
X
Y ′ (Yn)
Fig. 8.2 Coordinate axes with respect to magnetic field axis and electric
field gradient axis.
for iron and tin (ii) charges external to the MB atom or non-MB atom in noncubic symmetry. Such a contribution
is termed ligand/lattice contribution. The effective contribution is given by (Uzz )L (1 − g∞ ); where
n
) ∑q r
i
i i
−3
( ccos 2 θi − 1),
n
η /( )L ∑ q 3
i 2θi cos2ϕ i ,
i
If the distances ri and the angles fi and qi of all n ions i with respect to an atom are known from crystal-structure
determination, and if effective charges qi can be assigned to the parent ion then (Uzz)L and hL can be computed.
g∞ is called the Sternheimer anti-shielding factor; (1 − g∞) has been estimated to be in the order of 10 for iron
compounds. However, (Uzz)L is not the EFG contribution felt by the nucleus, and (iii) polarisation of the filled
inner shells of the atom. The filled inner shells being spherically symmetrical normally do not contribute to the
EFG, However, charges because of ligand/lattice and valence electrons cause polarisation of filled inner shells
which then contribute to the EFG at the nucleus and is known as polarisation contribution, i.e. (Uzz )pol.
In general, the Z-component of the total effective EFG at the nucleus may be written as
U zz (1 R )(U zz )val (1 ∞ )(U zz )L + (U zz )pol (8.7)
and the asymmetry parameter is given by
ηU zz U xx U yy (1 R ) (U zz )val ηval ( γ ) (U zz )L ηL + (U zz )pol η poll (8.8)
For all practical purposes, polarisation contribution being small is ignored. Similar expressions are also there
for the other two diagonal EFG elements.
The relative contribution of these factors depends upon the type of solid, e.g. the lattice contribution will pre-
dominate in ionic crystals or nontransition elements whereas the valence electron contribution will be dominant
in transition elements. In molecular solids, the asymmetric p-electrons or the presence of p-bonding contribute
mainly to the EFG.
The EFG elements and h, for electrons of different angular momenta, can be worked out from the above typed
expressions. Since in most chemical applications only d- and p-electrons are considered, the contributions to
1
(Uzz)val from these electrons expressed in the units of expectation value of 3 , i.e. <r −3> have been reported in
r
Table 8.1 where < > refers to integral over all space. For some cases h >1 the ordering (8.5) is not obeyed. In
Table 8.1 The magnitude of (EFG)val and h for the p- and d-electron orbitals.
Orbital (Uxx)val (Uyy)val (Uzz)val h
e <r−3> e <r−3> e <r−3>
px −4 +2 +2 −3
5 5 5
+2 −4 +2
py +3
5 5 5
+2 +2 −4
pz 0
5 5 5
−2 −2 +4
dxy 0
7 7 7
−2 +4 −2
dxz +3
7 7 7
+4 −2 −2
dyz −3
7 7 7
−2 −2 +4
dx 2 y2 0
7 7 7
+2 +2 −4
dz2 0
7 7 7
1
order that (8.5) is followed, the principal axis system should be chosen properly. The expectation value, , i.e.
r3
<r −3> = ∫R(r)r −3 R(r)r2 dr cannot be calculated precisely but may be obtained from experiments. The order of
expectation value, <r −3> for various orbitals is 4p < 3p and 3d, 3d < 3p. This is because of the greater radial extent
of the former orbitals relative to the latter. Similarly, we can interpret the sign of q in the X- and Y-directions also.
The total value of q for a completely filled or half-filled shell of electrons is zero. An excess of electrons in the
Z-axis (occupying pz, d z 2, dxz or dyz) gives a positive value of q whereas an excess in the XY plane (px, py, dxy or d x 2 y 2
) results in negative value of q.
It is to be noted that spherically symmetric s-electron gives no EFG. f electrons being closer to nucleus than
d-, p- and s-valence electrons of a particular complex ion are shielded by them from the influence of the ligand
hence rarely participate in chemical bonding. Consequently, f orbitals do not undergo ligand field splitting and,
therefore, electrons in the degenerate f-orbitals set do not contribute towards EFG.
8.4 NUCLEAR ELECTRIC QUADRUPOLE COUPLING

Nuclear quadrupole coupling constants in atoms were first observed in 1935 by Z Schmidt and H Schuler and for
the first time in molecules (HD and D2) by JMB Kellog, II Rabi, NF Ramsey and JHZacharias in 1940 by atomic
and molecular beam spectroscopy. However, with the discovery of pure nuclear quarupole resonance by H G Deh-
melt and H Kruger in 1951, the nuclear quadrupole coupling constants of molecules became readily accessible
experimentally.
In 1934, Harold Clayton Urey (USA) received the Nobel Prize in chemistry for the discovery of deuterium and
in 1944 Isidor Isac Rabi (USA) got the Nobel Prize in physics for resonance method for recording the magnetic
properties of atomic nuclei.
The interaction between nuclear electric quadrupole moment ‘eQ’ and electric field gradient ‘eq’ (i.e. Uzz)
is termed as nuclear quadrupole coupling and is measured in terms of a quantity ‘vQ’ called nuclear quadrupole
coupling constant. It is expressed as
e 2 qQ
vQ = (8.9)
h
vQ which contains the product of q and Q is measured experimentally. From a reliable theoretical value of q for
an atom, we can derive from Eq. (8.9), the value of Q for the nucleus of that atom. We may then use this value to
derive ‘q’ in compounds of the atom and compare with values calculated quantum mechanically.
8.5 ORIGIN OF NUCLEAR ELECTRIC QUADRUPOLE COUPLING

In quadrupole nuclei, there are two antiparallel dipoles. In an electric field, each of these dipoles senses a torque
tending to align them with the field. However, since the two turning torques are equal and opposite, there is no net
effect. On the other hand in an electric field gradient which is assumed to be axially symmetric about Z-axis, the
torques on two dipoles are not equal and the net turning torque exists and is proportional to EFG (= −eq, along Z
axis) and quadrupole moment ‘eQ’ i.e. to the magnitude | e2qQ |.
The nuclear electric quadrupole coupling causes a splitting of the Zeeman energy levels. Further depending
on the magnitude of the nuclear quarupole interactions relative to Zeeman energy, the field of nuclear quadrupole
interactions has been roughly divided into two areas, viz., combined magnetic and quadrupole interactions, and
pure quadrupole resonance.
8.6 COMBINED MAGNETIC AND QUADRUPOLE INTERACTIONS

When both magnetic and quadrupole interactions (both are directional effects) exist simultaneously, their
respective principal axes may or may not be in the same direction because of which the resulting behaviour
may be complex. Consequently, the general interpretation of the spectrum can be quite complex. The combined
effects may cause a change in line positions and/or in the relative line intensities or width of the hyperfine
pattern. The combined effects can operate in two ways, either the magnetic hyperfine interaction, i.e. Zeeman
splitting is perturbed by an electric quadrupole interaction or the quadrupole interaction is perturbed by mag-
netic dipole interaction, for instance by an applied external magnetic field. In case the quadrupole interaction
is very weak relative to magnetic hyperfine interaction (i.e. e2qQ << mH), the quadrupole coupling in such a
situation is taken as perturbation and causes a splitting in the NMR resonance signal into several components.
Quadrupole interactions in such systems are studied by NMR techniques.
The Hamiltonian operator of a nucleus with quadrupole moment ‘eQ’ in the presence of applied magnetic field
and nuclear quadrupole coupling is given by
HQM = HM + HQ
e 2QU zz ∧ ∧ ∧ ∧
= − g β HI
H + I z2 − I 2 + η I +2 + I −2 ) / 2 ] (8.10)
4II ( I − 1)
∧ ∧ ∧ ∧
where, I z is the z-component of the nuclear spin operator I and I + and I − are the raising and lowering opera-
tors defined as
Iˆz Iˆx ± iIˆy (8.11)
The eigen value of the operator Iˆ 2 is I(I + 1).
It is to be noted that in a high magnetic field the spin is quantised in the direction of the field. Let Z ′-axis be the
direction of the applied magnetic field (Fig. 8.2). If EFG has cylindrical symmetry, i.e. h = 0, the symmetry axis
of the EFG, i.e. Z-axis differs from Z′–axis (Fig 8.2). If X′-axis is in the plane containing Z and Z ′ then
Iz = Iz ′ cosq + Ix ′ sinq (8.12)
The Hamiltonian operator with h = 0 will then appear as
e 2qQ
H QM g N β N HI + (3Î 2z −Î 2 ) (8.13)
4I ( 2I − 1)
Squaring Eq. (8.12), we obtain
2
z = I z2 cos 2 θ + I x2 si 2 θ ( I z I x + I x ′ I z ′ ) sin θ cos θ
Substituting the value of I z2 into Eq. (8.13), we get
e 2qQ
H QM g N βN HI + .
4I ( 2I − 1)
⎡ 2 2 ⎛ 3 cos 2 θ − 1⎞ 3 3 2 ⎤
⎟⎠ + 2 sin θ cos θ {I z I + I I + I I z } + sin
i θ( 2 2
⎢3I z I ⎜ + )⎥ (8.14)
⎣ ⎝ 2 4 ⎦
Here, Ix′ is expressed in terms of I+ and I−.
The eigen values of the operator HQM, i.e. the energy of the sublevels as deduced from the standard perturbation
theory are given by
3e 2qQ ⎡ 2 1 ⎤ ⎡ 3 cos θ −1⎤ ...
2
EQM g N βN HM I + M − I ( I + 1) ⋅
⎥ ⎢ ⎥+ (8.15)
4I ( 2I − 1) ⎢⎣
1
3 ⎦ ⎣ 2 ⎦
where first and second terms represent the zeroth, and first-order contributions to the energy. The second and
higher-order contributions to the energy have been ignored. The terms and symbols in Eq. (8.15) have their usual
meanings. q is the angle between magnetic spin axis and the major axis (Z-axis) of EFG.
8.6.1 Shift in Zeeman Energy Levels Due to First-order Nuclear

Quadrupole Coupling
Let us now work out the shift in Zeeman energy levels due to first-order quadrupole interactions, for quadrupole
nuclei with different spins.
1
(a) For nuclei with I = , the number of substates due to Zeeman effect/first term of Eq. (8.15) is two according
2
to (2I + 1) rule. These substates are characterised by MI = ± 1 . The second term of Eq. (8.15) which is because of
2
1 1
first-order nuclear quadrupole interaction is zero for MI = ± . Therefore, for a nucleus with I = , the first order
2 2
quadrupole interaction is zero. Consequently, the Zeeman substates are not affected due to first order nuclear
quadrupole coupling [Fig. 8.3 (a)].
First Order
MI = −1/2
I = 1/2
H=0
MI = +1/2
Zeeman Splitting Magnetic and
H not Equal to Quadrupole
Zero Interactions
(a)
First Order
MI
−3/2
e2qQ
4
e2qQ
I=3 4 –1 Second Order
2 2
H=0
+1
2
Magnetic + +3/2
Zeeman
splitting quadrupole
(H ≠ 0) interactions
(b)
Fig. 8.3 Perturbation of Zeeman levels by the first-order quadrupole interaction in a
nuclear system with symmetry axis parallel to spin axis for (a) I = 1/2 (b) I =3/2.
3
(b) For nuclei with I = , the number of equally spaced sublevels due to Zeeman energy is four according to
2 3 1
(2I + 1) rule. These four substates are characterised by MI = ± , ±
2 2
3
Since, I = , so Eq. (8.15) reduces to
2
e 2qQ ⎡ 2 5 ⎤
EQM g N βN HM I + M − 2
θ − 1}
4 ⎣ 1 4 ⎥⎦
1 ⎛ 2 5⎞ 3 3
For MI = ± , the term ⎜⎝ M1 − ⎟⎠ = −1 and +1 for MI = ± . Therefore, Eq. (8.15) for I = can be expressed as
2 4 2 2
|M I |+
1
⎛ e 2qQ
Q ⎞ ⎛ 3 cos 2 θ − 1⎞
EQM g N βN HM I + ( − ) 2
.⎜ (8.16)
⎝ 4 ⎟⎠ ⎜⎝ 2 ⎟⎠
3
For MI = ±
2
⎛ 3⎞ 3 ⎛ e 2qQ
Q ⎞ ⎛ 3 cos 2 θ − 1⎞
EQM g N βN H + ⎜ (8.17a)
⎝ 2⎠ 2 ⎝ 4 ⎟⎠ ⎜⎝ 2 ⎟⎠
1
and for MI = ±
2
⎛ 1⎞ 1 ⎛ e 2qQ
Q ⎞ ⎛ 3 cos 2 θ − 1 ⎞
EQM g N βN H − ⎜ (8.17b)
⎝ 2⎠ 2 ⎝ 4 ⎟⎠ ⎜⎝ 2 ⎟⎠
3
Equations. (8.17a) and (8.17b) reveal that the substates with M I = ± will be raised by an amount
⎛ e 2 qQ 2
Q ⎞ ⎛ 3 cos 2 θ − 1 ⎞ 1
⎜⎝ 4 ⎟⎠ ⎜⎝ ⎟⎠ while those with M I = ± will be lowered with the same amount as shown in Fig (8.3 b).
2 2
1
Further, for cos q = , there is no shift in Zeeman substates due to first-order nuclear quadrupole coupling
3 −1 π
whereas the quadrupole coupling term will be in error by a factor of for q = . Therefore, without the inde-
2 2
pendent knowledge of the direction of the magnetic field relative to the symmetry axis of the field, the quadrupole
moment can’t be defined by the hyperfine field alone.
If the symmetry axis is parallel to the magnetic field H, i.e. q = 0, Eq. (8.16) becomes
|M I |+
1
⎛ e 2qQ ⎞
⎜⎝ 4 ⎟⎠
3
Consequently, the Zeeman level corresponding to MI = ± will be raised whereas lowered corresponding to
2
1 ⎛ e 2 qQ ⎞ ⎛ e 2qQ
Q ⎞ ⎛ 3 cos 2 θ −1 ⎞
MI = ± by an amount ⎜ rather than ⎜⎝ 4 ⎟⎠ ⎜⎝ ⎟⎠ Fig. (8.3b).
2 ⎝ 4 ⎟⎠ 2
It is to be noted that the treatment of second-order perturbation is beyond the scope of this book. However the
influence of second-order perturbation on the Zeeman levels has been shown in Fig. 8.3 b.
Further it is evident from Fig. (8.3 b) that in the presence of quadrupole interaction, the energy levels are no
longer symmetrical about the centroid.
8.6.2 Transition between Zeeman Energy Levels Perturbed

by First-order Quadrupole Interaction
We know that the energy of Zeeman levels perturbed by first-order quadrupole interaction is given by
3e 2qQ ⎡ 2 1 ⎤
EQM g N βN HM I + M − I (I + 1) ⎥ (8.18)
4I ( 2I − 1) ⎢⎣ 1 3 ⎦
provided the angle between the symmetry axis (i.e. Z-axis) and spin axis is zero.
(a) For nuclei with I = 1 , the number of magnetic substates is two according to the (2I + 1) rule. These states
2
1
are characterised by MI = ± 1 . Further for I = the second term in Eq. (8.15) vanishes, consequently the energy
2 2
expression resembles with that of an unperturbed Zeeman energy, i.e.
EQM = −gN bN HMI (8.19)
1
Since, MI = ±
2
Therefore,
⎛ 1⎞ 1
E QM g β H (8.20)
⎝ 2⎠ 2 N N
⎛ 1⎞ −1
E QM g β H (8.21)
⎝ 2⎠ 2 N N
Thus, according to the selection rule ΔMI = ±1, the transition frequency between the two states characterised
1
by MI = ± is given by
2
⎛ +1 −1⎞ ⎛ −1⎞ ⎛ +1⎞
ΔE ⎜ → ⎟ = EQM ⎜ ⎟ − EQM ⎜ ⎟
⎝ 2 2⎠ ⎝ 2⎠ ⎝ 2⎠
= gNbNH = hv0 (8.22a)
or
gN βN H
v0 = (8.22b)
h
1
Thus, the NMR spectrum of quadrupole nuclei with I = consists of a single line having frequency
gN βN H 2
v0 = and is independent of the first order nuclear quadrupole coupling. The zeroth order spectrum is
h
shown in Fig 8.4 (a).
3
(b) For quadrupole nuclei with I = , the energy of Zeeman levels perturbed by first order nuclear quadru-
pole interaction has been expressed as 2
1
|M I |+ e 2qQ
⋅ (Eq. 8.18)
4
3 1
The energy of four substates characterised by MI = ± , ± can be expressed as
2 2
⎛ 3⎞ 3 e 2qQ
EQM g N βN H + (8.23)
⎝ 2⎠ 2 4
⎛ 1⎞ 1 e 2qQ
EQM g N βN H − (8.24)
⎝ 2⎠ 2 4
The frequencies of NMR lines corresponding to the permissible transitions by the selection rule ΔMI= ±1 as
deduced from Eqs. (8.23) and (8.24) are given by
⎛ +1 −1⎞ g β H ⎫
v0 ⎜ → ⎟ = N N ⎪
⎝ 2 2⎠ h
⎪
⎛ −1 −3 ⎞ e 2
qQ ⎪
v1 → = v0 + ⎬ (8.25)
⎝ 2 2⎠ 4h ⎪
⎛ 3 +1⎞ e 2qQ ⎪
v2 ⎜+ → ⎟ = v0 − ⎪
⎝ 2 2⎠ 4h ⎭
The frequency separation between two adjacent peaks, i.e. quadrupole splitting, Δ, is given by
e 2 qQ
Δ= (8.26)
4h
3
Intensities of various lines are proportional to < MI | Ix | MI − 1>|2 and are in the ratio 3 : 4 : 3 for I = .
2
The set of Eqs. (8.25), shows that the NMR spectrum consists of three resonance lines with frequencies v0, v1
and v2. The lines with frequencies v1 and v2 are termed as satellite lines and are on the opposite side and equidistant
from the central line. The spectrum is shown in Fig. 8.4 (b).
(c) For quadrupole nuclei with I = 5/2, the number of substates are six by (2I + 1) rule. The six substates are
5 3 1
characterised by MI = ± , ± , ± .
2 2 2
Further, for I = 5/2, the Eq. (8.15) can be expressed as
3e 2qQ ⎡ 2 35 ⎤
EQM g N βN HM I + M − (8.27)
40 ⎣ I 12 ⎥⎦
The energies of six substates characterised by MI = ±5/2, ±3/2, ±1/2 can then be deduced from Eq. (8.27) and
are given by
⎛ 5⎞ 5 5e 2qQ
EQM g N βN H +
⎝ 2⎠ 2 20
⎛ 3⎞ 3 e 2qQ
EQM g N βN H − (8.28)
⎝ 2⎠ 2 20
⎛ 1⎞ 1 4e 2qQ
EQM g N βN H −
⎝ 2⎠ 2 20
The frequencies of NMR lines corresponding to the allowed transitions by the selection rule MI = ±1 can be
deduced from the set of Eqs. (8.28) and are given by
⎛ −3 −5 ⎞ 3e 2 qQ
v2 → = v0 +
⎝ 2 2⎠ 10h
⎛ −1 −3 ⎞ 3e 2 qQ
v1 → = v0 +
⎝ 2 2⎠ 20h
⎛ +1 −1⎞ g β H
v0 ⎜ → ⎟ = N N (8.29)
⎝ 2 2⎠ h
⎛ +3 +1⎞ 3e 2 qQ
v3 → v0 −
⎝ 2 2⎠ 20h
⎛ +5 +3 ⎞ 3e 2 qQ
v4 → v0 −
⎝ 2 2⎠ 10h
The set of Eqs. (8.29), shows that the NMR spectrum consists of five resonance lines with frequencies v2, v1,
v0, v3 and v4. The lines with frequencies v2 and v1 and v3 and v4 are on the opposite side and equidistant from the
central line. The quadrupole splitting for I = 5/2 is 3e2q Q/20.
The NMR lines with frequencies v2, v1, v0, v3, and v4 are in the intensity ratio 5 : 8 : 9 : 8 : 5. The NMR spectrum
for nuclei I = 5/2 is shown in Fig. 8.4 (c). The second-order perturbation shifts the central line in the spectrum.
⎡ 1 3 5 ⎤
Further, it is evident from Fig. 8.4 that the numbers of NMR lines in the spectra of nuclei with ⎢ I = , , ,…⎥
⎣ 2 2 2 ⎦
are given by 2 I, Thus the spin of a nucleus can be obtained from the number of lines in the NMR spectrum, e.g.
1
if the number of lines in the NMR spectrum is only one, the spin of the nucleus will be .
2
The energy expression for the cylindrically asymmetric systems is given by
3e 2qQ ⎡ 1 ⎤
EQM g N βN HM I + 2
+ s θ cos 2 ϕ ) M 12 I (I + 1) ⎥ (8.30)
8II ( 2I − 1) ⎣ 3 ⎦
I=1
2
ν0
(a)
I=3
2
ν2 ν0 ν1
(b)
I=5
2
ν4 ν3 ν0 ν1 ν2
(c)
1 3 5
Fig. 8.4 Schematic NMR spectra for nuclei with (a) I = (b) I = and (c) I = in the presence of zeroth, first and second-order qua-
2 2 2
drupole effects. The central line is not affected by the first-order quadrupole coupling. The dashed line shows the second-order shift of the
central line.
If we use single crystals that can be rotated with respect to the applied field, it is possible from the spectral
shapes for different orientations to determine the positions of the principal exes of the EFG in the crystal and the
values of the parameters h and e2qQ.
Problem 8.1: If the EFG is axially symmetric, prove that for I = 3/2 the frequency separation between two
adjacent lines, Δ, in an NMR spectrum showing first order quadrupole effects is given by
3 2qQ ( cos 2 θ − 1)
Δ Δ(θ ) =
4 hI ( − )
Solution From Eq. (8.15), we can easily deduce the required result as follows.
g N βN H 3e 2qQ
v1 = + ⋅ (3 c 2
θ 1)
h Ih ( 2I − 1)
4Ih
g N βN H
ν0 =
h
g N βN H 3e 2qQ
v2 = − ⋅ (3 c 2
θ 1
h Ih ( 2I − 1)
4Ih
Thus,
Q(( cos 2 θ − 1)
3e 2qQ
qQ(
Q
Δ = Δ (θ ) = v 1 − v 0 = v 2 − v 0 = (8.31)
4 hI ( − )
For a crystalline or powder sample where all directions of the EFG are equally probable, NMR absorption
occurs over a wide range, i.e. 3cos2q − 1 = 1, cos q = 2 or q = 35.3° or 144.7°, the powder sample behaves like
NMR liquid. 3
The quadrupole splitting between adjacent peaks is given by
3e 2 qQ
Δp = (8.32)
hI ( 2 I − 1)
4hI
8.7 QUADRUPOLE SPLITTINGS IN LIQUID CRYSTALS

NMR studies of quadrupole ions or molecules in liquid crystals are used to obtain information pertaining to
molecular geometry and EFGs. Order parameter (S) using known value of vQ and observed value of quadrupole
splitting can also be determined.
In quadrupole nuclei, all directions of the EFG with respect to the applied magnetic field are considered to be
equally probable. However, a slight anisotropy in the EFG may turn simple natured NMR signal into a complex
pattern, which is analogous to that observed in all solids. In liquid crystalline systems, a rapid molecular motion
also tends to average out the quadrupole interaction. However, for an anisotropic mesophase, complete averaging
out does not result. The residual interaction gives rise to NMR effects analogous to those obtained for rigid lattices
except that quadrupole splittings are generally much smaller for liquid crystals, because of the rapid molecular
motion.
For uniaxil liquid crystals like nematic mesophases, lamellar and hexagonal amphiphilic phases, the system is
cylindrically symmetric with respect to an axis called director. It is convenient to perform the transformation from
the molecular coordinate system (M) to the laboratory frame (L) via the director coordinate system. The M−D−L
(Z) transformation in the mesomorphous structure in the lamellar phase is shown in Fig. 8.5.
For a sample, (with I = 3/2) with uniform director orientation, the quadrupole splitting is given by
3e 2qQS
Δ( ) = c 2 θ LD − 1)
⋅ ( cos (8.33)
4 hI ( − )
Z D
L qDM
M
qLD
Fig. 8.5 The outline of the transformation of coordinates from the molecular to laboratory via the director coordinate system. qLD and qDM
are angles between Z-axis in the laboratory-director and director-molecular systems respectively.
For a powder sample, all values of qLD are equally probable and for qLD = 90°, the quadrupole splitting, Δp, is
given by
3e 2 qQS
Δp = (8.34)
hI ( 2 I − 1)
4hI
Further, the study of liquid crystalline phases is built up of amphiphilic compounds, e.g. surfactants and water
is important since phases of this type are important building stones in biological systems, i.e. membranes. The
counter cations in amphiphilic water mesophases can reside in different binding sites which are characterised by
different quadrupole splittings. In most cases, it is assumed that the rate of exchange of counter cations between
different positions is rapid than the difference between the quadrupole splittings between the different sites.
However, if the exchange is slow, the first-order quadrupole splittings are given by
3Pv
Pi Q ( )S i e 2qiQ
Δ ( ) = ( ccos 2 θ LD − 1)Σ ;v Q (i ) = (8.35)
4 ( − ) h
and
3Pv
i Q ( i ) Si
Δp = Σ (8.36)
4I ( 2 I − 1)
where, Pi is the probability for occupation of site i characterised by the field gradient qi, and the order param-
eter Si.
The observed quadrupole splitting, Δp, of Cl35 in the hexagonal mesophase of the dodecyltrimethyl ammonium
chloride ( +
)3 Cl − ) water system decreases as the concentration of surfactant is increased whereas the
reverse has been observed for 37Cl and 35Cl in the lamellar mesophases of the C8+NH3Cl−— H2O and (C8+NH3Cl−—
C10OH—H2O) systems respectively. The hexagonal mesophase is built up of rod-shaped amphiphilic aggregates
arranged in a hexagonal array whereas the lamellar phase is built up of alternating amphiphilic bilayers and water
layers. The latter gives rise to an asymmetric splitting of the 35Cl and 37Cl NMR signals. It is to be noted that these
substances behave as electrolytes in dilute aqueous solutions rather than mesophases.
Problem 8.2: Concentrated solutions of poly-g-benzyl-L-glutamate behave as liquid crystal of cholesteric type
and when placed in a magnetic or electric field, the cholesteric structure breaks up to give nematic phase. 35Cl—NMR
spectrum of CH2Cl2 in poly-g-benzyl-L-glutamate (10 CH2Cl2 molecules/monomer residue) at 24°C shows quadru-
pole splitting of 56 kHz. Calculate the order parameter provided vQ(CH2Cl2) = 72 MHz.

3 2 qQS ( cos 2 θ LD − 1)
Δ( ) = (Eq. 8.33)
4hI ( − )
3(3 cos 2 θ LD − 1) .vQ
=
(2 − 1)
4 (2
Rearrangement of Eq. (8.33) gives
4Δ (θ )I ( I − )
S=
3v Q ( cos 2 θ LD − 1)
Since the sample is under the influence of magnetic field, so qLD = 0.
Thus,
4 ΔII ( 2 I − 1)
S11 =
6 vQ
3
Substituting the values I = Δ = 56 kHz and vQ = 72 MHz, we get
2
3 3
4(56 103 ) ( 2 × − 1)
S11 = 2 2
6(72 106 )
= 1.55 × 10−3
8.8 NUCLEAR QUADRUPOLE RELAXATION

Nuclear quadrupole relaxation occurs through time dependent s-s and s-l interaction whereas quadrupole relax-
ation is due to time dependent electric field gradients at the nucleus. Both the magnitude and direction of EFG
play an important role in the quadrupole relaxation process, e.g. the changes in magnitude of EFG contributes to
the relaxation rate in halide ions whereas in covalent bonded halogens, the contribution to relaxation rate is due
to changes in direction of EFG caused by molecular tumbling.
The relaxation of quadrupole nucleus is in general expressed in terms of two (for 35Cl, 37Cl, 79Br, 81Br) and
three (for 127I) exponentials. However, if the frequency of modulation of quadrupole interaction is rapid relative
to Larmour frequency of the nucleus, the relaxation mechanism may be expressed by a single exponential, e.g. in
halogens. In such cases, the longitudinal and transverse relaxation rates, i.e. R1 and R2 respectively, become equal
and are given by
2
3π 2 ( 2I 3) ⎡ e 2qQ ⎤ ⎛ η 2 ⎞
R1 R 2 = ⋅⎢ ⎥ 1 + ⎟ τc (8.37a)
10 I 2 ( 2I 1) ⎣ h ⎦ ⎜⎝ 3⎠
Here, tc is the correlation time characterising the reorientation of the field gradient, i.e. molecular reorientation
e 2 qQ
around symmetry (i.e. EFG) axis. In terms of coupling constant, vQ= ,
h
3π 2 ( 2I 3) ⎛ η2 ⎞
R1 R2 = ⋅ v Q2 1 + ⎟ τc
1+ (8.37b)
2
10 I ( 2I 1) ⎝ 3⎠
1
The above expression is obeyed provided v L τ c ≥ otherwise fails for quadrupole halogen nuclei as well as
for any magnetic nuclei with I > 1. 2π
The above expression can be used to determine ‘t’ provided h, vQ and R1 or R2 are known. For axial symmetric
molecules, h = 0. However, if the halogen is attached to an unsaturated carbon, the partial double bond character
leads to loss of symmetry about the carbon halogen bond and h ≠ 0. For 35Cl, h = 0.25.
For Lorentzian shape NMR signals the values of the spin relaxation rates are expressed as:
R2 1
Δv 1 (Full width at half height) = = (8.38a)
2
π π T2
R2 1
ΔvPP (peak-to-peak line width of the derivative absorption curve) = = (8.38b)
3π 3ππ T2
The line broadening may also be either due to the dominance of correlation time or of ion exchange effects over
the quadrupole relaxation on broadening due to T1 or T2.
Equation (8.37) holds good only for rigid molecules. On the other hand, if the nucleus resides in a group which
undergoes internal rotation relative to the rest of the molecule, the rate of relaxation may depend on both the internal
rotation and on the rate of the rotation of the whole molecule. For isotropic motion of the whole molecule
3p 2 ( 2I + 3) 2 ⎡ 1 1 ⎞ ⎤
−1
⎧ 1 2⎫⎛ 1
R1 = v Q ⋅ ⎢ (3 c b 2 2
t rot ⎨1 − (3 cos b -1) ⎬ ⎜
2
+ ⎟ (8.39)
10I 2 ( 2I 1) ⎢⎣ 4 ⎩ 4 ⎭ ⎝ t rot t introt ⎠ ⎦
Here, b is the angle between the largest component of the field gradient and the axis about which the internal
rotation takes place
If τ introt << τ rot , (8.39) reduces to
2
3π 2 ( 2I 3)v Q 1
R1 = . (3 cos 2 β − τ rot
2
(8.40)
10 I 2 ( 2I 1) 4
Therefore, the effect of fast internal rotation is to average out components of field gradient which are perpendicular
to the axis of internal rotation.
The relaxation rates depend on temperature and increase with the increase of temperature. The effect is because
of increase in the acceleration of molecular reorientation with the growth of temperature and is given by
R1, 2 e + Ea / RT (8.41)
The relaxation rate of 127I− in aqueous solution of MI (M = Li, Na, K) varies with temperature approximately
as h/T (h denotes viscosity).
In general, the effect of temperature on quadrupole coupling constant is negligible. Therefore, the parameter tc,
i.e. correlation time is used to determine Ea.
1
∝ e + Ea / RT (8.42)
τc
8.8.1 Quadrupole Relaxation Models

The two mathematical models to predict the quadrupole relaxation times are the following:
(a) Electrostatic Model This model in due to H G Hertz. The model is based on the assumption that the EFGs
at the nucleus arise because of the distribution of charges set up by the surrounding species. For infinite dilute aque-
ous solutions, the expression for the randomly oriented water molecules around the relaxing monoatomic ion is
2 ⎛ C H Oτ H O ⎞
2
24π 3 ( 2I 3) ⎛ eQ ⎞
R1 R2 = ⎜⎝ ⎟⎠ ⎡⎣ M H2 O (1 γ ∞ )P ⎤⎦ ⎜ (8.43)
2 2
5I ( 2I − 1)
2 5 ⎟
h ⎝ r0 ⎠
2 3
where, P (polarisation factor) = is assumed to be 0.5 (D is the dielectric constant of the medium), e = 4.8 ×
5D
10 e.s.u. M H2 O is the dipole moment of water molecule and its value in the gas phase (M = 1.84 D = 1.84 ×
−10
10−18 e.s.u cm) is used, CH2 O is the water concentration in molecules/cm3. (1 + g∞) is the Sternheimer antishielding
factor and r0 is the distance between the relaxing nucleus and the water molecules in the first hydration sphere. It
is taken as the sum of the ionic radius and radius of the water molecule.
1 1 1
= + ,τ r is the time constant for the reorientation of water dipoles relative to the ion–water molecule
τ H2O τ r τ*c
vector direction and τ c∗ is the time constant for the reorientation of water dipoles relative to the rotational motion
axis of the ion–water molecule. The time variation of field gradient is either because of tr or τ c∗.
Further according to simple electrostatic model, in case of amphiphilic liquid crystals, the EFGs are assumed
to be due to charges of the amphiphilic ions and to the water dipoles. The quadrupole coupling constant arising
3
from point charge Z at a distance r from a nucleus with spin I = with quadrupole moment Q is given by
2
e 2qQ 0.69 × 106 ( H )( γ )QZ ⎡ 2D + 3 ⎤
vQ = = ⎢ 5D ⎥ (8.44a)
h r3 ⎣ ⎦
Here, vQ will be in Hz if Z is expressed in atomic units, r in cm, Q in m2. It can be assumed that D >> 1.
Further if we assume that the field gradient experienced by the bound chloride ion is solely to a positively
charged group on the protein, then the quadrupole coupling constant based on the simple electrostatic model is
given by
0.. (1 + γ ∞ ) ⎛ 2D + 3 ⎞ QZ
vQ ( )= ⋅⎜ ⎟ (8.44b)
I ( I ) ⎝ 5D ⎠ r 3
Here, Z is the point charge on the protein in atomic units, Q (in m2), r (in m) is the distance between the nucleus
and the point charge. For Cl− ion binding to an arginine group, the estimated value of vQ equals 1.4 MHz.
(b) Electronic Distortion Model This model is due to C Deverell. The model is based on the assumption
that the field gradients are produced at the halide ion nucleus by deformation in the electron cloud resulting from
collisions with other species in solution. Accordingly,
1 e 2Qσ p ΔE
E
vQ =
⋅ (8.45)
10 hα 2
⎛ 2π e ⎞
2
where, ΔE is the mean electron excitation energy and a is the fine structure constant ⎜⎝ α = hc ⎟⎠ .
and
2 2
3π 2 2I 3 ⎛ e 2Q ⎞ ⎛ σ p ΔEE⎞
R1 = ⋅ τc (8.46)
1000 I 2 ( 2I 1) ⎜⎝ h ⎟⎠ ⎜⎝ α 2 ⎟⎠
Here, tc for ion–water interactions is the reorientational correlation time of water molecules adjacent to the
−65.5 × 10 −4
relaxing ion. sp is the paramagnetic shielding constant sp(ClCl) = ; ΔE lies in the range of 10–12 eV.
ΔE
The modified version of electronic distortion model is due to Hertz. According to this version the distance
between the relaxing ion and the water molecules is assumed to vary with time.
2
3π 2 2I 3 ⎛ e Q ⎞
R2 = ⋅ 2 ⋅⎜ ⎟⎠ N H2 O P(a)τ a U a (8.47)
2
R1
10 I ( 2I 1) ⎝ h
τa
where, N H2 O is the total number of water molecules in the system. P (a) = is the probability of finding the
τ a + τb
water molecule in the position A. The contribution to relaxation rate from one water molecule is ∝ (Ua)2. Ua is the
field gradient in the position A. This model is obeyed by only weakly hydrated ion.
Problem 8.3: Calculate the correlation times of 35Cl-nuclei in (a) CH3Cl, and (b) C6H5Cl from the NQR data
(at 298 K) given below:
Molecule T2(m s) vQ(MHz)

CH3Cl 24.5 76.6
C6H5Cl 13 69.2

1
τc =
⎛ η2 ⎞
3.94384 ×T
T 2 v Q2 ⎜1 + ⎟
⎝ 3⎠
⎡ 3⎤
⎢ For I = 2 ⎥
⎣ ⎦
(a) For CH3Cl, h = 0
1
Therefore, τc = −6
3.94384 × 24.5 × 10 ( ) × (76
76.6 × 106 )2
= 1.76 × 10−12s = 1.76 ps.
(b) For C6H5Cl, h = 0.25
1
Therefore, τc =
⎡ (0.25) 2 ⎤
3.94384 × 13 × 10 −6 (s) × (69.2 × 106 ) 2 ⎢1 +
⎣ 3 ⎥⎦
10 −6
= s
3.94384 × 13 × (69.2) 2 × 1.0208
= 3.99 × 10−6 × 10−6s
= 3.99 × 10−12s = 3.99 ps
Problem 8.4: Calculate the quadrupole relaxation time of 35Cl− ion in an infinite dilute solution of water if Q
(cm2) = −0.079 × 10−24, 1 + g∞ = 57.6, r0(Å) = 3.21, H2O (s) = 2.5 × 10−12, P = 0.5.
Solution The negative sign of Q indicates that the charge distribution is flattened relative to the spin axis. We
know that the relaxation rate R1 is given by
2 ⎛ C H OτH O ⎞
2
24π 3 ( 2I 3) ⎛ eQ ⎞
R1 = . ⎜ ⎟ ⎡⎣ M H2O (1 γ ∞ )P ⎤⎦ ⎜ 2 5 2 ⎟ (Eq. 8.43)
5I ( 2I − 1) ⎝ h ⎠
2
(C ) ⎝ r0 ⎠
(A ) (B ) (D )
In order to determine R1 the values of four factors (A), (B), (C) and (D) need to be determined individually.
3
For Cl−, I=
2
⎛ 3 ⎞
24( .14)3 ⎜ 2 × + 3⎟
24 π (3
) ⎝ 2 ⎠
Therefore, ( )= =
5I 2 ( ) 9⎛ 3 ⎞
5 × 2 × − 1⎟
4⎝ 2 ⎠
24(3.14)3 6 × 4 32 × (3.14)3
= = = 198.1385
5×9×2 5
2 2
⎛ eQ ⎞ ⎛ 4.8 × 10 −10 × 0.079 × 10 −24 ⎞
( )=⎜ ⎟ =⎜ ⎟⎠
⎝ h ⎠ ⎝ 6.62 × 10 −27
[e = 4.8 × 10−10esu, h = 6.62 × 10−27 erg s]

= (0.0572)2 × 10−14 = 3.2811 × 10−17
M H2 O = 1.84 D = 1.84 × 10−18esu cm.
CH2 O = 55.5 moles/lit = 55.5 × 6.023 × 1023 molecule/103 cm3 =
334.2765 × 1020 molecule/cm3 = 3.342 × 1022 molecule/cm3
P =0.5
2 2
( ) ⎡M H O ( γ ))P ⎤⎦ = ⎡⎣1.84 × 10 −18 × 57.6 × 0.5⎤⎦ = 2.808 × 10−33
⎣ 2
C H2Oτ H2O 3.342 × 10 22 × 2.5 × 10 −12

( )= = =
r05 ( .21 × 10 −8 )5
8.355
× 1050 = 0.02451 × 1050 = 2.451 × 1048
340.820
Thus, R1 = (A) (B) (C) (D)
= (198.1385) × (3.2811 × 10−17) × (2.808 × 10−33) × (2.451 × 1048) s−1

= 4474.3 × 10−2 s−1 = 44.7 s−1
1
The quadrupole relaxation time of −Cl35 ion in an infinite dilute solution of water will be s = 0.0223 s.
44.7
Problem 8.5: Compute the 35Cl-quadrupole coupling constant for Cl− counter ion bound to N+ (CH3)3 group
in the amphiphilic [C12 N+ (CH3)3Cl−] − H2O system from the following data, g∞ (for chlorine) = 59, D >>1, r (the
distance between Cl− counter ion bound to amphiphilic surfaces and amphiphilic positive charge, i. e. N+ (CH3)3
end group) = 7.3 × 10−10 m. Q (for 35chlorine) = −0.0802 × 10−24 cm2. The Cl− counter ions are assumed to be
hydrated.
Solution We know that for the amphiphilic −H2O mesophases,
0.69 × 106 ( )(1 + γ ∞ )QZ ⎛ 2D + 3 ⎞

vQ = ⎜⎝ ⎟
r 3
5D ⎠
Substituting the values of respective parameters, we get
0.69 × 106 ×6 0 × (0.0802

60 080 ×10 0 −28
× 10 2
)⎛2 3 ⎞
vQ = ⎜⎝ + ⎟
(7.3 × 10 −10 m )3 5 5D ⎠
Further, ⎛ 2 3 ⎞ = 0.4 [since D >> 1] and Z = 1

⎜⎝ + ⎟
5 5D ⎠
0.69 × 106 ×6
60 × 10 −28
0 × (0.0802 ×10 2
)1 × 0.4
Therefore, vQ =
(7.3 × 10 −10 m )3
= 0.853 × 106 × 0.4 Hz
= 0.34 × 106 Hz = 0.34 MHz
8.9 PURE QUADRUPOLE RESONANCE

The field gradient responsible for large quadrupole interactions occur often in molecules with covalent bonding in
which valence electrons produce large asymmetric electric fields close to the nucleus. In such molecular systems,
the quadrupole interactions are very strong relative to Zeeman energy (i.e. e2qQ >> mH) and hence the Zeeman
hyperfine interaction is merely taken as perturbation. Therefore, in the limit of zero magnetic field, the quadrupole
interactions in such molecular systems are studied by pure nuclear quadrupole resonance (NQR) technique, which
deals with radio frequency transitions within the hyperfine multiplet of a nucleus in the ground state. Further it
is to be noted that in NQR, quadrupole moment is a probe to study the behaviour of electric field gradient in
molecular systems.
In order to deduce theoretically the pure NQR spectra of systems with quadrupole nuclei we are to proceed as
follows:
The classical expression for the quadrupole coupling Hamiltonian operator is given by
e 2 qQ ⎡ 1 U xx U yy ⎤
HQ = ⎢1 − ⎥ (8.48)
6 ⎣ 2 U zz ⎦
where terms independent of nuclear orientation have been ignored.

The quantum mechanical analogue of the classical quadrupole Hamiltonian is
eQU zz
HQ = ⎡⎣3I z2 − I 2 + η(I +2 I −2 ) / 2⎤⎦ (8.49)
4I ( 2I − 1)
where terms and symbols have their usual meanings. The asymmetry parameter h which is a measure of the
departure from the axil symmetry has already been defined earlier and is expressed as
( − )
η= xx yy
(8.50)
U zz
and is restricted to 0 ≤ h ≤ 1.
The eigen values of the Hamiltonian operator HQ i.e. the energy of the substates caused by pure nuclear qua-
drupole interaction is given by
eQU zz
EQ = ⎡3M 2 − I (I + 1) ⎤⎦ f ( ) (8.51)
4I ( 2I − 1) ⎣ I
where f(h) is a function of h.

Here, the substates are not discernible by the sign of MI, since Eq. (8.51) contains only the second power of MI.
Therefore, the nuclear substates arising from nuclear quadrupole splitting remain doubly degenerate. The two-
fold degeneracy of substates can only be lifted by magnetic perturbation.
For spherical symmetrical fields, (i.e. Uxx = Uyy = Uzz), all the quadrupole levels are degenerate, i.e. they have
the same energy. Consequently the degeneracy corresponding to MI cannot be lifted, and no resonance can take
place, e.g. systems with closed shell structure, i.e. Cl− ion, whereas for axil symmetrical fields, (i.e. Uxx = Uyy ≠ Uzz
e.g., the EFG at the chlorine nucleus along the F−Cl bond in FCl molecule is defined as Z-axis), I = 3/2, 5/2, 7/2 ...,
the number of substates = 2I + 1. The substates split into a series of Kramer’s doublet, i.e. the degenerate levels
lie very close to each other. The energy level splitting due to pure quadrupole interaction only for nuclei with
I = 1/2, 3/2 and 5/2 has been shown in Fig. 8.6.
(a)
I=1 ±1
2 Q.I. 2
(b)
±3
2
I=3
2
±1
2
Q.I.
(c)
±5
2
I=5 ±3
2 2
No field or only spherically

symmetric field ±1
2
Q.I.
Fig. 8.6 Energy-level splitting under pure quadrupole interactions (Q = I) only for nuclei
1 3 5
with (a) I = (b) I = ,(c) I = .
2 2 2
Further, since h = 0 for spherical symmetric and axil symmetrical fields, so for such systems, Eq. (8.51)
becomes
eQU zz
EQ = ⎡3M 2 − I ( I + 1) ⎤⎦ (8.52)
4I ( 2 I − 1) ⎣ I
3 5
With the aid of this equation, let us deduce the pure NQR spectra of nuclei with I = 1, , 2 and .
2 2
(a) For nuclei with I = 1, MI = −1, 0, +1. The two allowed transitions are: 0 → −1, +1 → 0. From Eq. (8.52), it
can be proved that the frequencies corresponding to these transitions have the same frequencies, i.e.
3 e 2 qQ
v1 (0 1) = v2 (1 → 0 ) =
4 h
The quadrupole energy levels and transitions for I = 1 are shown in Fig. 8.7 (a).
3 3 1 −1 −3 +1 −1 +3 +1
(b) For nucleus with I = , MI = ± , ± . The allowed transitions are → , → , →
2 2 2 2 2 2 2 2 2
+1 −1
The frequency corresponding to transition → is zero whereas the other two transitions have the same
frequency, i.e. 2 2
⎛ −1 −3 ⎞ ⎛ +3 +1⎞ e 2 qQ
v1 → v2 ⎜ → ⎟ =
⎝ 2 2⎠ ⎝ 2 2⎠ 2h
The quadrupole energy levels and transitions for I = 3/2 are shown in Fig. 8.7 (b).
(c) For nucleus with I = 2, MI = ±2, ±l, 0, The allowed transitions are: −1 → −2, 0 → −1, +1 → 0, +2 → +1.
The frequencies of transitions −1 → −2, +2 → +1 and 0 → −1, +1 → 0 are the same, i.e.
3 e 2 qQ
v1 ( 1 2) = v2 ( +2 → 1) =
8 h
1 e 2 qQ
and v3 (0 1) = v4 ( +1 → 0 ) =
8 h
The quadrupole levels and transitions for I = 2 are shown in Fig. 8.7 (c). Thus for I = 2, the NQR spectrum
consists of two lines and v1/v2 = 3 [v3/v4].
5 5 3 1
(d) For a nucleus with I = , MI = ± , ± , ± , the allowed transitions are
2 2 2 2
−3 −5 −1 −3 +1 −1 +3 +1 +5 +3 +1 −1
→ , → , → , → , → . The frequency corresponding to transition →
2 2 2 2 2 2 2 2 2 2 2 2
−1 −3 +3 +1 −3 −5 +5 +3
is zero whereas the transitions → , → and → , → have the same frequency,
2 2 2 2 2 2 2 2
i.e.
⎛ −1 −3 ⎞ ⎛ +3 +1⎞ 3 e 2 qQ
v1 → v2 ⎜ → ⎟ =
⎝ 2 2⎠ ⎝ 2 2 ⎠ 20 h
⎛ −3 −5 ⎞ ⎛ +5 +3 ⎞ 3 e 2 qQ
v3 → v4 ⎜ → ⎟=
⎝ 2 2⎠ ⎝ 2 2 ⎠ 10 h
±2
3 e2qQ
8 h
±1
1 e2qQ
8 h
0
±1 (c)
3 e2qQ
4 h ±5
2
0
(a) 2
3 e qQ
10 h
± 3
2
±3
2
1 e2qQ 2
2 3 e qQ
h 20 h
± 1 ±1
2 2
(b) (d)
3 5
Fig. 8.7 Quadrupole energy levels and transitions for (a) I = 1, (b) I = , (c) I = 2, and (d) I = .
2 2
The quadrupole energy levels and transitions for I = 5/2 are shown in Fig. 8.7 (d).
Thus, for I = 5/2, the NQR spectrum consists of two lines and the ratio between two frequencies is v3/v4 = 2v1/v2.
Consequently, one can write
3e 2qQ
v M( I )I M I −1 = [2 | M I | −1] (8.53a)
4I ( 2I − 1)h
3e 2qQ
and v M( I )I′ M I′ −1 = [2 | M I′ | −1] (8.53b)
4I ( 2I − 1)h
The ratio of these frequencies is
v M( I )→ M I −1 2 | M I′ | −1
= (8.54)
v (I )
M ′→ M I′ −1 2 | M I′ | −1
The above examples, i.e. (a)−(d), reveal that the pure NQR transition frequency depends on the magnitude of
the magnetic quantum number MI. The frequencies vary within very large limits that depend on the magnitude of
quadrupole moment and nature of the system under study.
Uptil now we have predicted the NQR spectra of integral and half-integral spin systems assuming h to be zero,
i.e. axially symmetrical fields. Let us now consider the case for nucleus with I = 3/2 when h ≠ 0, i.e.
Uxx ≠ Uyy ≠ Uzz (asymmetrical field)
This is the most complex case. The energy of pure NQR is given by
1
⎡ 1
⎤
e 2qQ ⎛ η2 ⎞ 2 ⎢ ⎛ η2 ⎞ 2 ⎥
EQ = [3M 1 − I (I + 1)] ⎜1 + ⎟ From Eq. (8.51), f (η) = 1 + ⎟
2
(8.55)
4I ( 2I − 1) ⎝ 3⎠ ⎢ ⎝ 3⎠ ⎥
⎢⎣ ⎥⎦
3 3 3 1
The energy difference ΔEQ between the two substates , ± > and , ± > can be worked out as follows.
From Eq. (8.55), we write 2 2 2 2
1
2
Q ⎛ η2 ⎞ 2
⎛ 3 ⎞ 3e qQ 3 3
EQ ⎜ ± ⎟ = 1 + ⎟ for I MI = ±
⎝ 2⎠ 12 ⎝ 3⎠ 2 2
1
2
Q ⎛ η2 ⎞ 2
⎛ 1 ⎞ −3e qQ 3 1
EQ ⎜ ± ⎟ = 1 + ⎟ for I MI = ±
⎝ 2⎠ 12 ⎜⎝ 3⎠ 2 2
Thus, the quadruple transition frequency is given by
⎛ 3⎞ ⎛ 1⎞
EQ EQ ± ⎟ 1
ΔEQ ⎝ 2⎠ ⎝ 2⎠ Q ⎛ η2 ⎞ 2
e 2 qQ
v= = = 1+ ⎟ (8.56)
h h 2h ⎝ 3⎠
Since there are two unknowns, i.e. h and Q, the values of both and e2qQ and h cannot be obtained from the
measured frequency. In such cases, the results from NQR experiments on a sample in a weak magnetic field are
necessary to obtain h.
5 7 9
For other half-integral spin systems with I = , , , the secular quadrupole energy equations are very
2 2 2
5
complex and cannot be solved exactly. However, for I = and h2 ≤ 0.25, the energy-level expressions are
2
⎛ 5⎞ ⎡ 5 85 4 715 62935 ⎤
EQ = A 10 + η 2 + η − η6 − η8 + ...⎥ (8.57)
⎝ 2⎠ ⎣ 9 2916 472392 306110016 ⎦
⎛ 3⎞ ⎡ 23 449 6 44675 8
EQ A ⎢ −2 + 3η 2 − η 4 + η − η + (8.58)
⎝ 2⎠ ⎣ 12 216 15552
⎛ 1⎞ ⎡ 32 1376 4 122656 6 13740640 8 ⎤

EQ A ⎢ −8 − η 2 + η − η + η + ...⎥ (8.59)
⎝ 2⎠ ⎣ 9 729 59049 4782969
9 ⎦
e 2 qQ
where, A=
4I ( 2 I − 1)
For h ≤ 0.1, the second term in the brackets is significant. Therefore, for I = 5/2, it can easily be shown that
⎛ 3 1 ⎞ 3e 2 qQ ⎡ 59 2 ⎤
v⎜+ → + ⎟ = 1+ η (8.60)
⎝ 2 2⎠ 20 h ⎣ 54 ⎥⎦
⎛ 5 3 ⎞ 3e 2 qQ ⎡ 11 2 ⎤
v⎜+ → + ⎟ = 1− η (8.61)
⎝ 2 2⎠ 10 h ⎢⎣ 54 ⎥⎦
e 2 qQ
Thus, for a small value of h, we can compute both and h with the aid of Eqs. (8.60) and (8.61) provided
both the frequencies are measured experimentally. h
The ratio of frequencies for these two transitions is;
⎛ 5 3⎞
v⎜+ → + ⎟
⎝ 2 2⎠ ⎡ 35 ⎤
= 2 ⎢1 − η 2 ⎥ (8.62)
⎛ 3 1⎞ ⎣ 27 ⎦
v⎜+ → + ⎟
⎝ 2 2⎠
For a small value of h, Eq. (8.62) reduces to:
⎛ 5 3⎞
v⎜+ → + ⎟
⎝ 2 2⎠
=2 (8.63)
⎛ 5 1⎞
v⎜+ → + ⎟
⎝ 2 2⎠
Therefore, any deviation from the observed ratio of these two frequencies from the value of 2 suggests that h
is finite.
5
For low h (say < 0.25) and I = , we also have
2
3 ⎛ e 2 qQ ⎞
v1 = (1 0.09259η 2 − 0.63403
. 4
) (8.60a)
20 ⎜⎝ h ⎟⎠
3 ⎛ e2 qQ ⎞
v2 = (1 0.20370η2 + 0.16215
. 4
) (8.61a)
10 ⎜⎝ h ⎟⎠
7
and, for I = , η < 0.25
2
1 ⎛ e 2 qQ ⎞
v1 = ⎜ ⎟ [1 3.63333η 2 − 7.26070η 4 ]
14 ⎝ h ⎠
2 ⎛ e 2 qQ ⎞
v2 = ⎜ ⎟ [1 0.56667η 2 + 1.85952η 4 ] (8.64)
14 ⎝ h ⎠
3 ⎛ e 2 qQ ⎞
v3 = [1 0.1001η 2 − 0.01804η 4 ]
14 ⎜⎝ h ⎟⎠
Here, also the harmonic ratio of the frequencies v1, v2 and v3 is no longer observed due to the asymmetry param-
e 2 qQ
eter h. However, both h and can be separately derived provided the theoretical equations. (8.64) can be
h
fitted to the observed frequencies.
For integral spins I = 1, 2, 3 ..., etc., with nonzero h, the secular quadrupole energy equations except I = 1
are also complex and cannot be solved exactly. The behaviour of the energy levels and transition frequencies for
h = 0 and finite h is shown in Fig. 8.8.
When, h ≠ 0, we observe three transition frequencies of
1 ⎛ e 2 qQ ⎞ ν3 = (3/4)[e 2qQ/h][1 + (η/3)]
v1 = η
2 ⎜⎝ h ⎟⎠ M I = ±1
M I = +1
3 ⎛ e qQ ⎞ ⎛ η ⎞
2
ν1= (1/2)[e 2qQ/h]η
v2 = ⎜1 − ⎟⎠ (8.65)
4 ⎜⎝ h ⎟⎠ ⎝ ν= (3/4)[e 2qQ/h] M I = −1
3
No field or
only spherically ν2= (3/4)[e 2qQ/h][1 − (η/3)]
3 ⎛ e 2 qQ ⎞ ⎛ η ⎞ symmetric field
MI = 0
MI = 0
v3 = ⎜ ⎜1 + ⎟⎠
4 ⎝ h ⎟⎠ ⎝ 3 (a) (b)
Fig. 8.8 Quadrupole energy levels and transitions for nuclei
When, h → 0, a single transition is observed at with I = 1 (a) h = 0. (axially symmetric fleld) (b) h ≠ 0 (axially
3 ⎛ e 2 qQ ⎞ nonsymmetric field).
. Thus h and vQ can once again be deter-
4 ⎜⎝ 4 ⎟⎠
mined from the NQR spectrum of the powder sample. 14N-quadrupole spectra mostly consist of two lines corre-
sponding to frequencies v2 and v3.
Note that the MI degeneracy of the energy levels is removed for nuclei with integral spins if h ≠ 0. For half-
integral spins, the MI degeneracy is unaffected whether h is zero or not.
The sign of the coupling constant in a pure quadrupole resonance experiment (a) have been compared with the
sign of the coupling constant in a nuclear magnetic resonance experiment in which the magnetic field lies along
the field gradient principal axis is depicted in the schematic diagrams in Fig. 8.9.
(a)
E MI MI
±5/2 ±1/2
±3/2
±3/2
±1/2 ±5/2
e2qQ > 0 e2qQ < 0
(b)
MI
E
–3/2
–1/2
−1/2
−3/2
mN > 0 e2qQ > 0 mN > 0
e2qQ > 0
Fig. 8.9 Schematic diagram of the coupling constants (a) in pure quad-
rupole resonance experiment and (b) in a nuclear magnetic resonance
experiment in the presence of magnetic field along the field gradient axis.
Problem 8.6: Two groups of transitions at 198.736, 199.999 and 397.430, 399.993 MHz have been observed
in the 127I-quadrupoie spectrum of SiI4 at 77 K. Determine the values of asymmetry parameter and coupling
constants.
Solution The two groups of frequencies are in harmonic ratio. Therefore, h = 0.

For 127
I (I = 5/2), we observe two transitions of
3 ⎛ e 2 qQ ⎞
v1 =
20 ⎜⎝ h ⎟⎠
3 ⎛ e 2 qQ ⎞
v2 =
10 ⎜⎝ h ⎟⎠
Thus, the values of coupling constants corresponding to each of the frequencies in the group will be
⎛ e 2 qQ ⎞ 20
⎜⎝ h ⎟⎠ = 198.736 × = 1324.90 MHz
198.736
3
⎛ e 2 qQ ⎞ 20
⎜⎝ h ⎟⎠ = 199.999 × = 1333.32 MHz
199.999
3
The two coupling constants may be attributed to the lower point symmetry of the molecule in the crystal com-
pared to the vapour.
Co (I = 7/2) quadrupole spectrum of cobalticinium perchlorate ⎡Co (C6 H 5 )2 ClO 4− ⎤ exhibits

+
Problem 8.7: 59
⎣ ⎦
three groups of frequencies near 12, 24, and 36 MHz. Calculate the values of h and nuclear quadrupole coupling
constants.
7
Solution The three groups of frequencies are in harmonic ratio. Therefore, h = 0. For I = , the three transi-
tions as worked out from Eq. (8.64) are 2
1 ⎛ e 2 qQ ⎞
v1 = ⎜
14 ⎝ h ⎟⎠
2 ⎛ e 2 qQ ⎞
v2 =
14 ⎜⎝ h ⎟⎠
3 ⎛ e 2 qQ ⎞
v3 =
10 ⎜⎝ h ⎟⎠
⎛ e 2 qQ ⎞
Thus, ⎜⎝ h ⎟⎠ = 14 × v1 = 14 × 12 = 168 MHz.
Problem 8.8: I-quadrupole spectra of BI3 show two frequencies at 212.6 and 340.1 MHz. Find the values of
127
quadrupole coupling constant and asymmetry parameter.
5
Solution For 127I, I = . The frequencies are not in harmonic ratio, so h ≠ 0
2
340.1 ⎡ 35 ⎤
= 2 1 − η2 ⎥ .
212.6 ⎣ 27 ⎦
35 340.1
1 − η2 = = 0.7998
27 212.6 2
35 2
η = 0 20
27
0 20 27
η2 = = 0.1543
35
or h = 0.392
Substituting h = 0.392 into Eq. (8.61), we get
3e 2 qQ ⎡ 11 ⎤
340.1 = 1− × (0.392)2 ⎥
10 h ⎢⎣ 54 ⎦
e 2 qQ 10 × 340.1 3401
or vQ = = = = 1169.93
3 MHz
h ⎡ 11 2⎤ 3 × 0.969
3 ⎢1 − (0.392) ⎥
⎣ 54 ⎦
8.10 EQUIVALENCE OF LARMOR’S THEOREM OF NMR

IN NQR SPECTROSCOPY
In the quadrupole case, the precessional frequencies are different in different orientations whereas in magnetic case
the frequency is independent of orientations. If nuclear spin axis is ‘set’ at one or more fixed orientations defined
by Iz, the component of nuclear angular momentum along Z-axis, then at one of these positions, there will be one
precessional frequency. This precession of the nucleus carries with it nuclear magnetic moment, thereby producing
a small magnetic field at right angles to Z-axis rotating in phase with nuclear quadrupole at the same frequency.
Now a linearly oscillating field
H = 2 H1 cos ωt (8.66)
can be resolved into two rotating, i.e. clockwise (cW) and counter clockwise (ccW) fields.
H( W ) H1 cos ω
iH i ωt ⎫
i 1 sin
⎬ (8.67)
H ( W ) = H1 cos ω iH i ωt⎭
i 1 sin
where, H = H (cW) + H (ccW).

Such a field is obtained by enclosing the sample in a small induction coil through which an alternating current
is passed. The component of Eq. (8.67), which rotates in phase with the precessing nucleus exchange energy with
it, resulting into resonance process occurring at a specific quadrupole frequency v.
8.11 NQR SPECTROMETER

MI degeneracy in NMR is lifted by external magnetic field whereas in pure NQR, it is removed by EFGs which
are due to the electrons in the molecules. Since the transitions between the energy states in NQR are fixed by the
electronic structure of the molecules, they are studied by the radio frequency (Rf) sweep method rather than the
field sweep method. The frequency range that can be studied by NQR is a few MHz to 1000 MHz.
The NQR spectrum which is a plot between intensity (on the vertical axis) and frequency (on the horizontal
axis) is obtained by a device called NQR spectrometer. A block diagram of an NQR spectrometer is shown in
Fig. 8.10. The five basic elements of the spectrometer are (i) radio frequency (Rf) oscillator, (ii) bridge circuit,
(iii) amplifier, and (iv) oscillator detector which is usually a single-tube arrangement employing frequency
modulation. Conventional phase-sensitive detector arrangement is used for weak outputs and (v) recorder.
The sample is placed in the induction coil of the bridge circuit. For better results, the coil is surrounded by
the sample.
Power from the Rf oscillator is supplied to a pre-balanced bridge circuit where it is resonantly absorbed by the
sample causing an unbalance in the circuit also. The resulting output is amplified, fed to the detector and then
recorded.
Oscilloscope
Rf Chart
Amplifier Detector
Oscillator recorder
Computer
Fig. 8.10 Schematic diagram of NQR spectrometer.
Radio-frequency pulse techniques (spin-echo) are used in limited cases. The frequency shifts under the
influence of external applied electric field (10 kV/cm) are also studied. This type of electric field effect may
be of use in the study of noncentrosymmetric crystals by NQR. Measurements have been made on single
crystals of sodium chlorate and on p-dichlorobenzene to verify the Stark effect in pure nuclear quadrupole
resonance.
Only solids can be studied and samples may be in the form of crystals or crystalline powder but not ampor-
phous powder. Gases and liquids cannot be studied by NQR method since at normal pressure, due to molecular
collisions the axis of rotation changes continuously because of which the net EFGs at the nucleus averages to zero.
This factor limits the applications of NQR compared to NMR. e2qQ of gases is obtained from the fine structure of
rotational spectrum arising from the interaction of EFGs with rotational motion of the molecules and has already
been reported in Chapter 2.
Moderately large samples are required in NQR studies, e.g. a 50 mg sample of p-dichlorobenzene produces
observable 35Cl resonances and several grams of haxmethylene tetra-amine are needed to produce observable
14
N-resonances. Moreover, the liquid samples need to be cooled to 77 K in order to obtain a crystalline solid.
The sample should be pure otherwise the impurities present in it will broaden the spectrum. The broadening
may be due to charge as well as size of the impurities. The extra charge due to impurities generates additional
EFGs at the nucleus. Since the foreign nuclei are asymmetrically distributed around the sample nuclei, the
EFGs contribution due to foreign nuclei to different sample nuclei will vary, and consequently the asymmetric
distribution of EFGs due to foreign nuclei over the sample nuclei will cause spectrum broadening. The size
effect due to foreign nuclei will become appreciable provided the size of the foreign nuclei is comparable to
that of the parent nuclei. The foreign nuclei will distort the electron cloud of the parent nuclei and consequently,
the EFGs at the parent nuclei. It is to be noted that broadening not only depends on the charge and size of the
impurity nuclei but also on the composition of the impurity since intermolecular interactions between foreign
nuclei and neighbouring molecules depend on the quadrupole moments of nuclei in both the molecules, i.e.
sample and impurity.
8.12 EFFECTS OF VARIOUS FACTORS ON NUCLEAR

QUADRUPOLE RESONANCE FREQUENCY
The main parameters which affect the NQR resonance frequencies are temperature, pressure, internal rotation,
substituents in the molecule, etc.
8.12.1 Effect of Temperature and Pressure

Frequency variation with temperature and pressure is interpreted on the basis of lattice and torsional oscilla-
tions and due to restricted rotation around C−C bond in some chloroethanes is interpreted in terms of torsional
oscillations.
8.12.2 Conjugative Effects

The electron releasing and attracting groups in substituted halobenzenes affect the nuclear quadrupole coupling
constants. The electron-releasing groups have negative Hammett-s values and give rise to smaller vQ values. This
may be attributed to increase of ionic character in the C−X (X–Cl, Br, I) bond which makes the X-atom more
negatively charged.
In case of substituted halobenzenes, linear relationships exist between s-constant and nuclear quadrupole
resonance frequencies, i.e.
v (35Cl) = 34.826 (1 + 0.0287 s) MHz (8.68)
v (81Br ) = 226.932 (1 + 0.0336 s) ± 2.98 MHz (8.69)
v ( I) = (267.0 + 1.424 s) MHz
127
(8.70)
1 3
Here, v(127I) corresponds to the MI = ± ↔ ± transitions only.
2 2
For multichlorobenzenes, the resonance frequency of a particular chlorine atom is well represented by an equa-
tion of the form
v (35Cl) = v (= 34.622 MHz; chlorobenzene) + a Δvortho + b Δvmeta + g Δvpara ( 8.71)
where, Δvortho = 1.30 MHz, Δvmeta = 0.45 MHz, Δvpara = 0.25 MHz and a, b and g are the numbers of chloro atoms
ortho, meta and para respectively to the chlorine atom whose frequency is being calculated.
The theoretical ratios of Δvs obtained from simple electrostatics combined with the known geometries of the
chlorobenzenes are
Δvortho : Δvmeta : Δvpara : : 1 : 0.27 : 0.20
The experimental ratios are 1 : 0.35 : 0.19.

It is to be noted that for chloro and bromo benzenes, h is not much different from 0.1, so vQ = 2v.
Problem 8.9: Calculate 35Cl-pure quadrupole resonance frequencies of (a) m-dichlorobenzene, (b) p-dichlo-
robenzene, (c) o-dichlorobenzene, (d) 1, 2, 3-trichlorobenzene, and (e) 1, 2, 3, 4-tetracloro benzene.
Solution From Eq. (8.71), we obtain

(a) v (35Cl) = 34.622 + 0.45
= 35.072 MHz (observed (average) = 34.904 MHz; 34.809, 34.875, 35.030 MHz)
(b) v ( Cl) = 34.622 + 0.25 = 34.875 (observed (average) = 34.799 MHz)
35
(c) v (35Cl) = 34.622 + 1.30 = 35.922 (observed (average) = 35.719 MHz; 35.580, 35.755, 35.824 MHz)
(d) v (35Cl) = 34.622 + 1.30 + 0.45 = 36.372 MHz (observed (average) = 36.547 MHz; 36.214, 36.238, 36.268,
36.563, 36.973, 37.031 MHz).
(e) v (35Cl) = 34.622 + 1.30 + 0.45 + 0.25
= 36.622 MHz (observed (average) = 37.34 MHz; 37.013, 37.455, 37.557 MHz).
Problem 8.10: (i) Predict the numbers of chloro atoms ortho, meta and para respectively to the chlorine atom
whose frequency is being measured experimentally for the following chlorobenzenes (provided v1, 2, 3, 4 > v1, 2, 3, 5 > v1, 2, 4, 5 )
(ii) Assign also the names to the tetrachlorobenzenes.
Chlorobenzene 35
Cl-NQR frequencies (MHz)
Tetrachlorobenzene (a) 37.013, 37.455, 37.557

Tetrachlorobenzene (b) 36.185, 36.386, 36.816, 36.872, 37.020, 37.289, 37.483, 37.657
Tetrachlorobenzene (c) 36.702, 36.738, 36.843, 36.898
Solution The average values of frequencies for (a), (b) and (c) are 37.3416, 36.9635 and 36.7952 respectively. Thus
(i) For tetrachlorobenzene (a), the average value of v (35Cl) is 37.3416 MHz.
From Eq. (8.71), 37.3416 = 34.622 + 1.30 a + 0.45 b + 0.25 g
Choose different numbers for a, b and g so that the above expression is mathematically balanced.
The best fit is obtained for a = 1, b = 1 and g = 1, i.e. the right-hand side becomes equal to 36.622 MHz.
(ii) The name of the tetrachlorobenzene (a) will be 1, 2, 3, 4-tetrachlorobenzene.
(i) For tetrachlorobenzene (b), the average value of observed v (35Cl) is 36.9635 MHz.
From Eq. (8.71), we write 36.9635 = 34.622 + 1.30 a + 0.45 b + 0.25 g
The best fit is obtained for a = 1, b = 2 and g = 0, i.e. the right-hand side of the above expression becomes
equal to 36.822 MHz.
(ii) The tetrachlorobenzene (b) is 1, 2, 3, 5-tetrachlorobenzene.
(i) For tetrachlorobenzene (c), the average value of observed v (35Cl) equals 36.7952 MHz. From Eq. (8.71),
we write
36.7952 = 34.622 + 1.30 a + 0.45 b + 0.25 g
The best fit is obtained for a = 1, b = 1 and g = 1, i.e. the right-hand side of the above expression becomes
equal to 36.62 MHz.
(ii) The tetrachlorobenzene is 1, 2, 4, 5-tetrachlorobenzene.
8.12.3 Inductive Effects

The coupling constants of the substituted chloroalkanes, i.e. series RCH2Cl is a measure of the inductive effect of
the substituent R, since the presence of the methylene (−CH2) group prevents conjugation with the chlorine atom.
The linear relationship between the Taft’s s * parameter and the NQR frequency of the 35Cl atom under measure-
ment is given by
v (35Cl) = 32.70 + 1.715 | s * | ( MHz) (8.72)
where 32.70 MHz is the 35Cl − NQR frequency of CH3CH2Cl. s * values of some typical groups are given in Table 8.2.
Table 8.2 s* values of some typical groups.
R s*
CH3CH2CH2CH2— −0.130
CH3CH2CH2— −0.115
CH3CH2— −0.110
CH3— 0.0
CH3CH==CH— 0.36
ClCH2CH2— 0.385
CH2==CH— 0.41
H— 0.49
CH3OCH2— 0.52
C6H5— 0.60
C6H5OCH2— 0.85
HO CO CH2— 1.05
ClCH2— 1.05
CH3CO— 1.65
+
HN (CH3)2CH2— 1.90
CH3OCO— 2.00
CF3— 3.1
35
Cl-NQR frequency of a substituted methylchloride R1 R2 R3CCl is related to Taft s* parameter of the substituents
by the expression.
v( 35 ( i
)
.0317 ∑ i* ± 0.35 MHz (8.73)
Problem 8.11: Calculate the 35Cl-NQR frequency of (i) CF3CH2Cl (ii) CH3COCH2Cl (iii) CH3CH2CH2Cl

v (35Cl) = 32.70 + 1.715 | s* | MHz (Eq. 8.72)
Thus, for
(i) CF3CH2Cl
v (35Cl) = 32.70 + 1.715 × 3.1 = 32.70 + 5.316
= 38.016 MHz (obs: 37.98 MHz, the average of two transitions)
(ii) CH3COCH2Cl
v (35Cl) = 32.70 + 1.715 × 1.65 = 32.70 + 2.829
= 35.529 MHz (obs: 35.28 MHz, the average of two transitions)
(iii) CH3CH2CH2Cl
v (35Cl) = 32.70 + 1.715 × 0.110 = 32.70 + 0.188
= 32.888 MHz (Obs: 32.968 MHz).
8.13 APPLICATIONS OF NQR SPECTROSCOPY

As already pointed out that when quadrupole interaction is small relative to Zeeman energy, the NQR is studied
by NMR spectroscopy whereas in the reverse case, the splitting of energy levels is due to EFGs only. In such a
case, the NQR is studied by pure NQR techniques, The applications of NQR in the former case are based mainly
on the intensity and width of NMR lines whereas in the latter case are based mainly on three NQR parameters,
viz., quadrupole splitting, EFG’s and quadrupole coupling constant. Applications are being described here under
two subtitles, namely NQR through NMR and pure NQR.
8.13.1 Applications of NQR through NMR

(a) Study of Chemical Equilibria The intensity of 127I NMR signal can be used to study to equilibrium
K
H 4 IO6 (aq ) IO 4 (aq ) 2H 2 O
⎡⎣IO 4− (aq ) ⎤⎦
K= (8.74)
[H 4 IO6− (aq )]
Thus, K can be obtained from Eq. (8.74) provided concentrations of the products and reactants are known.
The exchange between two iodine sites in the system is assumed to be slow. The signal of 127I corresponding
to H 4 IO6− (aq) in the NMR spectrum is too broad to be detected. The signal of 127I of IO 4− (aq) gives directly the
concentration of IO 4− . Thus, the values of K can be obtained from Eq. (8.74).
(b) Study of Contact Ion-Pairs Formation in Non-aqueous Media The rapid increase in the 35Cl relax-
ation rate (i.e. line width) with the increase in the concentration of electrolyte in an organic solvent is attributed
to the formation of contact ion pairs, e.g. LiClO4 in methanol, propanol, butanol, acetonitrile, nitro methane, ethyl
acetate, etc.; Mg (ClO4)2 and NaClO4 in propanol, butanol and ethyl acetate.
(c) Study of Interaction of 35Cl Nuclei with the Electron Spins The addi-
( CH2 CH2 )n
tion of Mn2+ to an aqueous solution of ClO 4− causes an increase of 35Cl relaxation
rate (i.e. line broadening). The decrease of 35Cl line broadening in the presence Cl
of Mn2+ with the rise in temperature suggests that the relaxation results from the H2C C O D C Cl
interaction of 35Cl nuclei with the electron spins. This interpretation finds support Cl
from the ratio of the 35Cl and 37Cl relaxation rates which is found to be 1.35. H2C CH2
(d) Study of Molecular Interactions The NQR relaxation tool is most suited to study the molecular
interaction of a small molecule with a macromolecule provided there is a marked difference between correla-
tion times of the free and bound states of the small molecule, e.g. binding of CDCl3 to polyvinylpyrrolidone.
35
Cl relaxation rate is less affected relative to 2H relaxation rate on binding of CDCl3 to the polymer. This suggests
that chloroform reorientation is slowed down around the symmetry axis than around the perpendicular axis due to
hydrogen bonding as shown below:
(e) Kinetics of Quadrupole Nuclei Chemical Exchange Process Recall from Chapter 7 on NMR that if
magnetic nuclei reside in two chemically different environments say A and B, they will show two distinct NMR
signals provided no exchange of nuclei occurs between the two sites. However, after the onset of exchange pro-
cess, the signals will begin to broaden and at high exchange rates will merge together into a one narrow signal. If
PA and PB are the mole fractions of nuclei at sites A and B respectively then PA + PB = 1. If PA >>PB, then we can
write PA ∼
− 1. In the halide ion-exchange process, environment A corresponds to ‘free’ halide ions in solution and
environment B corresponds to a halide ion coordinated to metal ion. Let us now consider the following exchange
system.
X−(A) + Mn+ = MX(n – 1)+ (B) (8.75)

where, M stands for metal.
1 1
In the absence of chemical exchange, the transverse relaxation rates and in the two environments will
1 1 T2 A T2B
be different for quadrupole nuclei and >> .
T2B
2 T2 A
Under these conditions, the signal near or at A with frequency vA will be observable whereas the signal at B
with frequency vB will be too broad to be easily observed. The NMR spectrum in the absence and after onset of
exchange process, have been illustrated in Fig. 8.11. The line width of the major signal may be expressed as
Δn
(a) (b)
2
T2A
2
T2A
2
T2B
nB nA nB nA
Fig. 8.11 Schematic NMR spectrum of nucleus which resides at two chemically different environments A and B with
frequencies vA, and vB respectively (a) in the absence of exchange, and (b) after onset of exchange process.
PB ⎡ 1 ⎛ 1 1⎞ 2⎤
⎢ ⎜ + ⎟ + 4π 2 ( Δv AB ) ⎥
1 1 τB ⎣ T2 B ⎝ T2 B τ B ⎠ ⎦
= + 2
(8.76)
T2 T2 A ⎛ 1 1⎞
⎜⎝ T + τ ⎟⎠ + π ( Δ )
2 2
2B B
1
where, is the rate of chemical exchange of nuclei from environment B, and ΔvAB = vA − vB. The position of the
τB
observed signal near vA will be shifted from its original position by an amount Δv and the following expression
holds good:
1 1
2πΔv AB × ×
τA τB
Δv = 2
(8.77)
⎡ 1 1⎤
⎢T + τ ⎥ + π (Δ )
2 2
⎣ 2B B ⎦
1
Here, is the chemical exchange rate of nuclei from environment A.
τA
In the steady state,
τA P 1
= A = , so Eq. (8.77) may be written as
τB PB PB
2
⎛ 1⎞
2π Δv AB ⎜ ⎟ PB
⎝ τB ⎠
Δv = 2 (8.78)
⎡ 1 1⎤
⎢T + τ ⎥ + π (Δ )
2 2
⎣ 2B B⎦
1 1
According to Eqs (8.77) and (8.78), the appearance of a spectrum depends on the values of , , P and
τ B T2 B B
ΔvAB. Parameter PB is considered to be only a multiplicative constant and can be altered by changing the compo-
sition of the solution. All the four parameters depend on temperature but the most temperature-dependent param-
1
eter is the rate of chemical exchange, . Thus, if the temperature dependence of the other three parameters is
τB
ignored then the influence of the exchange rate on the observed NMR spectrum can be conveniently discussed in
terms of two dimensionless parameters defined as
⎛ 1⎞
⎜⎝ τ ⎟⎠ T
ξ= B
= 2B (8.79)
⎛ 1 ⎞ τB
⎜⎝ T ⎟⎠
2B
and
2πΔv AB
η= (8.80)
⎛ 1 ⎞
⎜⎝ T ⎟⎠
2B
Thus, Eq. (8.77) may be rewritten as
⎛ 1⎞ 1 1 P ⎡ ξ + ξ + η2 ) ⎤
Δ⎜ ⎟ = − = B ⎢ 2 ⎥
(8.81)
⎝ T2 ⎠ T2 T2 A T2 B ⎣ (1 + ξ + η ⎦
2
⎛ 1⎞
where, Δ ⎜ ⎟ is the excess relaxation rate of the observed signal. The excess line broadening, Δve, of a signal is
⎝ T2 ⎠
then defined as
1 ⎡ ⎛ 1 ⎞⎤
Δve = ⎢ Δ ⎜ ⎟ ⎥ (8.82)
π ⎣ ⎝ T2 ⎠ ⎦
Similarly, Eq. (8.77) in terms of dimensionless parameters x and h will appear as
Δv ⎡ ξ2 ⎤
= PB ⎢ 2⎥
(8.83)
Δv AB ⎣ (1 + ξ + η ⎦
2
⎛ 1⎞
Let us now discuss the effect of temperature of the system on Δ ⎜ ⎟ and Δv in terms of x which is a measure
of exchange rate relative to the relaxation rate. ⎝ T2 ⎠
(i) When x << 1, the exchange rate will be very slow relative to the relaxation rate in site B. Therefore, in this
region, Eq. (8.81) may be written as
⎛ 1⎞ ξ PB P
Δ⎜ ⎟ = = B (8.84)
⎝ T2 ⎠ T2 B τB
Accordingly, the excess line broadening is very small and is independent of h. It will increase linearly with the
Δv
chemical exchange rate. The relative chemical shift displacement, in this region will be small.
Δv AB
For h2 << 1, Eq. (8.83) reduces to

Δv P ξ2 (8.85)
= B 2
Δv AB (1 + ξ )
Δv
i.e. ∝ PBξ 2
Δv AB
For larger values of h,
Δv P ξ2
< B 2 (8.86)
Δv AB (1 + ξ )
(ii) When x = 1, the exchange rate will be comparable in magnitude to the relaxation rate at the site B. There-
fore, in this region, Eq. (8.81) may be written as
⎛ 1⎞ P ⎡ η2 + 2 ⎤
Δ⎜ ⎟ = B ⎢ 2 ⎥ (8.87)
⎝ T2 ⎠ T2 B ⎣ η + 4 ⎦
i.e. the observed excess line broadening in this region depends on h. Further, if (a) h << 1, we obtain
⎛ 1⎞ P ⎡ ξ ⎤
Δ⎜ ⎟ = B ⎢ (8.88)
⎝ T2 ⎠ T2 B ⎣1+ξ ⎥⎦
⎛ 1⎞ PB
or Δ⎜ ⎟ = (8.89)
⎝ T2 ⎠ τ B + T2 B
i.e. the excess relaxation rate becomes appreciable and its value is about half its limiting value (PB /T2B) at a very
rapid chemical exchange (x >>1).
Δv
The corresponding relative chemical shift change in this case is expressed as
Δv AB
Δv ⎡ ξ2 ⎤
= PB ⎢ ⎥ (8.90)
Δv AB ⎢⎣ (1 + ξ ) ⎥⎦
2
From this it can easily be shown that

Δv
= 0.25 PB, i.e. it has reached about 25 per cent of its maximum value (= PB).
Δv AB
(b) For h >>1, we obtain
⎛ 1⎞ Pξ
Δ⎜ ⎟ = B (8.91)
⎝ T2 ⎠ T2 B
i.e. in this subcase, the excess relaxation rate or width is independent of h. The broadening may exceed the limit-
ing excess relaxation rate at very rapid chemical exchange.
Δv
The relative chemical shift change , in this subcase will be given by
Δv AB
Δv ⎡ ξ2 ⎤
= PB ⎢ 2 ⎥ (8.92)
Δv AB ⎣η ⎦
i.e. the relative chemical shift change in this subcase will be very small.
(c) For x � 1 and h � 1, the expressions (8.81) and (8.83) reduce to
⎛ 1 ⎞ 3 PB
Δ⎜ ⎟ = (8.93)
⎝ T2 ⎠ 5 T2 B
Δv 1
and = PB (8.94)
Δv AB 5
In this subcase, both the excess relaxation rate and the relative chemical shift change will depend on h.
PB
The excess relaxation rate is about 60 per cent of its maximum value = whereas, the relative chemical shift
T2 B
change has reached 20 per cent of its maximum value (= PB). Studies of NMR spectra at different frequencies may
show frequency-dependent line broadenings and shift changes.
(iii) When x >>1, the chemical exchange rate at the site B will be very fast relative to the relaxation rate in this
site. Assuming x >>h2, Eqs (8.81) and (8.83) reduce to
⎛ 1⎞ P
Δ⎜ ⎟ = B (8.95)
⎝ T2 ⎠ T2 B
Δv
and = PB (8.96)
Δv AB
PB
Thus, the maximum observable values of excess relaxation rate and relative shift change are and PB
T2 B
respectively. The signal now simply occurs at a weighted mean of the frequencies in environments A and B.
⎛ 1 ⎞
Further, when h x ⎜ i.e. Δv AB = , we obtain
⎝ 2πτ B ⎟⎠
⎛ 1 ⎞ P ξ ⎡ 1 + ξ ξ + 1) ⎤
Δ⎜ ⎟ = B ⎢ ⎥ (8.97)
⎝ T2 ⎠ T2 B ⎣1 + 2ξ (ξ + 1) ⎦
⎛ Δv ⎞ ⎡ ξ2 ⎤
and ⎜⎝ Δv ⎟⎠ = P ⎢ ⎥ (8.98)
⎢⎣ (1 + ξ ) + ξ 2 ⎥⎦
B 2
AB
The excess relaxation rate as well as the relative chemical shift change will now depend on h but will never
P
exceed their respective limiting values, i.e. B and PB. The above results are summarised in the form of plots of
T2 B
⎛ 1 ⎞ T2 B ⎡ Δv ⎤
Δ⎜ ⎟ and ⎢ ⎥ as a function of log ξ for different values of h (Figs 8.11 and 8.12). It is to be noted that
⎝ T2 ⎠ PB ⎣ PB Δv AB ⎦
the effect of chemical exchange rate on excess line width is small when h lies between 0 and 1 (Fig. 8.11), whereas
the effect on relative chemical shift displacement is independent of h (Fig. 8.12).
(i) (ii) (iii)

T2B
PB
T2
1
Δ
PB ΔnAB
Δn
1
h = 10
0 h = 0 3.3 10 100
− 0 +
− 0 +
log ξ
log ξ
Fig 8.12 The behaviour of excess relaxation rate
Fig 8.13 The behaviour of relative chemical
⎛ 1⎞ Δν
Δ ⎜ ⎟ of the dominant NMR signal with the dimen- shift displacement of the dominant NMR
⎝T ⎠
2 Δν AB
sionless exchange parameter x in a chemical exchange signal with x in chemical exchange process for
process for different values of h. different values of h.
(f) Effects of Different Isotopes on Relaxation Rate In the limit of fast exchange, the relaxation rates of
different isotopes in diamagnetic halogen compounds are proportional to the square of the respective nuclear qua-
drupole moments provided the effects of isotopes on EFGs’ are ignored.
Thus, for chlorine isotopes
35 ⎛ 1⎞ ⎛ 1⎞
Cl ⎜ ⎟ 37 Cl ⎜ ⎟
⎝ Ti ⎠ ⎝ Ti ⎠ (8.99)
= [Q(35Cl) Q(37Cl)]2 = 1.610

and for bromine isotopes
⎛ 1⎞ ⎛ 1⎞
( ) ( )
2
79
Br ⎜ ⎟ 81
Br ⎜ ⎟ = ⎡⎣Q
B 79
Br Q 81
Br ⎤⎦ (8.100)
⎝ Ti ⎠ ⎝ Ti ⎠
= 1.433
where, i = 1 or 2.
The experimentally observable line widths for the two isotopes may be calculated from Eq. (8.101) as
1
T2 B ( I )
= k(
k k > 1) (8.101)
1
T2 B ( II )
where, I and II stand for two isotopes and k = 1.61 for chlorine and k = 1.433 for bromine.
The ratio of the observed excess line broadening when h << 1 is obtained from Eq. (8.81) and is given by
⎛ 1 ⎞
Δ⎜ ⎟
⎝ T2( I ) ⎠ 1 + kξ ( I )
= (8.102)
⎛ 1 ⎞ 1 + ξ (I )
Δ⎜ ⎟
⎝ T2( II ) ⎠
1
where, the dimensionless parameter x (I) refers to the relaxation rate, .
T2 B( )
For h >>1, the above ratio equals unity and for h − ∼1, the ratio becomes a function of h.
In general, we can state that for fast exchange rates, the isotope ratio of relaxation rates equals 1.610 for chlorine
and 1.433 for bromine. For slow exchange rates, this ratio becomes equal to unity.
Further, since the nuclear quadrupole moments and hence the relaxation rates in the halide complexes increase
in the order I >Br >Cl, the fastest exchange reactions which can be studied are iodine-exchange reactions.
The typical chemical exchange systems in which quadrupole relaxation is measured as a function of concentra-
tion and temperature by NMR technique in relation to chemical kinetics are halide ion exchange in metal halide
complexes, e.g. exchange rate of Br− and I− in the complexes with Hg2+ and halide-ion exchange in systems involving
halogen molecules, e.g. iodide-triiodide exchange.
(g) Biological Applications NQR spectroscopy has also entered the homeland of biological sciences. The
biological applications of quadrupole nuclei-NMR roughly deals with studies of halide ion binding to metal free
and metal containing proteins or to other molecules and proteins where an artificial metal label is attached. The
parameters that may be obtained from halide probe techniques are stoichiometry and binding constants of halide
ions, stoichiometry of metal reactive groups on a protein, rotational correlation times (s) of the macromolecule,
quadrupole coupling constant (s) of halide ion (s) in different molecular binding sites, binding constants rates and
activation parameters of halide ion exchange, and stoichiometry of other ions or molecules through competition
experiments. However, before describing the applications, the definition and the role of proteins in biological
systems demands a little attention.
Proteins, the keystones of life, are the most complex substances known to man. Thousands of different proteins
which are composed of thousands of amino acids go into the make up of a living cell. They perform thousands of dif-
ferent acts in the exact sequence that causes the cell to live.
Many amino acids are linked together side by side to form a long protein chain. This linear chain then folds in
a globular fashion.
The linkage between amino acids in a Dis-assembly Unfolding

protein chain is cleavable, therefore, pro-
tein can be dis-assembled in its building
blocks. The assembly of amino acids for
the genesis of proteins and dis-assembly Unfolded
of proteins into amino acids as shown in protein
Fig. 8.14, and are among the most crucial Amino
Globular
acids
processes during the cell cycle. Therefore, protien
any foreign material may cause distur-
Assembly Folding
bances in the cell cycle which ultimately
leads to abnormal functioning of the cell Fig. 8.14 Genesis and dis-assembly of proteins.
in the biological systems.
Combined magnetic field and NQR effects where NQR nearly acts as a perturbation is a useful tool to study
small NQR nuclei binding to native metal-containing or metal-free proteins or other molecules and to proteins or
other molecules where an artificial metal label is attached. This has been demonstrated with the help of halide ion
probe method in which the relaxation of NQR nucleus in the biological system occurs either via chemical exchange
process or correlation times, NQR relaxation times.
When the exchange rate of halide ions between a macromolecule binding site and the bulk exceeds the NMR
relaxation rate in the binding site, then ts, the correlation time at the site, say ‘s’, is neglected. The exchange rates
1
increase usually rapidly with the increase in temperature. This suggests a low free energy of activation for the
τs
exchange process at room temperature. The fraction of ions bound in different sites, Ps, could increase as well as
decrease with temperature. The latter is more probable since the binding of small ions to macromolecules at least
to proteins is accompanied by an enthalpy decrease. However, at elevated temperatures, a conformational change
of macromolecule may occur resulting in new binding sites being accessible for binding. If the latter possibility
is ignored then the halide ion NMR line width decreases with temperature for fast-exchange conditions whereas
increases with temperature for slow exchange conditions.
1 1
The exchange processes are usually studied using either or or in reality usually the halide signal
T1 ( s ) T2 ( s )
1 1
line width as a function of temperature. and become smaller with increasing temperature, the same
T1 ( A) T2 ( A)
1 1
tendency is usually assumed also for and .
T1 ( s ) T2 ( s )
Isotope effects on excess relaxation rates can also be fixed a criterion for the halide ion exchange rates in bio-
logical systems.
For fast exchange processes,
⎛ 1⎞ ⎛ 1⎞
( ) ( )
2
33
Cl ⎜ ⎟ 37
Cl ⎜ ⎟ = ⎡⎣Q 35
Cl Q 37
Cl ⎤⎦ (Eq. 8.99)
⎝ Ti ⎠ ⎝ Ti ⎠
and = 1.610 (Eq. 8.100)
⎛ 1⎞ ⎛ 1⎞
( ) ( )
2
79
Br ⎜ ⎟ Br ⎜ ⎟ = ⎡⎣Q
81
1 79
Br Q 81
Br ⎤⎦
⎝ Ti ⎠ ⎝ Ti ⎠
= 1.433
where, i = 1 or 2.
For slow exchange processes, the isotope ratio of excess relaxation rates is equal to unity.
The parameter ‘molar relaxivity’, i.e. ν i (Hz mole−1) is defined as
Δve = ∑ vi ⎡⎣( M − L )i ⎤⎦ (8.103)

i
for fast exchange processes, may be fixed a criterion for the halide ions interaction with metal binding sites on
macromolecule and also to fix the geometry of complexes so formed. Here, Δve is the excess line width of 35Cl
signal. (M − L)i-stands for a particular complex between M2+ and a ligand L. vi is proportional to the product of
correlation time (tc), square of quadrupole constant, (e2qQ/h)2 and fraction of halide ions bound to a particular M
complex (i).
The importance of molar relaxivities to gain understanding of the interaction of halide ions with metal
binding sites on macromolecules can be realised from the following example. The molar relaxivities of the
Cl− ion interactions with Zn complexes having different bi- tri- and quadridentate ligands as obtained from
35
Cl line width at different pH in 0.5 NaCl, containing Zn2+ and a ligand in appropriate proportions are given
in Table 8.3.
Table 8.3 35
Cl-molar relaxivities ( vi in kHz) for Zn-complexes with different bi- tri- and quadridentate
ligands.
Ligand v (ZnL) v (ZnL2) v (ZnHL)

Bidendate
Cysteine 19.06 0 20.35
Glycine 1.72 0.69 —
Glutamic acid 2.65 3.82 3.82
Tridendate
Aspartic acid 0.54 0 4.26
Iminodiacetic acid 1.64 0 —
Citric acid 0.63 — 1.15
Quadridentate
Nitrilotriacetic acid 2.21 — —
For Zn2+(aq): vi = 2.0 kHz mole−1, ZnL, ZnL2, Zn (HL) are the concentrations of the different liganded species.
HL denotes a protonated ligand.
The following inferences can be drawn from the data in Table 8.3. (i) Except glycine, vi values for 1:1 bidentate
ligand complexes are higher than that of Zn2+(aq). This suggests that the increase of the product vQ2 τ C is more than
sufficient to compensate the expected decrease in the number of bound Cl− ions on chelation. The exceptionally
high value of cysteine is because of high value of vQ for this complex. (ii) For tridendate ligands, the variation
in v suggests that changes in the effective charge of metal ion depend on the nature of the tridendate ligand
bound to the metal ion. (iii) The complexes of quadridentate and bidendate (for 1:2 complex) ligands may have
either tetrahedral or octahedral geometry. In case of octahedral geometry, two water molecules enter into the first
coordination sphere. The large v values for tetracoordinated chelates, Zn (glycinate)2, Zn (glutamate)2 and Zn
(nitrilotriacetate) suggest that the geometry of these complexes is not of coordinately saturated tetrahedral type.
On the other hand, v value of zero for Zn (cysteine)2 complex suggests that it has a tetrahedral type of geometry
(binding of Cl− ion is not included).
The interaction of halide ion with proteins can be obtained from the halogen-NMR excess relaxation rates or
excess line width Δνe. It has been demonstrated by the following examples: (i) Carbonic anhydrase is a metal-
loenzyme. It has one zinc atom per protein molecule of a molecular mass of about 30,000. It catalyses the revers-
ible hydration of carbon dioxide to form bicarbonate in the biological systems. 35Cl-NMR studies on the zinc
free solutions of bovine carbonic anhydrase (in aqueous NaCl) and on the enzyme containing zinc shows that
line broadening is observed in the latter relative to the former. This is attributed to the coordination of Cl− ion
to zinc of the enzyme. Further, line narrowing takes place when the enzyme solution is titrated with acetazole-
amide (a strong inhibitor of enzyme) and cyanide and the equivalence point is observed at stoichiometric ratio of
acetazoleamide : carbonic anhydrase :: 1 : 1. This indicates that these ligands completely eliminate the Cl− ions
bound to the enzyme molecule. The pH dependence of line width also shows that at high pH, the Cl− ions gets
eliminated.
The —S —S —cross linkages of blood serum albumin (molecular mass = 65000) are responsible for resistance
towards perturbations from environments. 35Cl-NMR studies on excess line width (Δve) of NMR signal reveal that
the anion binding follows the sequence
− − − −
O42− < F− < CH3COO− < Cl− < Br− < NO3 < I < ClO 4 < SCN < (CN ) 4 < Au (CN)−2 << CH3(CH2)11 OSO3 .
2− −
SO
for high as well as low affinity binding sites of the protein.
35
Cl-NMR excess line width (Δve) as a function of pH (4.8−10.5) in solutions of human oxy and carbonmonox-
ide haemoglobin in 0.5 M NaCl indicates that the line width increases with the decrease in pH as expected since
the number of positively charged groups on the protein molecule increase at lower pH. At about neutral pH, the
consistently larger line width for carbon monoxide haemoglobin (HbCO) relative to deoxyhaemoglobin (Hb) is
interpreted in terms of the finding that in oxyhaemoglobin (HbO2) conformation a number of interchain and inter-
chain salt links are broken and subsequently more charged residues will be available for the interaction with other
ions. Further 35Cl-NMR line width in 0.2 M NaCl decreases markedly as the concentration of ATP in the solution
is increased at a pH about 6.0 to 6.3. This suggests that ATP and Cl− compete for some of the high-affinity bind-
ing sites. The binding constant is about 6 × 103 mole−1 for ATP-Hb and 4 × 102 mole−1 for ATP-HbO2 systems, the
order of binding constant being higher than for Cl−.
Horse liver alcohol dehydrogenase (LADH), a zinc metalloenzyme, catalyses the reversible interconversion of
several alcohols and the corresponding aldehydes in the presence of a coenzyme which is reduced (NADH) or
oxidised (NAD+) nicotinamide adenine dinucleotide. The titration of solution containing LADH (in aqueous NaCl
⎛ 1⎞
or NaBr) with NADH has been followed by 35Cl or 79Br longitudinal relaxation rates ⎜ ⎟ . The decrease in excess
⎝ T1 ⎠
− −
bandwidth on gradual addition of NADH suggests the elimination of Cl or Br ions bound to Zn of LADH. The
complete elimination of halide ion occurs at a molar ratio of NADH : LADH :: 2 : 1.
The interaction of a mercury-labelled hapten (I), with antidinitrophenyl anti-
NO2
body can be studied by the 35Cl-probe method. The increase in 35Cl-NMR line
width with the increase in the concentration of hapten is due to rapid relax- O2N N HgCl
–
ation of 35Cl interacting with Hg in the hapten-antibody complex. 35Cl-NMR H
line width as a function of the molar ratio of hapten (2, 4-dinitro- (I)
4′-(chloromercuri) diphenylamine) to antidinitrophenyl antibody
suggests the stoichiometric ratio to be 2 : 1 (Hapten : antibody).
−
Note that the line broadening may be dominated by Cl exchange Macro-
instead of relaxation. molecule
Further it is to be kept in mind that when halide ion is liganded θ2
to an Hg-label attached to a macromolecule, the halide ion at the
binding site may experience internal motion with respect to the
θ1
overall motion of the macromolecule. This is illustrated schemati-
cally in Fig. 8.15.
⎛ 1 ⎞
When the rate of internal rotation ⎜ ⎟ >> overall motion (X)
⎝τ ⎠
introt Fig. 8.15 Schematic diagram of halide nucleus (X)
of the macromolecule tR (i.e. tintrot << tR), the effective quadru- attached to a binding site on a macromolecule with
pole interaction in the Hg−X bond is reduced by a factor of the possibility of rotation around two axes.
1
( )
2
3 2 θi 1 for each mode of internal rotation, q1 is the angle between the Hg−X bond (assuming h = 0) and
4
the first internal rotation axis, q2 is the angle between the first and second rotational axes, etc. When q equals the
tetrahedral angle (109°28′), the above factor becomes 0.11. This result is valid for either unrestricted rotational
diffusion around the axis of internal rotation or of random jumps between three equally deep potential minima
around the C3-axis. In practice, steric interactions between the halide label and neighbouring group of atoms may
restrict the internal rotation and give rise to potential minima of appreciably different depths around the axis of
internal rotation. Thus, in such a case, the effects of internal rotation on the quadrupole coupling constant may be
partly quenched.
8.14 APPLICATIONS OF PURE NQR

The transition frequency in the pure quadrupole spectrum of spin 3/2 particles like chlorine and bromine nuclei
is given by 1/ 2
Q ⎛ η2 ⎞
e 2 Qq
v= 1+ ⎟ (Eq. 8.56)
2h ⎝ 3⎠
for a polycrystalline powder. For spin I = 3/2, h must be obtained from studies on single crystals, whereas for
127
I with I = 5/2 both vQ and h may be obtained from the spectrum of polycrystalline powder. For gases, the
most important method for obtaining νQs’ is microwave spectroscopy. However, only a small number of halogen
quadrupole coupling constants are known for the gaseous state.
8.14.1 Study of Chemically Inequivalent Quadrupole Nuclei in the Unit Cell

Chemically inequivalent quadrupole nuclei in solids may be attributed to interaction between the neighbouring
molecules or the presence of different conformers of the same molecule in the unit cell. Chemical inequivalence
may also arise when an atom containing the quadrupole nucleus is present at chemically different sites in the free
molecule. Thus, the chemically inequivalent NQR active nuclei in solids should give rise to multiple lines in the
NQR spectrum and the same has been observed indeed, e.g. the 35Cl-NQR spectrum of trichloroacetyl chloride
exhibits four lines at 33.721, 40.132, 40.474 and 40.613 MHz. The appearance of four lines suggests that four
chlorine atoms of the molecule are inequivalent. The first line is attributed to chlorine atom of the −COCl group
whereas the last three lines are attributed to inequivalence of three chlorine atoms of −CCl3 group caused by
intermolecular interactions. The expected numbers of lines in the 35Cl-NQR spectrum of dichloroacetylchloride
is three but the observed spectrum contains six lines at 32.147, 32.962, 38.353, 38.521, 39.189 and 39.386 MHz.
The two low-frequency and four high-frequency lines suggest that there are at least two types of chemically
inequivalent conformers in the unit cell. Similarly, the two high and low frequencies (30.437 and 37.517 MHz)
of monochloroacetylchloride are interpreted in terms of two chemically inequivalent conformers in the unit cell.
Oxalyldichloride (ClCO−COCl) shows only one line (at 33.621 MHz) in its 35Cl-NQR spectrum suggesting that
the two NQR active Cl nuclei in the unit cell are chemically equivalent. The four-line multiplet in the halogen qua-
drupole spectrum of each SiBr4, GeCl4, GeBr4, SnCl4, SnBr4, SnI4, etc., is attributed to the chemical inequivalence
of nuclear sites in the free molecule.
8.14.2 Nature of Chemical Bond

Pure NQR tool is frequently used to study the electronic configuration of molecules. The extent of hybridisation
and the ionic character in the covalent bond is obtained from the comparison of the quadrupole coupling constants
of the nucleus under consideration in the molecule and the isolated atom. A number of relationships between the
molecular and atomic quadrupole constants of the same nucleus to describe the covalent bond involving quadru-
pole nucleus are expressed as
⎛ e 2 Qq ⎞ ⎛ e 2 q0 Q ⎞
⎜⎝ h ⎟⎠ (
= 1 2
) ( )⎜
1 − i
⎝ h ⎟⎠ atom
(8.104)
molecule
where, a is the fractional s-character of the bonding hybrid orbital on the quadrupole atom and ‘i’ is a measure of
the ionic character of the covalent linkage.
For halogen nucleus in the molecule and a free halogen atom, the relation is,
⎛ e2 qQ ⎞ ⎛ e2 q0 Q ⎞
⎜⎝ h ⎟⎠ = ⎡⎣1 s d i (1 s − d )⎤⎦ ⎜ (8.105)
molecule
⎝ h ⎟⎠ atom
Here, d stands for the d-character of the bonding orbital. vQ for 35Cl(3p) = +109.746, 37Cl(3p) = +86.510,
79
Br (4p) = −769.76, Br81 (4p) = −643.032 and 127I (5p) = +2292.712 MHz.
For an electropositive atom, ‘i’ changes sign so Eq. (8.105) is written as
⎛ e 2 qQ ⎞ ⎛ e 2 q0 Q ⎞
⎜⎝ h ⎟⎠ = ⎡⎣1 s d i (1 s − d )⎤⎦ ⎜ (8.106)
molecule
⎝ h ⎟⎠ atom
Note that for a molecule AB, ψ = aψA + bψB and i = a2 − b2 if A is more electronegative than B and ψ is the
bonding orbital for the molecule AB.
If p-bond is present in the molecule, Eq. (8.105) becomes
⎛ e 2 qQ ⎞ ⎛ e 2 q0 Q ⎞
⎜⎝ h ⎟⎠ = (1 − s d − i − π ) ⎜ (8.107)
molecule
⎝ h ⎟⎠ atom
The exact nature of chemical bond cannot be deduced from NQR data since a large number of unknowns are
involved in the above expressions. As the ionic character in a molecule increases, the electronic environment tends
towards spherical symmetry, i.e. q → 0, consequently the molecular quadrupole coupling constant decreases.
sp hybridisation also causes a decrease in the value of molecular quadrupole coupling constant.
The contribution of d orbital to the bonding orbital increases the electric field gradient and hence molecular
quadrupole coupling constant. In such a case, the asymmetry parameter h also becomes appreciable and cannot
be ignored.
Let us now see how NQR data are utilised to study the nature of chemical bond in H2S. The electronic configu-
ration of sulphur is [Ne] 3s23p4 and the bond angle in H2S in about 90°.
Prior to the 33S -NQR studies on H2S, it was assumed that in H2S only p-orbitals of sulphur participate in bonding.
However, NQR studies reveal that H2S has a large value of asymmetry parameter, i.e. h = − 0.60. Theoretically, h can
be expressed in terms of electron population in the px, py and pz orbitals of the atom. The components of molecular
electric field gradient along the three axes in terms of electron population are given by
⎛ e 2U xx Q ⎞ ⎛ Ny Nz ⎞ ⎛ e 2 q0 Q ⎞
⎜⎝ h ⎟⎠ = −⎜ − Nx⎟ ⎜
molecule
⎝ 2 ⎠ ⎝ h ⎟⎠ atom
⎛ e 2U yy Q ⎞ ⎛ Nx Nz ⎞ ⎛ e q0 Q ⎞
2
⎜ ⎟ = − ⎜⎝ − N y⎟ ⎜
⎝ h ⎠ molecule 2 ⎠ ⎝ h ⎟⎠ atom
⎛ e 2U zz Q ⎞ ⎛ Nx Ny ⎞ ⎛ e 2 q0 Q ⎞
= −⎜ − Nz⎟ ⎜ (8.108)
⎜⎝ h ⎟⎠
molecule
⎝ 2 ⎠ ⎝ h ⎟⎠ atom
where, Nx, Ny, and Nz are the effective electron populations in the px, py and pz orbitals of the atom in the molecule.
The set of Eqs (8.108) yields
h = (Uxx − Uyy) Uzz /
=
3 ( − )
N x + N y − 2N z
If pz contains the lone pair, we have Nz = 2 and 1.50 Ny = Nx +1.
This suggests that Nx cannot be equal to Ny , otherwise h will become zero. Therefore, pure p-orbitals of sulphur are
not involved in bonding. The large value of h is interpreted in terms of 15 per cent d character in the bonding orbitals.
Thus, p3.40 d0.60 hybridised orbitals rather than pure p-orbitals of sulphur are involved in bonding in the H2S molecule.
8.14.3 Structural Information from NQR Spectrum

The similarity in the number and pattern of lines in the NQR spectra of different molecules having the same
anions but different cationic part can be taken as a criterion to obtain information about their crystal structure, e.g.
NQR spectrum of each AlBr3 (97.497; 113.790, 115.450 MHz) and GaBr3 (120.602, 120.635; 168.375, 168.430
MHz) has three lines, at 77 K two of which are closely spaced whereas the third one is far apart from the doublet.
According to the principle of similarity in the NQR spectra, they must have similar structure. The same has indeed
been observed on the basis of x-ray diffraction studies on these molecules. They exist as bridged dimers rather
than monomers. Thus, the two closely spaced lines correspond to bromine atoms attached at the end positions of
the bridge and the third line is due to bromine atoms in the bridge sites.
The doublet in the 14N-NQR spectrum of solid BrCN may be attributed either to two inequivalent nitrogen
atoms in the crystal lattice or to splitting of the nitrogen resonance signal. The former possibility is ruled out
since single crystal x-ray diffraction studies suggests a linear chain structure of the type Br—C≡N| ... Br—C≡N| ...
Br—C≡N|. Accordingly, nitrogen has axial symmetry and only one line is expected. Therefore, the only alterna-
tive left to explain the doublet in the spectrum is in terms of induced asymmetry at the nitrogen nucleus arising
because of interactions between the chains.
8.14.4 Study of Charge-Transfer Complexes

The advantage of nuclear quadrupole resonance spectroscopy in the study of charge transfer complexes, particu-
larly when the crystal structure is known, is that the changes in NQR frequency produced by complex formation
can provide evidence of the direction and extent of electron transfer.
Electronic absorption spectroscopy shows the formation of charge-transfer complexes in CCl4—p-xylene
and CBr4—p-xylene systems. However, the NQR(35Cl/Br79˙ Br81) spectra of frozen equimolar solution of these
systems show that the resonance frequencies of the halogens in the mixture (CCl4—p-xylene: 40.472, 40.689 MHz,
CBr4—p-xylene: 264.95, 269.18 MHz) are not appreciatively different from the halogenated component CCl4 :
35
Cl: 40.465 to 40.817 MHz (15 lines); CBr4 : 79Br : 317.00, 322.09, 81Br : 265.45, 266.06, 266.4, 267.06, 267.97,
268.30, 269.25 MHz, suggesting thereby that charge transfer complexation does not occur in the ground state.
The crystal structure of SnCl4 ⋅ 2POCl3 complex, measured at room temperature shows that the tin atom has
octahedral configuration with the POCl3 groups occupying cis positions and bonded to the tin by oxygen (i.e.
(Cl3P==O)2 → SnCl4). 35Cl-NQR frequencies (in MHz at 77 K) of SnCl4 and POCl3 are observed at 24.294,
24.226, 24.140, 23.719 and 28.9835, 28.9378 respectively. The 35Cl-NQR frequencies in the SnCl4 ⋅ 2POCl3 com-
plex appear at 30.213, 30.117, 21.146, 19.807, 19.035 MHz at 77 K. If possible p-bonding in the P−Cl and Sn−Cl
bonds is neglected, both the magnitude and sign in the shifts of the resonance of the chlorine atom attached to tin
(−3 to −4 MHz) are in accordance with the structure (Cl3P—O)2 → SnCl4, which leads to greater ionicity of the
Sn—Cl bond. The positive shift (+1 MHz) in the POCl3 35Cl resonance frequency is also in accordance with the
above hypothesis. The smaller magnitude in the shift is because of the fact that chlorine atom is one bond further
removed from the site of the complex formation, i.e. Cl—P—O … Sn—Cl.
8.14.5 Study of Phase Transitions

NQR frequencies undergo abrupt changes at the transition temperature. These changes in the number and pat-
tern of frequencies occur because the electron distributions within the molecules are affected by intermolecular
interaction and depend upon the environment around each molecule, e.g. 79Br-NQR spectrum of K2SeBr6 exhibits
a single line at room temperature and two lines at dry ice temperature. This difference in the number of lines is
attributed to crystalline phase change in going from room temperature to dry-ice temperature. At dry-ice tempera-
ture, there are at least two different bromide environments in the unit cell. The clearly visible 35Cl-NQR resonance
frequencies (30.213, 30.117, 21.146, 19.807, 19.035 MHz) of SnCl4. 2POCl3 complex at 77 K suddenly disappear
at 183 K, This is attributed to the phase change in going from 77 to 183 K.
8.14.6 Study of Ionic Character of Chemical Bonds in the Gaseous

and Solid States
A comparison of halogen (35Cl, 79Br, 127I)-nuclear quadrupole coupling constants in halomethanes in relation
to ionic character of chemical bonds both in the gaseous and solid states met with great success, e.g, the 35Cl-
quadrupole coupling constants of CH3Cl in the solid state (−68.4 MHz) is large relative to in the gaseous state
(−74.74 MHz). The decrease of quadrupole coupling constant on vaporization suggests that the ionic character of
C—Cl (C+Cl−) bond is more in the solid state as compared to in the gaseous state. This increase of ionic character
in going from gaseous to solid state is attributed to enhancement in the Madelung constant (M) in the binding or
cohesive energy (U) of an ionic crystal.
⎡ − N A Me 2 ⎤
U=⎢ ⎥ + Ae p ( a / ρ) (8.109)
⎣ a ⎦
1
where, A and r are constants, a is the equilibrium lattice parameter and M = ∑ ± where ri’s are the distances
i ri
from the origin to the points of a lattice with unit lattice parameter. The same has been observed in ICl (+ICl−).
Further, it is observed that in CFCl3 the difference between quadrupole coupling constants in the gaseous and
solid states (−78.05 − (−77.58) = −0.47 MHz) is small as compared to that in CH3Cl (−6.34 MHz). This indicates
that C—Cl bond in CH3Cl becomes more ionic on solidification relative to the C—Cl bond of CF3Cl in the solid
state, the reason being carbon is more electropositive in CFCl3 relative to CHCl3 where it is slightly electrone-
gative with respect to hydrogen. Moreover, on solidification, both the C—Cl and C—F bonds of CFCl3 tend to
acquire more ionic character since F and Cl atoms compete to remove electrons from the carbon atom. F being
more electronegative than Cl, the C—F bond will have less ionic character relative to C—Cl bond.
8.14.7 Partial Double-Bond Character and Sigma Electron Population

for Chlorine Atom Bonded to an sp2 Hybridised Carbon Atom
The partial double-bond character of the C—Cl bond, and sigma electron population for chlorine atom in conju-
gated organochlorine compounds, namely substituted chloroethylenes, chlorobenzenes, chloropyridines, chlora-
nil, etc. can be estimated from their nuclear quadrupole coupling constants.
The fractional p-bond character rp is related to the molecular nuclear quadrupole coupling constant by
2 e 2 Qq
ρπ η per cent (8.110)
3 e 2 Qq0
where, e2Qq0 for 35Cl (atom) = −109.74 MHz.
The total p-electron population on the chlorine atoms in substituted chlorobenzenes is given by
eQq
∑ρ P 6
e Qq0
[η + 1] (8.111)
The sigma electron population rs for chlorine atom bonded to an sp2 hybridised carbon atom in organochlorine
compounds is given by
e 2 Qq ⎛ η⎞
ρσ = 2 − ⎜⎝1 + ⎟⎠ (8.112)
e 2 Qq0 3
Problem 8.12: (A), Calculate the partial double-bond character of C—Cl bond and sigma electron population
for chlorine atom bonded to an sp2 hybridised carbon atom in (a) H2C = CHCl (b) O = CCl2 (c) 1, 2, 4, 5-Cl4C6H2
(d) 2, 4, 6-trichlohloro-s-trizene; from the following data.
Compound v (MHz) h
(a) 36.0 0.14

(b) 36.23 0.25
(c) 36.283 0.116
(d) 36.30 0.23
(B) Calculate the total p-electron population on chlorine atoms in (a) p-dichlorobenzene, (b) p-hydroxychloroben-
zene, and (c) p-aminochlorobenzene from the data given below:
Compound v (MHz) h
(a) 34.25 0.067
(b) 34.367 0.058
34.145 0.223
(c) 33.767 0.0489
Solution (A) From Eqs (8.110) and (8.112), we obtain for

(a) H2C = CHCl
2 72
ρπ = 0.14 × 100 = 6.12
3 109.74
72 ⎛ 0 14 ⎞
ρσ = 2 − ⎜⎝1 + ⎟ = 1.313
109.74 3 ⎠
(b) O = CCl2
2 72.46
ρπ = 0.25 × 100 = 11.00
3 109.74
72.46 ⎛ 0 25 ⎞
ρσ = 2 − ⎜⎝1 + ⎟ = 1.284
109.74 3 ⎠
2 72.566
(c) ρπ = 0.116 × 100 = 5.11
3 109.74
72.566 ⎛ 0.116 ⎞
ρσ = 2 − ⎜⎝1 + ⎟ = 1.313
109.74 3 ⎠
2 72.60
(d) ρπ = 0.23 × 100 = 10.14
3 109.74
72.60 ⎛ 0 23 ⎞
ρσ = 2 − ⎜1 + ⎟ = 1.287
109.74 ⎝ 3 ⎠
(B) From Eq. (8.111), we obtain for
(a) p-dichlorobenzene
68.50
∑ρ p = 6 − (1.067) = 5.333
109.74
(b) p-hydroxychlorobenzene
68.734
∑ρ p1 =6− 058) = 5.337
(1.058
109.74
68.290
∑ρ p2 =6− 223) = 5.238
(1.223
109.74
67.534
(c) ∑ρ p =6− (1.0489) = 5.354
109.74
PROBLEMS
1. Determine the spin of the quadrupole nuclei if the 5. x-ray diffraction studies on the halides of group three,
number of NMR lines in their respective spectra is (a) i.e. MX3 (M = In; X = Cl, Br, I) show that these halides
four, (b) five, and (c) seven, exist as bridged dimers in which there are two types of
−
2. Calculate the quadrupole relaxation times for (a) 81Br , halogen atoms corresponding to the bridge and end
127 −
and (b) I ions at infinite dilution in water from the fol- sites. The 35Cl-NQR spectrum of InCl3 consists of three
lowing data. lines (at 104.112, 128.110 and 128.608 MHz at 77 K)
two of which lie very close to each other whereas the
third line is far away from the doublet. Interpret the NQR
τ
Nucleus Q × 1024 (cm2) 1+g∞ r0(Å) H 2 O ×1012 (s) spectra in the light of X-ray studies on MX3 systems.
6. Calculate the paramagnetic shielding term in Cl2 mol-
Br−
81
0.28 100 3.36 2.5 ecule if the mean excitation energy ΔE for the 1∑ − 1P
127 −
I −0.75 180 3.56 2.5 transition in Cl2 is (a) 3.76 eV, and (b) 10 eV.
[Ans: (a) sp(Cl) = −17.4 × l0−4 [b] −6.55 × l0−4 is more rea-
sonable]
Hint: Assume P = 0.5
7. F-NMR spectrum (58.3 MHz, 26°C, CFCl3) of BrF+6 in HF
19
[Ans: (a) 7.4 × 10−4 s (b) 1.77 × 10−4 s]
solution consists of two overlapping 1 : 1 : 1 : 1 quartets at
3. Compute the values of correlation times of 35Cl nuclei from
very low field. Splittings within the quadruplets have also
the NQR data of the systems given below. What inference
been observed. The values of two coupling constants
can be drawn from the correlation times so obtained?
J79Br −19 F and J81 have been found to be 1575 and 1679
Br −19 F
ppm respectively. Interpret the spectral data.

System T2(m s) vQ (MHz) Temperature (K)
79
CHCl3 24.5 76.6 298 [Ans: BrF6+ and 81BrF6+ ; Large coupling constants means
CHCl3⎯C6H6 16 79 298 F6+ has octa-
Br has long relaxation times, hence BrF
hedral symmetry. Splitting between the quadruplets
complex is because of bromine (I = 3/2) spin coupling to
CHCl3 bound to 3.3 77 300 equivalent fluorines].
polyvinyl 8. The coupling constants of halomethane series are tabu-
pyrrolidone lated as follows.
[Ans: 1.76, 2.53 and 12.9 p s] Halomethane vQ(MHz)

4. PbCl4, SnCl4 and TiCl4, exhibit four lines in their CH3Cl 68.0
respective 35Cl-NQR spectra. The values of na in the
respective systems are 45.4, 48.2 and 12.2 MHz. CH2Cl2 72.0
Comment on the number of lines in the spectra and CHCl3 76.6
what inference can be drawn regarding the structure
CCl4 81.2
of these molecules?
Calculate the average value of the increase in qua- (b) Assign also names to the chlorobenzenes.
drupole coupling constant of chlorine atom in the [Ans: 1, 2, 3-trichloro − and 1, 2, 4, trichlorobenzenes]
halomethane series and comment. [Ans: ∼
− 4 MHz]
9. Predict and explain the ionic character of C⎯X(X =
Br, I) bond of each of the following pairs of halogenated Chlorobenzene 35
Cl-NQR frequencies (MHz)
methanes both in the gaseous and solid states from
their respective quadrupole coupling constants. Trichlorobenzene (a) 35.189, 35.6l7, 36.173, 36.380,
36.400, 36.623
Pair Nucleus vQ (MHz) Trichlorobenzene (b) 35.545, 35.894, 36.115
Gas Solid
16. Calculate the 35Cl-NQR frequencies of pentachloro-
CH3Br 79
Br 577.15 529 and hexachloro benzenes
CF3Br 79
Br 619 604 [Ans: 37.072 MHz (Obs (average) 37.760 MHz);
38.372 MHz (Obs (average):38.422 MHz)]
CH3I 127
I −1929 −1766
17. Calculate the partial double-bond character of C⎯Cl bond
CF3I 127
I −2143.8 −2069 and sigma electron population for chlorine atom bonded
to an sp 2 hybridised carbon atom in (a) H2C == CFCl
10. Determine the 35Cl-quadrupole coupling constant for (b) o ⎯ ClC6H4OH (c) chloranil, from the following data.
Cl− counter ion bound to N+H3 group in the amphiphilic
[Ans: (a) 3.75,1.323 (b) 4.23, 1.342 (c) 9.38,
octacylammonium chloride. (C8N+H3 Cl)−H2O system 1.282; 9.40, 1.281]
from the following data, g∞ (for chlorine) = 59, D >>1, r
(the distance between Cl− bound to amphiphilic sur- Compound u (MHz) η
faces and amphiphilic positive charge, i.e. N+H3 end (a) 36.35 0.085
group) = 6.0 × 10−10 m, Q (for 35chlorine) = −0.0802 (b) 34.84 0.10
× 10−28m2. The Cl− counterions are assumed to be
(c) 36.79, 36.86 0.21, 0.21
hydrated [Ans: 0.6 MHz)
11. Calculate the 81Br-quadrupole coupling constant for Br−
+ +
counter ion bound to (CH3)3 in the amphiphilic (C16 N 18. The 35
Cl-NQR spectrum of X Clgives
(CH3)3Br−) − H2O system from the following data: (1 +
+
− uQ = 69.116 MHz. Determine the Hammett −s constant
g∞) for bromine = 110, D >>1, r[Br − N (CH3)3] = 7.5
× 10−10m, Q (for 81bromine) = 0.282 × 10−24cm2. The Br− for the substituent X. [Ans: s = 0.268]
counter ions are assumed to be hydrated.
19. Calculate (a) Cl-NQR frequency for p-aminochlo-
35
[Ans: 2.0 MHz] robenzene, (b) 81Br-NQR frequency for p-aminobromo

12. Draw the energy-level diagram depicting the amount of (81) benzene (c) 127I-NQR frequency for p-aminoiodo-
shift in the Zeeman levels due to first-order quadrupole benzene. spara(NH2) = −0.66
interaction for a nucleus with I = 5/2.
[Ans: (a) 34.166 MHz (obs: 34.146 MHz)
13. Calculate the quadrupole relaxation rates of 81Br − and
127 −
(b) 221.899 MHz (obs:22 1.862 MHz)
I ions in infinite dilute solutions of H2O from the data
given below: (c) 266.060 MHz (Obs: 261.14 and521.125 MHz)]
20. At 20 K two quadrupole resonances are observed for
Nucleus Q × 10 (cm ) 1+ g∞ 24 2
r0 (Å) H2O × 10 (s)
12 t-butylchloride, (CH3)3CCl, the stronger at 31.195 MHz
and the weaker at 24.586 MHz. Interpret these obser-
81
Br − 0.28 100 3.36 2.5 vations.
127 −
I −0.75 180 3.56 2.5 [Ans: Cl and 37Cl resonances, |νQ|: 62.39 and 49.17
35
MHz respectively, rp = 5.431]

[Ans: 1348 s−1, 5635 s−1] 21. (a) 127
I-nuclear quadrupole coupling constants (MHz)
⎛ 14
⎞ of NH4I3 are 466.8 (h = 0.130), 1725.0 (h = 0.151),
14
14. N HC = N- NH quadrupole spectrum of 1, 2, 4-triazole and 2458.7 (h = 0.0 18). Determine the rs popula-
⎝ ⎠
at 77 K shows three lines at 3.8759, 2.4692 and 1.4070 tion in the linear I−3 ion.
MHz. Find the values of quadruple coupling constant (b) 35Cl and 127I quadrupole coupling constants (MHz)
and asymmetry parameter. in KICl2 are 36.81, 37.53, 38.00, 39.25 and 3081,
[Ans: uQ = 4.2301 MHz, h = 0.6651] 3129 respectively. Calculate the rs population in the
linear ICI−2 ion.
15. Predict (a) the positions of chloroatoms with respect to
the chlorine atom whose frequency is being measured Hint: Take the average of coupling constants in Part (b).
experimentally for the following trichlorobenzenes. [Ans: (a) 1.78, 1.20, 0.92(b) 1.65, 0.65, 0.635]
CHAPTER 9 ELECTRON SPIN
No hypothesis can lay claim to any value unless it assembles many phenomena under one concept.
—Goethe
Immerse yourself in facts, especially the baffling facts, in the hope that illumination would come.
—Charcot
9.1 INTRODUCTION
Electron Spin Resonance (ESR), also called Electron Paramagnetic Resonance (EPR), was discovered in Kazan
(USSR) in 1944 and the first experiments dealt with absorption resonance in salts of iron group ions. It deals with
the study of absorption of electromagnetic radiation in the microwave region (0.1−100 cm) by free radical(s) under
the influence of an external magnetic field. A free radical is defined as an atom, a group of atoms, or a molecule
in a certain state containing one unpaired electron which occupies an outer orbital, e.g. atomic hydrogen (H•),
hydroxyl (HO•) or methyl (H3C•) radicals. By definition, naturally occurring stable and long-lived molecules with
odd number of electrons in their normal state, e.g. nitrous oxide (NO), nitrogen dioxide (NO2), nitrogen difluoride
(NF2), di-t-butylnitroxide [(CH3)3CN•OC(CH3)3], chlorine dioxide (ClO2) are free radicals. A biradical contains
two unpaired electrons in outer orbitals, e.g. methylene radical H2C: or oxygen (O2) in the ground state. A radical
ion is a freei radical with positive or negative charge, e.g. protonated amine radical (H3N+•) or naphthalene anion
radical(C10 H8 ). Some of the transition metal ions and their complexes, e.g. Mn2+, Cr3+, Co2+, Fe3+, [Fe(CN)6]3−,
Cu2+ copper (II) acetate monohydrate, etc., have also unpaired d electrons and have, therefore, some properties
common with free radicals. Although all transition metal ions have unpaired electrons, but according to a theorem
by Kramer, the most suitable ions for ESR study are those containing an odd number of unpaired electrons.
Naturally occurring free radicals, being stable and long-lived, can easily be studied by ESR whereas free radicals
formed as intermediates in a chemical reaction or produced artificially by irradiation of a normal molecule with
electric discharge or UV or x-ray or with a beam of nuclear particles are unstable, and can be detected by ESR
provided their lifespan is >10−6 s. The lifetimes of many organic free radicals and ionic radical intermediates are
too short for observation by ESR spectroscopy. Such free radicals and radical ions may be stabilised by the fol-
lowing methods: (i) stabilisation of free radicals by cooling to low temperatures, e.g. liquid nitrogen (− 196°C =
77 K). (ii) stabilisation of free radicals on synthetic zeolites. They are good stabilising matrices for free radicals
at relatively high temperatures. (iii) Stabilisation of free radicals and ionic radicals in organic glass matrices. Free
electrons are trapped only in amorphous frozen (glassy) states but never in crystalline states, e.g. ESR spectra
of γ-irradiated 3-methyl pentane glass at 77 K. (iv) Stabilisation of free radicals in organic crystalline matrices,
e.g. alcohols. Methanol and alcohol glasses are frequently used as trapping matrices in radiation and photochemi-
cal studies. Generally, electrons and hydrogen abstraction radicals RC•HOH of the alcohols are formed by ionis-
ing radiation; so alcohol glasses as trapping matrices should be used very cautiously.
9.2 SIMILARITIES BETWEEN ESR AND NMR

ESR is very similar to NMR in theory. The discussion and equations on NMR are equally valid for ESR except
that the system under consideration is an electron rather than proton.
The electron is characterised by its rest mass, m = 9.109558 × 10−28 g; charge e = 4.803 × 10−10 esu, intrinsic
⎛ 1⎞
angular momentum or spin ⎜ S = ⎟ and its magnetic moment (me).
⎝ 2⎠
The magnetic moment for the unbounded electron is related to its magnetic spin quantum number (MS) by the
following equation:
μe = − g e βs M S erg G −1 (9.1)
where, ge is dimensionless constant called the Lande g-factor of the electron or electron free spin g-factor
or spectroscopic splitting factor of the electron, ge = 2.0023192778, be is the Bohr magneton of the electron,
eh ⎛ 1⎞
βe = = 9.2740 × 10 −21 erg G −1 , where c is the velocity of light = 2.997925 × 1010 cm s−1. For ⎜ S = ⎟ ,
4π m e c ⎝ 2⎠
⎛ 1 ⎞ 1
MS = (2S + 1) = ⎜ 2 × + 1 2. Thus, the two values of MS are ± .
⎝ 2 ⎠ 2
The minus sign in Eq. (9.1) indicates that the magnetic moment vector of an electron is in opposite direction to
the angular momentum vector. The energy of the electron spin state is given by
E g e βe H M S (9.2)
H is the strength of the applied magnetic field.

In the absence of a magnetic field, the electron may exist in one of the two degenerate states corresponding to
1
M S = ± . However, when the electron is placed in an external magnetic field (H), it precesses about the applied
2 1
field axis (Z-axis) with a component of its spin angular momentum either parallel ( M s = − ; low energy state
1 1 1 2
of − g β H ) or antiparallel (M s = + , high energy state of + g β H ) to applied field axis. The oscillating
2 2 2
magnetic field of electromagnetic radiation having frequency, ν, when applied at right angles to the applied field
axis (Z-axis) induces transitions between the two spin states provided the frequency of the field is at or near the
Larmor frequency of the precessing electron, i.e. ΔE = hv = ge be H. Since the ratio of Larmor frequency of elec-
tron to proton is
ω( ) γ( )
= = 103 (9.3)
ω( ) γ( )
the NMR experiments are carried out in the MHz range (radio frequency region), ESR experiments require
frequencies in the GHz range (1 GHz = 109 Hz = 103 MHz), i.e. microwave frequency region. In spite of the
difference in the frequency regions of NMR and ESR, the s-s and s-l relaxation times play a key role in deter-
mining the nature of ESR signal. Although, as in NMR, the ESR spectrum can be obtained by varying the
frequency or the field strength, in practice it is simpler to vary the latter.
More in common with the methods of pure quadrupole resonance and nuclear magnetic resonance, e2qQ, η and
the directions of the field gradient axes can all be determined by electron spin resonance measurements conducted
on single crystals. In favourable cases, the absolute sign of coupling constant can be determined. The most general
application of this method is the determination of quadrupole coupling constants of paramagnetic complexes of
transition metals and the lanthanons.
Let us now see how an unpaired electron in different chemical environments behaves under the influence of
an external magnetic field. But prior to that, certain features of ESR of bare electron and the various parameters
associated with it deserve consideration.
9.3 BEHAVIOUR OF A FREE ELECTRON IN AN EXTERNAL MAGNETIC FIELD

The electron energy (E) of an electron under the influence of magnetic field (H) is
E = ge be H MS (Eq. 9.2)
1
In the absence of a magnetic field, the two spin states of the electron corresponding to M S = ± will have
2
equal energy and are said to be degenerate. However, on the application of a magnetic field, the two MS states split.
The down spin of an electron is more stable than the up spin because of electron negative charge. The high-energy
1 1
state will thus correspond to M S = + whereas the low-energy state will correspond to M S = − . The selection
2 2
rule for the transition between spin states is Δ S = ±1. If E+1/2 and E−1/2 are energies of higher and lower states of
the electron respectively, then from Eq. (9.2) we write
1
E g e βe H (9.4)
2
1
E g e βe H (9.5)
2
ΔE = E +1/ 2 − E − = g e βe H (9.6)
Now, when the electron in magnetic field (H) is irradiated with microwave radiation of frequency ν the energy
1
hν will be absorbed by the electron in state M S = − and the transition between the two spin states will occur
provided 2
ΔE = hν = g β H
Δ (9.7a)
This is the condition for resonance in ESR. The transition between the spin states and the ESR signal correspond-
ing to it is shown in Fig. 9.1.
(a)
Ms = + 1
2
E+1/2
Ms = + 1
2
E=0 ΔE = hν = geβeH
Ms = − 1
2
H=0
E−1/2
(b) Ms = − 1
2
Energy
Absorption
Spectrum
Hr
First Derivative
Spectrum
H
Fig. 9.1 (a) Splitting of electron (system S = 1/2, I = 0) energy levels by variable applied magnetic field and constant frequency
hn = constant) (b) Absorption and first derivative ESR spectra of a free electron.
Problem 9.1: (a) Calculate the relation between field strength and frequency for ge = 2.0023. (b) Compute the
frequencies for ESR and PMR transitions if the strength of the applied magnetic field is 10 kG. (c) The magnetic
moment of an electron is about −9270 × 1024 compared to +14.1 × 1024 erg G−1 for proton. Determine the ratio of
Lande splitting factor of proton to electron.
Solution (a) From Eq. (9.7a),
v g e βe 2.0023 × 9.274 × 10 −21 ( −11

)
= = = 2.80 × 106 Hz(G−1) = 2.80 MHz (G−1)
H h 6.625 × 10 −27 (erg s)
Thus, 1 G = 2.80 MHz
(b) The resonance frequency (ve) for an electron is given by
gβH
ve = (9.7b)
h
Substituting the values of respective parameters into Eq. (9.7b),
2.0023 × 9.274 × 10 −21 ( −11
) ×10 4 ( )
ve = = 2.80 × 1010 Hz = 28000 MHz
6.625 × 10 −27 (erg s)
Recall Eq. (7.16) from Chapter 7, i.e.

g p βp H
vp =
h
5.585 × 5.05 × 10 −24 ( −11
) × 10 4 ( )
= = 4.6 × 107 Hz = 46 MHz
6.625 × 10 −27 (erg s)
(c) We know that
mp = gp bp MI (erg G−1) (Eq. 7.8)
me = − ge be MS (erg G−1) (Eq. 9.1)
1
For electron and proton, MS = MI = ± . Since we are to compare the magnitudes of g, for proton and
2
electron, so the minus sign in Eq. (9.1) may be ignored.
Comparing Eqs (7.8) and (9.1), we obtain
g p μ p βe e g G −1 ) ×
14.1 × 10 24 (erg × −21 (erg G −1 )
= × = = 2.79 × 103
g e μe β p 9.270 × 10 24 (erg G −1 ) 5.05 10 24
24
( 1
)
9.4 INTENSITY OF ESR LINES AND FACTORS AFFECTING IT

The properties of ESR absorption lines for Lorentzian and Gaussian line shapes (Figs 9.2 and 9.3) are listed in
Table 9.1. Like NMR, the intensity of the ESR absorption line for these two types of line shapes is given by
I0
I= Lorentzian shape (9.8)
T ( H − H r )2 + 1
2
2
I I 0 exp [ −b ( H − H r ) 2T 22 ] Gaussian shape (9.9)

where terms and symbols have their usual meanings, i.e. I0 is the intensity of the absorption line at its centre, Hr
is the field of resonance at line centre, T2 is the spin-spin relaxation time, b is a constant.
The intensity of ESR line depends on the population of the unpaired electrons in the ground (N−1/2) and excited
(N+1/2) spin states. The ratio of two populations in thermal equilibrium is governed by the Boltzmann law.
N
−
1
⎡⎛ ⎞ ⎤ ⎛ ΔE ⎞ ⎛ g e βe H ⎞
2
= exp ⎢ E 1 E 1 ⎟ / κT ⎥ = exp ⎜ ⎟⎠ = exp ⎜⎝ (9.10a)
N 1 ⎣⎝ 2
+ − ⎠
2 ⎦ ⎝ κT κT ⎟⎠
+
2
(a)
Normalised
absorption curve
1 (a)
l lo
2 o Normalised
absorption
curve 1
l lo
First-derivative
2 o
curve (b)
lpp First-
derivative (b)
curve
lpp
Second-
Second-derivative derivative
curve A curve (c)
(c)
ΔH 1 A
2
ΔHpp B
B
ΔHpp
ΔH 1
2
Hr H
H
Hr Fig. 9.3 Gaussian line shape: (a) absorption spectrum,
Fig. 9.2 Lorentzian line shape: (a) absorption spectrum (b) first-derivative spectrum, and (c) second-derivative
(b) first-derivative spectrum (c) second-derivative spectrum. spectrum.
Table 9.1 Properties of Lorentzian and Gaussian lines.
Property Lorentzian Shape Gaussian Shape
Half-width at half-height Δ 1/ 2
ΔH Δ 1/ 2
ΔH
1
2 ⎛ 2 ⎞2
Peak-to-peak width Δ = ΔH 1 / 2
Δ Δ =⎜ ΔH 1 / 2
Δ
⎝ ln 2 ⎟⎠
PP
3 PP
1
1 ⎛ ln 2 ⎞ 2 1
Peak amplitude I0 = I0 = ⎜
πΔ
ΔH
H1/ 2 ⎝ π ⎟⎠ ΔH H1/ 2
I 0 ( ΔH )2 ⎡ (− ) (H Hr ) ⎤
2
H
Equation for absorption line I= I I 0 exp ⎢ ⎥
( ΔH
H ) (H − H r )
2 2
⎢⎣ ( ΔH )2
⎥⎦
1
3 3 1 ⎛ 2 ⎞2
Peak-to-peak amplitude I PP = I PP ΔH
H1/ 2
4π ( ΔH
H ) 2
⎝ ln 2 ⎠
I′ =
( −I ) 2 ( ) (H − H r ) × exp
( −I ) 2 ( ΔH
H ) (H H r )
2
( ΔH H )
2
Equation for first derivative absorption line I′ = 2

⎡( ΔH ) ( H − H )2 ⎤ ⎡ (− ) (H H r ) ⎤⎥
2 2
⎣ r
⎦ ⎢
⎢⎣ ( ΔHH )
2
⎥⎦
⎡ 1 ⎤ ⎡ 4e −33 / 2 l 2 ⎤
Peak amplitude of positive lobe A I0 ⎢ ⎥ A I0 ⎢ ⎥
⎢⎣ 2 ( ΔH ) ⎥⎦ ⎢⎣ ( ΔH
H ) ⎥⎦
2 2
H
⎡ 2 ⎤ ⎡ 2l 2 ⎤
Peak amplitude of negative lobe B I0 ⎢ ⎥ B I0 ⎢ ⎥
⎢⎣ ( ΔH ) ⎢⎣ ( ΔH
H ) ⎥⎦
2 2
H ⎥⎦
Hr is the field resonance at the centre.
Accordingly, the intensity of ESR signal increases as the ratio between the two populations, i.e. N−1/2/N+1/2
decreases. The intensity also increases when the temperature decreases,
The population difference (Δn) between the spin states is given by
Ng e βe H
Δn = N 1 −N 1 = (9.10b)
−
2
+
2 2κT
where, N−1/2 + N+1/2 is the total number of unpaired electrons in both the spin states.
The signal intensity is directly proportional to Δn which is inversely proportional to temperature. Thus as the
temperature decreases, the signal intensity increases. For this reason, the ESR experiments are performed at low
temperatures. The intensity of ESR signal at liquid nitrogen temperature (77 K) is about four times higher than at
room temperature.
The probability of the transition between two spin states is proportional to the incident microwave energy. As
the microwave energy increases, the signal intensity also increases and becomes maximum. A further increase of
microwave energy broadens and decreases the ESR signal till it completely disappears. This is the result of satura-
tion process, where the population of both the spin states becomes equal. The return of the system to equilibrium
is called relaxation process. The two types of relaxation processes, i.e. s-l and s-s are exponential in nature and
are characterised by their relaxation times, T1 and T2 respectively.
9.5 ESR LINE WIDTH AND FACTORS AFFECTING IT

The line width of ESR lines is expressed in two ways: (i) Half line width ( ΔH 1/ 2 ) , i.e. the width at half the height
of the absorption line (Figs 9.2 and 9.3 and Table 9.1). (ii) Peak-to-peak width (ΔHpp), i.e. the full width between
the two extremes of the first derivative curve (Figs 9.2 and 9.3 and Table 9.1).
According to Heisenberg’s uncertainty principle, the line width of ESR line is given by
h h
g β ΔH 1/ 2 Δt ≥ or ΔE Δt ≥ (9.11)
2π 2π
where ΔH1/2 is the line width (G), Δt is the lifetime of the excited spin state (s), ΔE is the energy difference
between the two spin (electron) states. According to Eq. (9.11), ΔH1/2, depends on the lifetime of the excited spin
state, i.e. if Δt is small, ΔH1/2 will be large. Thus the parameters which decrease the lifetime of the excited spin
state cause broadening of the ESR resonance.
As already discussed in NMR (Chapter 7), the line width depends on either or both the spin-lattice (T1) and
spin-spin, i.e. dipolar (T2) relaxation processes via which the spin in the excited state dissipates its energy to return
to the ground state without emission of any energy and the exchange process.
Line broadening due to s-l process takes place due to the interaction of the paramagnetic ion with the thermal
vibrations of the crystal lattice or with the rotation and vibration motions of solvent molecules in liquids. The s-l
relaxation time depends on the intensity of s-l interactions. Thus the factors which decrease the s-l interactions,
i.e. T1 will increase the line width and vice versa. The s-l process is temperature dependent and is efficient at
room temperature. Its efficiency decreases as the temperature is lowered and becomes several minutes at liquid
nitrogen temperature. Thus at low temperatures, s-l process (T1 = ∞) does not contribute to line broadening. So
in order to obtain a good ESR spectrum, the sample should be cooled to liquid N2, or H2 or He temperature.
T1 is also sensitive to the presence of neighbouring electronic excited states. The relaxation time is small when the
electronic excited state is close (∼100 cm−1) to the magnetic nuclei. T1 is also dependent on the symmetry of radi-
cals. The reduction in symmetry of the radical leads to weakening of the spin-orbit coupling and to an increase
in the s-l time and consequently narrowing of lines. T1 for liquids (10−4−10 s) is less than that for solids (10−2−104 s)
due to random motion of molecules in liquids. As a consequence, liquids exhibit broad line spectra whereas
solids exhibit narrow line spectra. On the other hand, the s-s process (T2) is very efficient and has relaxation times
T2~ 10−4 s for solids and T2 ∼ T1 for liquids. In the s-s interaction, the excited state unpaired electron shares its
energy with the neighbouring electron spins rather than with the lattice as in the s-l process. Thus the larger the
sharing rate, the smaller will be the relaxation time and broader will be the ESR line. Note that the local magnetic
field (dH) produced by a paramagnetic centre, i.e. say paramagnetic ion, is given by
3βe
δH = (9.12)
r3
where r is the distance between the paramagnetic centres. The local field of the magnetic dipole influences the
neighbouring electron spins and alters the field there. This magnetic dipolar interaction reduces the lifetime of
the excited state electron spin and broad ESR line results. This is also called concentration broadening since
1
δ H ∝ 3 . The effect of s-s interaction or concentration broadening can be reduced by increasing the distance
r
between the paramagnetic ions by magnetically diluting the sample. In solids, this is achieved by diluting the
paramagnetic ion in a diamagnetic isomorphous material, e.g. small amounts of CuSO4 in a ZnSO4 crystalline lat-
tice. The dilution factor should be in excess of 100. The interaction of the magnetic nuclear spins in a crystalline
compound can also broaden the ESR line and this effect is about 103 times smaller than the electron s-s interaction
effect and is negligible in magnetically dilute samples.
In conclusion, the ESR linewidth is related to T1 and T2 as
1 1 1
Δ 1/ 2 = = + (9.13)
τ T1 T 2
At low temperature, the s-l relaxation time T1 = ∞. Therefore, by Eq. (9.13) we obtain
1
Δ 1/ 2 = (9.14)
T2
i.e. at low temperature the ESR line width is governed only by s-s interaction.
The ESR lines are wider than the NMR lines, i.e. bandwidth of ESR lines > bandwidth of NMR lines, e.g. the width
of ESR absorption line for a typical relaxation time say 10−7 s is (2pΔt)−1 ≈ 1 MHz whereas the width of NMR line for
a relaxation time of 10−1 s is 1 Hz (normal line width for a liquid sample). The advantage of wider ESR line is that the
magnetic field homogeneity of 1 in 105 would suffice whereas this figure in NMR is 1 in 108 over the sample. On the
disadvantage side, however, it is much more difficult to detect a wide line than a narrow line of the same intensity. Due
to this reason, ESR spectrum is recorded in the derivative mode. We have already stated that ESR line width is also
affected by exchange process between neighbouring unpaired electrons. In undiluted crystals, the paramagnetic ions
are so close to one another that overlapping (a)

of orbitals containing the unpaired electrons
takes place. When exchange occurs between
equivalent ions, the exchange energy will be
large as compared to the value of hyperfine
splitting, the hyperfine lines broaden at the
base and become narrow at the centre. This is
known as exchange broadening. On the other (b)
hand, when exchange takes place between non-

equivalent ions, the exchange energy is almost
equal to the value of hyperfine splitting, the
resonances of separate hyperfine lines merge
to produce a single broad line. This is called
exchange narrowing phenomenon. Such an (c)
effect has been observed in the electron transfer
reaction between tetracyanoethylene (TCNE
or N) anion radical and its neutral molecule
i
( ) . As TCNE is added to
the solution containing TCNE anion radical,
the sharp nine lines in the spectrum (due to four
equivalent nitrogen atoms) will begin to broaden (d)
and coalesce into a broad spectrum and finally
at very high concentration of TCNE, only one
sharp line is obtained (Fig. 9.4). CuSO4· 5H2O
also exhibits such an effect since it has two
distinct sites per unit cell. Since the exchange
interaction depends on the distance between
the paramagnetic centres, it can be minimised Fig. 9.4 ESR spectra of tetracyanoethylene anion radical ( TCNE or N )
i i
by magnetic dilution. Note that in dilute solu- as a function of the transfer rate: no neutral TCNE is present (a) and the
tions, the paramagnetic centres may come concentration of TCNE grows from (b) to (d).
into contact with one another due to random
molecular motion in a liquid. If the temperature is raised, the increased rate of molecular collisions results into
greater exchange interaction and broadening of hyperfine lines.
Hlocal
H0
(a)
(b)
Fig. 9.5 Hole burning in inhomogenously broadened line: (a) initial line,
and (b) localised saturation developes and appears as hole.
The residual anisotropy in a viscous medium produces line broadening. This is termed as motional narrowing
since free motion of the paramagnetic molecules if they are large is restricted. This effect can be reduced by rais-
ing the temperature.
Incident microwave energy also influences the line width. When the microwave power is increased, the lifetime
of spin state increases and broadening of ESR line occurs up to the moment when the line completely disappears.
This process is known as microwave power broadening or saturation broadening of a homogeneous line. The
width of homogeneous broadened line is determined by Eq. (9.11).
Note that broadening of ESR lines may occur homogeneously or inhomogeneously. Homogeneous broadening
results when all the spins have identical environment whereas inhomogeneous broadening arises when spins are in
different environment [Fig. 9.5 (a) and (b)]. The inhomogeneous broadening arises due to the selective saturation
of some particular population of spins. The intensity of line corresponding to such spins becomes zero and has
indeed been observed. The resonance line shows typical holes and the process is called ‘burning a hole’. Most of
the ESR spectra are inhomogeneously broadened, which may be due to irregularities in the external magnetic field
and also due to heterogeneities in the samples. Anisotropies in the dipolar interaction in most of the powdered
samples alter the inhomogeneous broadening. Inhomogeneous line broadening is also shown by viscous liquids.
9.6 g-VALUE AND FACTORS AFFECTING ESR LINES

Various ESR spectrometers use different frequencies and various types of samples require different frequencies
for the experiment. For that reason the position of ESR lines cannot be indicated by the frequencies. The line posi-
tions are instead denoted in terms of ‘g-value’ which is expressed as a function of microwave frequency (v) and
external magnetic field (H) at resonance. From Eq. (9.7), we obtain
hv
ge = (9.15)
βe H
As mentioned earlier, the ge value of a free electron is 2.0023 due to a relativistic correction, whereas the
g-values for free radicals in atoms, molecules and crystals depend on their electronic structure. This quantity
specifies the strength of the interaction of a given electron with the applied magnetic field and is the spectroscopic
gauge of the magnitude of the magnetic moment of the electron under study.
μ βe S (S + ) (9.16)
The quantum mechanical expression for the Lande splitting factor is

J ( J + 1) + S ( S + 1) − L( L + 1)
g = 1+ (9.17)
2JJ ( J + 1)
According to Russell–Saunders coupling, the total angular momentum of atom (characterised by quantum
number J) is the sum of orbital angular momentum (characterised by quantum number L) and spin angular
momentum (characterised by quantum number S), i.e.
J=L+S
J = L + S, L + S − 1, L + S − 2, L + S − 3,..., L S
For L S J S +1
L S J 2L + 1
Accordingly, a set of terms of an atom with one and two electrons is given in Tables 9.2 to 9.4.
Table 9.2 Singlet terms of an atom with two electrons (S = 0, J = L).
Term L J
1
S 0 0
1
P 1 1
1
D 2 2
1
F 3 3
Table 9.3 Triplet terms of an atom with two electrons (S = 1, J = L + 1, L, L − 1).
Term L J
S
3
0 1
P
3
1 0 1 2
D
3
2 1 2 3
F
3
3 2 3 4
Table 9.4 Doublet terms of an atom with one outer electron (S = 1/2, J = L + S) and the Lande splitting factor.
J 1 1 3 3 5 5 7
Term , , ,
L 2 2 2 2 2 2 2
S
2
0 2
2 4
P
2
1 g ,
3 3
4 6
D
2
2 ,
5 5
6 8
F
2
3 ,
7 7
1
For a free electron L = 0, J = S = . Therefore, g = 2.0 (by Eq. 9.17).
2
All free radicals and some ionic crystals have g = 2.0 ± 0.003. In free radicals, the electron is not localised in
a particular orbital but it can move about freely over the orbitals encompassing the whole molecule so the spin
orbit coupling constant of the parent atoms is small and the orbital degeneracy, i.e. electronic excited states, is
removed or quenched. Consequently, the other contributions to magnetic moment other than those from elec-
tron spin are very small (<1 : 104); the electron behaves as an electron in fiee space having L = 0, i.e. g = 2.0
[by Eq. (9.17)].
In transition metal systems, the electron is localised in a particular orbital about the atom, so the parent nucleus
has large spin-orbital coupling constant and accessible orbital degeneracy. Consequently, the magnetic moment
contains additional contribution and observed g-value differs significantly from 2.0. It is found to vary in the
range 0.2−8 in some ionic crystals. In such cases, L couples with S to give g-values according to Eq. (9.17). The
g-value of 2.0 in some ionic crystals may come in two ways: (i) When the electron in an ion is in S state, (L = 0);
g = 2, e.g. ground state of Mn (II) with five d unpaired electrons (i.e. S = 5/2, 2S + 1 = 6) has zero orbital momen-
tum. Thus L = 0, J = S and the term symbol is 6S5/2. By Eq. (9.17), g = 2. (ii) The internal crystal field (electric
field due to ions in a crystal) may break down the L-S coupling and on application of external magnetic field, the
electron spin precesses independently about the applied field direction. In such a case the value of L is immate-
rial and g becomes ~2. On the other hand, if internal crystal field is week or if the paramagnetic electron is well
shielded from the field (rare earth metals), then L couples with S to give resultant J which itself precesses about
the direction of the applied field and g is given by Eq. (9.17). Further when L and S are only partly uncoupled, the
residual orbital contribution to the energy results in a g-value not easily predictable theoretically.
In transition metal systems, the additional contribution to g-value (2.0023) by orbital degeneracy and spin-
orbital coupling is termed as orbital paramagnetism. This shifting of g-value is because of mixing via spin orbital
coupling of metal orbitals involved in molecular orbitals containing the unpaired electron (s) with the empty or
filled ligand orbitals. When mixing takes place with empty ligand orbitals, the result is negative shift (g − Δg)
whereas positive shift (g + Δg) results when mixing is with filled ligand orbitals. The shift depends upon the
amount of unpaired electron density at the donor sites of the ligands, i.e. on the degree of covalency of the
complex. The degree of covalency increases as the value of Δg decreases.
Furthermore, the orbital paramagnetism reflects the ability of an unpaired electron to migrate from its parent
orbital into other orbitals related to the original orbital by rotations about one or another coordinate axis. The
rotational relationships of the d-orbitals are listed in Table 9.5, which shows that the rotational tendencies may
be different about the X, Y, Z axes, e.g. rotation about Z-axis has no effect on d z 2 while rotation about X or Y axis
takes d z 2 into dyz or dxz with a strength of 3.
Table 9.5 First-order perturbation solutions matrix elements for the g-tensor (S = 1/2) rotational properties of the real d-orbitals.
Term Lx Ly Lz
eg −i 3 d yz i 3 d xz 0
z2
dx 2 −i d yz 2id xy
y2 −idxz
t 2g −2id x 2 − y 2
dxy idxz −idyz
−id x 2 − y 2 −i 3 d z 2 + id x 2 − y 2 id yz
dxz
id x 2 + i 3 dz 2
dyz y2 idxy −idxz
ni λ
; i x , y , z ; ni = 2. < matrix element > , ni is an integer and may have positive or negative value. + refers to the mixing
2
gi = ge +
Δi
of the electron with an empty orbital and − refers to the mixing of the electron with a filled orbital. The value of ni may also be obtained from
the magic pentagon (Fig. 9.6). For electronic transition between d x 2 y 2 and dxy; n = 8. For electronic transition between d x 2 y 2 and dyz; n = 2;
10 Dq = Δ = E e g E t 2 g , λ-spin-orbital coupling constant of free ion, e.g. for an electron in d x 2 y 2 the matrix element for rotation into dxy is:
2 2 2 8λ
< xy L z x − y > = 2i ; n 2.( 2i ) = 8; g z = 2 − . For square planar Cu(II), λ = − 830 cm−1, Δ 2 = 16000 cm ; g =
−1
Δ ( xy → x 2 − y 2 )
2
( → − ) z
2 + 0.42 = 2.42. 1 cm−1 = 10 kG.
The total magnetic moment has contribution from both the spin and orbital dz 2
magnetic moments and will also thus vary with the direction of the molecule 6
6
in the applied magnetic field. Consequently, g-factor being the spectroscopic
manifestation of magnetic moment will also be orientation dependent, i.e. it 2 dyz ± 1
dxz
is anisotropic.
In single crystals, the g-value depends on the direction of the applied field
2 2
in relation to the crystallographic or molecular axes X, Y and Z. Z-direction 2 2
is defined coincident with the highest fold rotation axis which can be deter-
mined by x-ray technique. In polycrystalline samples (powdered samples)
a set of g1, g2 and g3 [Fig. 9.7(a)] are used to define the anisotropy of the dx -y dxy ± 2 2 2
8
g-value, i.e. Fig. 9.6 Magic pentagon: MI stands for
values of different orbitals.
2
= g12l x2 g 22 l y2 g 32l z2 (9.18)
where l x2 2 2
i 2 θ sin 2 ϕ and l z2 = cos 2 ϕ
l y2 = sin
are the normalised direction cosines of the g-tensor. The
anisotropy of g-value causes an asymmetry of ESR absorp- g1 g||
H H
tion line one for each of the three mutually perpendicular
orientations of the paramagnetic centre with respect to the θ
θ
magnetic field. For cubic crystals, g1 = g2 = g3 or gmin = gmid
= gmax or gx = gy = gz, where the subscripts min., mid., max.
refer to the smallest, intermediate and largest values of g. g⊥
The ESR spectrum is being described as cubic or isotropic g3

[Figs 9.2/9.3(a) and (b)]. For crystals having axial symme-
g2
try, i.e. tetragonal crystals, g1 = g2 < g3, or g1 = g2 > g3; the
spectrum is described as tetragonal. The identical pair, i.e. (a) (b)
g1 = g2 is referred as g⊥ whereas the unique g-value is called Fig. 9.7 (a) Relations between the directions associated
g [Figs 9.7(b), 9.8, 9.9(a) and (b)]. For rhombic crystals, g1 with g , g , g and H, and (b) relations between the directions
≠ g2 ≠ g3, the spectrum is described as rhombic [Figs 9.9(c) associated with g , g and H.
1 2 3
⊥
and (d)].
Experimentally observed, ESR spectrum of some molecule or ion in solution phase is also shown in Figs 9.9(e)
and (f).
g|| g⊥
A|| A⊥
Fig. 9.8 ESR spectrum of Ag (II) in nitric acid at 77K.
(a) (c)
(e)
g|| g⊥ H g3 g2 g1 H
H
(b) (d) (f)
H H
g1
g|| g⊥ g3 g2 H
Fig. 9.9 (a) and (b) Axial symmetry g ≠ g ( g > g⊥ ) : (a) absorption spectra; (b) first derivative spectrum. For g < g⊥ : ESR spectra are
obtained by inverting curves (a) and (b). (c) and (d): An asymmetry g-tensor g1 < g2 < g3, (c) absorption spectrum, (d) first derivative
spectrum, (e) experimentally observed absorption spectrum and (f) first derivative spectrum of some free radical in the solution phase.
In bulk susceptibility measurements, a powdered sample or frozen solution is used and hence g works out as
an average (gav). The gav value for a tetragonal complex is given by
1 2
g2 ( g x + g y2 + g z2 )
3
=
1
3
( + ) (9.19)
or approximately as,
g
1
3
(
g + g ) (9.20)
1
For S = systems, anisotropy in g can be considered in terms of two limiting cases:
2
(i) Δ >> λ; e.g. Cu2+, Mo5+ and low spin Co2+.
ni λ
Then, g = ge + i x,y z (From Table 9.5) (9.21)
Δi
λ has a positive value for d1−d4. The use of this equation has been elaborated in Table 9.5. This relation
holds good provided λ/Δ ≤ 0.1. Covalency effects reduce the magnitude of λ and this is accommodated in
Eq. (9.21) by multiplyin g the last term by α2, the fraction of unpaired electron confined to the metal. Thus
Eq. (9.21) becomes
ni λα 2
g = ge + i x, y z (9.22)
Δi
(ii) Δ ≈ λ, e.g. low-spin Fe(III) as found in haeme proteins. In low-spin Fe(III), the single unpaired electron is
found in the configuration ( g )5 , specifically (dxy)2, (dxz)2, (dyz)1. These orbitals are nondegenerate and the
differences in energy between them is described in terms of two parameters V and Δ. Δ is a measure of the
strength of the Z-directed axial ligand field perturbation which stabilises dxy with respect to the centre of
gravity of dxy and dyz, whereas V is a measure of the strength of the rhombic (X, Y) perturbation and is the
energy separation between dxy and dyz. V and Δ can be obtained directly from g-values.
V gx gy
= + (9.23)
λ gz + g y g gx
Δ gx gy
= − (9.24)
λ 2( + ) 2 (g − g x )
Δ is an index of the crystal field intensity at the iron centre and ratio V/Δ is a measure of the magnitude of
the distortion imposed upon the iron atom by the six pseudo-octahedrally disposed ligands.
9.7 HYPERFINE INTERACTION

When the unpaired electron in the molecule is delocalised over the whole or at least a large part of it, it may inter-
act with the many nuclei possessing magnetic moment in the molecule to cause a further splitting of the electron
resonance line. This type of magnetic interaction between the electron and the nuclear spins in the same molecule
is known as hyperfine coupling and produces ESR spectra with hyperfine splitting, i.e. spectra with a number of
lines. From the number of lines, their relative intensities and their separation, one can determine the type and
number of nuclear spins which interact with the electron, i.e. the structure of free radical. The line separation
depends on the interaction between the electron spin and each nuclear spin, and on the magnetic moment of the
involved nuclei.
The nuclei spin (I) in an external magnetic field have the restricted orientations: I, I − 1, I − 2, I − 3,.....0.....
–(I−1), − I. In this way, the number of nuclear spin quantum numbers MI = 2I + 1, i.e. they split an electron
magnetic level into 2I + 1 sublevels.
The energies of a coupled level are given by
E g e βe H M S + E loc (9.25)
The second term on the right-hand side of Eq. (9.25), i.e. Eloc, appears due to local magnetic fields at the elec-
tron produced by the nuclei.
1
Single nucleus of I = , e.g. 1H, 19F, 13C and 31P and single deuterium or nitrogen nucleus (2H or 14N; I = 1) exhibit
2
hyperfine splitting. Hyperfine splitting in poly-atomic radicals by 6Li, 7Li, 10B, 11B, 23Na and 39K have also been reported
as have been splittings by nuclei of transition elements, rare earths, and trans uranic elements in compounds.
9.8 TYPES OF HYPERFINE INTERACTIONS

There are two types of hyperfine interactions that give rise to hyperfine splittings: (i) Contact or Fermi hyperfine
interaction, and (ii) anisotropic hyperfine interaction.
9.8.1 Contact or Fermi Hyperfine Interaction

This is a quantum mechanical interaction that depends on the electron spin density in the immediate vicinity of
the nucleus. This interaction is isotropic (orientation independent) and, therefore, also known as isotropic hyper-
fine interaction. We know that the electron spin density is nonzero for electrons in s-orbitals, and consequently
the Fermi interaction can take place. On the other hand, the wave function for electrons in p, d, f and higher
orbitals have a node at the nucleus so that the electron density vanishes at that point and no contact interaction
takes place.
(a) ESR Spectrum due to Single Proton The contact interaction energy (Eloc) for hydrogen atom is given by
E loc hA MS MI (9.26)
where A (in MHz) is a frequency factor (also called the isotropic hyperfine coupling constant and is written as A ,
when the interaction is parallel to the direction of the applied field) that can be computed from the square of the wave
function evaluated at the nucleus. Thus, Eq. (9.25) can be written as
E g e βe H M S + hA M S M I (9.27)
In a strong applied magnetic field, the first-order energies, i.e. ge be H MS, are 103 fold or more than the contact
interaction energies.
1 1
For a system of one paramagnetic electron and a nucleus with S = and I = , there are two possible orienta-
2 2
⎛ 1⎞ 1
tions ⎜ ± ⎟ for the magnetic moment of nucleus in an external magnetic field, i.e. M I = ± . The energy of these
⎝ 2⎠ 2
nuclear spin states affects the energy levels of an electron (E+1/2 and E−1/2) and split them into four sublevels (E1,
E2, E3 and E4). This process of splitting electron energy levels into sublevels by nuclear magnetic levels is known
as the Zeeman hyperfine splitting for nucleus, i.e. proton in the present case [Fig. 9.10(a)].
The energies of the four sublevels can be written in order of decreasing energy as
1 hA 1 1
E g e βe H + for M S d MI = +
2 4 2 2
1 hA 1 1
E g e βe H − for M S d MI = − (9.28)
2 4 2 2
1 hA 1 1
E g e βe H + for M S d MI = −
2 4 2 2
1 hA 1 1
E g e βe H − for M S d MI = +
2 4 2 2
Transition between energy levels E1 and E3 and levels E2 and E4 are not permissible since they require a simultaneous
absorption and emission of energy by both the electron and the nucleus. As stated earlier, when the electron absorbs
microwave energy and flips from one spin state to the other, it does so with no change in MI. However, MI will change
if irradiation is done with oscillating radio frequency field of energy (hv). Thus, the selection rules are
(i) ΔMS = ±1 and ΔMI = 0 for allowed transitions
(ii) ΔMS = ±1 and ΔMI = ±1 for forbidden transitions (9.29)

Applying the selection rules from Eq. (9.29) to the Eq. (9.28), we obtain the energy change for the two transi-
tions as,
hA 1
Δ 1− 4 = E1 − E = g e βe H + fo M I = + (9.30)
2 2
hA 1
Δ 2−3 = E 2 − E = g e βe H − for M I = − (9.31)
2 2
At constant external magnetic field, the transition from level E4 to level E1 occurs at higher frequency (v2) than
the transition from level E3 to level E2 which occurs at (v1). The difference between the two frequencies is called
the hyperfine frequency (Δv) and is equal to A(MHz). In other words, we can say that in a constant magnetic
field, the transition in energy differs by the isotropic hyperfine coupling constant (A).
In an ESR experiment, the frequency v usually is fixed and the applied field is varied until resonance is achieved
where ΔE = hv and H = Hr. Substituting these relationships in Eqs (9.30) and (9.31), we obtain
hA
ΔE1− 4 = hv = g e βe H 1− 4 + (9.32)
2
hA
ΔE 2 −3 = hv = g e βe H 2 −3 − (9.33)
2
Rearranging Eqs (9.32) and (9.33), we get the resonant fields at which the electron resonance transitions
occur:
hv a 1
H1 4 = − for M I = + (9.34)
g e βe 2 2
hv a 1
H2 3 = + for M I = − (9.35)
g e βe 2 2
At constant frequency [Fig. 9.10(a)], the transition from level E1 to E4 occurs at a lower magnetic field than the
transition from level E2 to E3. The two normal absorption and their corresponding first derivative signals for the
two transitions are shown in Fig. [9.10(b)]. The separation between the two signals is
hA
ΔH = H 2 −3 − H 1− 4 = =a (9.36)
g e βe
hA
The value of ΔH = a = is called the hyperfine splitting constant (HSC) and expresses in guass (G) the sepa-
g e βe
ration between two hyperfine lines in the spectrum.
(b) Quantum Mechanical Considerations of ESR Spectra of Hydrogen Atom The general Hamiltonian
used in the analysis of ESR spectra due to radicals is given by
H = H Zeeman + H d d Hc (9.37)
where, HZeeman refers to the interaction of electron and the nucleus with the applied magnetic field, H.
H g e βe SH − g β IH
I (9.38)
where the symbols have their conventional meanings; S and I are respectively, the electronic and nuclear spin
vectors, ge and gN are the Lande factors for the electron and nucleus, be and bN are the electron Bohr magneton
and nuclear magneton.
Hd−d refers to the classical d−d interaction between electron and nuclear magnetic moments.
⎡ S I 3(S r )(I r ) ⎤
H g e βe g N β N ⎢ − (9.39)
⎣ ( re rn )
3
( re rn )5 ⎥⎦
(a)
E+1 MI = + 1 (1), E1
2
Ms = + 1 2
2 hA
2
Ms = + 1
2 MI = −1/2(1), E2
ΔE = 0 ΔE = hν = geβeH
MI = −1/2(1), E3
Ms = − 1
2
H=0 hA
Ms = − 1 2
Energy
2 E−1
2 MI = + 1 (1), E4
2
(b)
Absorption spectrum
H
H1−4 Hr H2−3
Low field High field
First-derivative
spectrum
H
ΔH = a +a
−a
2 2
Fig. 9.10 (a) Energy levels of the hydrogen atom (system: S = 1/2, I = 1/2) in a variable applied magnetic field and constant frequency
(n = constant), and (b) absorption and first derivative spectra of hydrogen atom.
where, re and rn, refer to the vector position of electron and nucleus respectively. For dilute solutions of free
radicals, the electron-electron interaction, i.e. interaction between unpaired electrons on different molecules can
be neglected.
Hc refers to the Fermi contact interaction and does not vanish in solution.
8π
H g e βe g β S I δ ( re − rn ) (9.40)
3
Here, d(re − rn) is known as the Dirac delta function. d(re − rn) ≠ 0, provided the wave function of the unpaired
electron in the radical has a non-vanishing value at the nucleus.
The total energy due to various magnetic interactions is given by
E ∫ ψH ψ d τ (9.41)
where ψ contains both the space and the spin part of the wave function of the electron.
If the spin part of the wave function is denoted by ψa or ψb then
1
∫ψ ( β ⋅ ψα τ =
2
gβ H
(9.42)
1
∫ψ β ( β ⋅ ψβ τ=− g β H
2
In general, we write
∫M g e βe S H ) M S d τ = g β H M S
(g (9.43)
In solution, Hd−d = 0. The nuclear Zeeman interaction, being very small as compared to its electron counterpart
is also neglected. In a strong applied field, there is no interaction between the orbital and spin angular momenta
but both the momenta interact independently with the applied field. This is known as Paschen–Back effect. This
effect manifests itself in the ESR and NMR spectra. In the Paschen–Back region the strength of applied field is
greater than the multiplet splittings and only the normal Zeeman effect is observed even when doublets studies
doubts or higher spin states are present. In principle, the ESR alike NMR studies can be performed at any mag-
netic field but the usual studies are always preferred in the Paschen–Back region. Equation (9.41) in the Paschen–
Back region becomes
8π
E ∫ ψ g e βe S H + g β g N β N S I δ ( re rn )] ψdτ
3
= g β H MS + aM IMS (9.44)
Here, a is the hyperfine splitting constant and is given by

8π 2
a g e βe g N βN ψ(0) (9.45)
3
|ψ(0)|2 is the probability of finding the unpaired electron in contact with the nucleus giving rise to isotropic coupling.
This term vanishes except when the atomic orbital containing the unpaired electron has some s-character. Hartree
Fock calculations have made available the theoretical values of |ψ(0)|2 for elements of the Periodic Table. Applying
Eq. (9.44) to hydrogen atom, we get
1 a 1 1
E g e βe H + for MS d MI = +
2 4 2 2
1 a 1 1
E g e βe H − for MS d MI = − (9.46)
2 4 2 2
1 a 1 1
E g e βe H + for MS d MI = −
2 4 2 2
1 a 1 1
E g e βe H − for MS d MI = +
2 4 2 2
By selection rules ΔMS = ±1 and ΔMI = 0, the two permissible transitions are
a
Δ 1− 4 = E1 − E = g e βe H + (9.47)
2
a
Δ 2−3 = E 2 − E = g e βe H − (9.48)
2
The two resonance lines are presented in Fig. 9.11.
For a system in which more than one type of proton interacts with the unpaired electron, Eq. (9.44) may be
written as
n n a
E g e βe HMS + a MS ∑M
i =1
I i a MS ∑M
i =1
I (i ) + ... (9.49)
where n is the number of protons of a particular type, a1, a2 ... are gebeH
the hyperfine constants for protons of type 1, 2, ...a is expressed
either in MHz or in gauss. The conversion from one unit to the Fig. 9.11 First derivative ESR spectrum of hydrogen atom.
other is given by:
⎛g ⎞
a (G ) = 0.35683 ⎜ free electron ⎟ a( MHz ) (9.50)
⎝ g radical ⎠
The hydrogen atom produced by electric discharge or trapped in solids exhibits two-line ESR spectrum. The
separation between the two lines has been found to be 506 G (calc. 539.866). This is incidently the largest splitting
for a proton since the unpaired electron is in the 1s orbital and so has 100 per cent s character. When two unpaired
electrons are present in a molecule, i.e. triplet molecule, then we write
H g e βe S S H g N βN IH (9.51)
8π
H g e βe g β (S + S ) I ⋅ δ ( re rn ) (9.52)
3
⎡S S 3 S 1 r )(S 2 r ) ⎤
3(
and H g e2 βe2 ⎢ 1 3 2 − ⎥ (9.53)
⎣ r r5 ⎦
Here, S1 and S2 (the spin angular momentum vectors) refer to electrons 1 and 2 and r is the vector joining them.
Problem 9.2: Express the hyperfine coupling frequency (Δν) in terms of hyperfine splitting constant (a) for
g = 2.0023.
hA
Solution We know that Δv = A and a = .
g e βe
g β a
Thus, Δv =
h
For g = 2.0023
2. 23 × 9.274 × 1 −21
(erg G −11 )a (G )
Δv = −27
6.625 × 1 (erg s)
= 2.80 × 106 a s−1 = 2.80 a Mc s−1

= 2.80 a MHz
.
(c) ESR Spectrum of C H (System with S = 1/2, I = 1/2) When the magnetic moment of a paramagnetic
electron is perturbed by the magnetic field of a nucleus with I = 1 (say proton), the transitions occur at two reso-
nance fields [Fig. 9.10(a)]: 2
hv a 1
H1 4 = − for MI = + (9.54)
g e βe 2 2
(a)
MI = +1(1),E1
Ms = + 1 hA
2
MI = 0(1),E2
Ms = + 1 E+1 hA
2 2 MI = −1(1),E3
ΔE = 0 ΔE = hν = geβeH
MI = −1(1),E4
Ms = − 1
2 E−1 hA
2
H=0 MI = 0(1),E5
Ms = − 1 hA
Energy
2
MI = +1(1),E6
(b)
Absorption spectrum
H
H1−5 H2−5 H3−4

First-derivative
spectrum
H
ΔH = a ΔH = a
Fig. 9.12 (a) Energy levels of the deuterium atom (system S = 1/2, I = 1) in variable applied magnetic field
and constant frequency (n = constant), and (b) absorption and first-derivative spectra of deuterium atom.
hv a 1
H2 3 = + for MI = − (9.55)
g e βe 2 2
The absorption signals corresponding to these transitions are shown in Fig. 9.10(b). ESR spectrum for a system
1
with S = and I =1 [the 2H-deuterium, nitrogen (14N)]. The number of sublevels in this case = (2I + 1) = (2 × 1
2
+ 1) = 3 and MS = +1, 0, −1. The transition occurs at three resonance fields [Fig. 9.12(a)].
hv
H1 6 = − a for M I = +1 (9.56)
g e βe
hv
H2 5 = for MI = 0 (9.57)
g e βe
hv
H3 4 = + a for M I = −1 (9.58)
g e βe
Thus, the ESR spectrum for such a system consists of three lines of equal intensity as shown in Fig. 9.12(b).
Generally speaking, for a system where a single nucleus interacts with one unpaired electron, 2I + 1 lines of
equal intensity are obtained. They are separated by the hyperfine splitting constant (a).
.
(d) ESR Spectrum of C H2 (System with S = 1/2, I = 1/2, I = 1/2) In this system, the total nuclear spin of
two equivalent protons = 2 × 1/2=1 so MI can take on (2 × 1 + 1) = 3 possible values, viz.,, +1, 0, −1. Thus the
⎛ 1⎞
magnetic field of two equivalent protons will split the two paramagnetic electron spin states ⎜ M S = ± ⎟ into six
⎝ 2⎠
sublevels characterised by the nuclear spin quantum numbers MI = +1, 0, −1, −I, 0, +1. The splitting is displayed
1 1 • •
in Fig. 9.13(a). Just like S , I = system, i.e. C H , the transitions for C H 2 system also occurs at three reso-
2 2
nant magnetic fields H1−6, H2−5 and H3−4 except that the degeneracy of the middle transition is two.
hv
H1 6 = − a for M I = +1 (degeneracy αα 1) (9.59)
g e βe
hv
H2 5 = for M I = 0 (degeneracy αβ
β , βα
β 2) (9.60)
g e βe
hv
H3 4 = + a for M I = −1 (degeneracy ββ 1) (9.61)
g e βe
Thus, the absorption spectrum of C H 2 consists of three peaks with intensity ratio 1 : 2 : 1 [Fig. 9.13(b)].
The above treatment may be extended to the interaction of a paramagnetic electron with varying number of
⎛ 1⎞
equivalent protons. The splitting of unpaired electron levels ⎜ M S = ± ⎟ into sublevels by magnetic field of
varying numbers of equivalent protons is shown in Fig. 9.14. ⎝ 2⎠
(a)
MI = +1(1),E1
Ms = + 1 hA/2
2
MI = +0(2),E2
Ms = + 1 E+1 hA/2
2
2 MI = −1(1),E3
ΔE = 0 ΔE = h ν = geβeH
MI = −1(1),E4
Ms = − 1
2 E−1 hA/2
2
H=0 MI = 0(2),E5
Ms = − 1 hA/2
2
MI = +1(1),E6
(b)
Energy
Absorption spectrum
H1−5 H2−5 H3−4

First-derivative
spectrum
ΔH = a ΔH = a
Fig. 9.13 (a) Energy levels of the free radical with two equivalent protons (system: S = 1/2, I = 1/2
•
l = 1/2) in a variable applied magnetic
field and constant frequency (n = constant), and (b) absorption and first derivative ESR spectra of C H2 .
+ 5 (1)
2
+2(1) 3
+ 3 (1) + (5)
2 2
+1(4)
+1(1)
Ml = + 1 + 1 (3) + 1 (10)
2 2 2
Ms = + 1 0(2) 0(6)
2
Ml = − 1 − 1 (3) − 1 (10)
2 2 2
−1(1) −1(4)
− 3 (1) − 3 (5)
2 2
−2(1)
− 5 (1)
2
Ms = ± 1
Energy
2
− 5 (1)
2
−2(1)
− 3 (1) − 3 (5)
2 2
−1(1) −1(4)
Ml = − 1 − 1 (3) − 1 (10)
2 2 2
Ms = − 1 0(2) 0(6)
2
Ml = + 1 + 1 (3) + 1 (10)
2 2 2
+1(1) +1(4)
+ 3 (1) + 3 (5)
2 2
+2(1)
+ 5 (1)
2
0 1 2 3 4 5
Number of Protons
Fig. 9.14 Hyperfine energy levels, arising from the interaction of an unpaired electron with varying numbers of equivalent protons.
The quantities with brackets represent the intensity of the lines in a multiplet.
In conclusion, for a system where n equivalent nuclei First order (a)

First order (b)
(protons) interact with one paramagnetic electron, 2nl + 1
lines are obtained. The intensity of the lines corresponds to
the coefficients of the binomial expansion (1 + x)n.
(e) Second-order Splitting When hyperfine coupling

energy (hA) is very large as compared to that of the electron
energy (ge be H ) or when a small magnetic field is used, Second order Second order
additional splitting of some line may occur, which is known Fig. 9.15 Second order splitting (a) 1:2:1 triplet arising
as ‘second-order splitting’ (Fig. 9.15). This is a quantum from two equivalent nuclei and I = 1/2, (b) for 1:3:3:1 quartet
mechanical effect. arising from three equivalent nuclei, I = 1/2.
9.8.2 Anisotropic Hyperfine Interaction

This is a nonquantum mechanical interaction and is not restricted only to s-electrons, The contribution to split-
ting by anisotropic splitting as worked out from classical physics is given by h(mN(Z)/r3) (3cos2q −1) where mN(Z)
is the Z-component of nuclear magnetic moment, mN at a distance r from a paramagnetic electron with magnetic
moment me,. r subtends an angle q with the direction of the applied field. Both mN and me are parallel to the applied
field direction (Fig. 9.16). mN(Z) can take on 2I + 1 values. The anisotropy in coupling constant has the same angular
dependence as the g-value. Thus Eq. (9.27) may be written to take account of anisotropy in both g and A for an
axial system as
⎛1 2 ⎞
E g + g β H M S hA h MI MS
⎝3 3 ⎠
⎡1
+ (g − g ⊥ ) βe H M S + hB M I M S ⎤⎥ 3 2
θ − 1) (9.62)
⎣3 ⎦
Here, B is the anisotropic frequency factor (MHz). It is denoted by B⊥, when this interaction is perpendicular
1
to the direction of the applied field. For M S = ±
2
⎛1 2 ⎞ ⎡1
ΔE = hv =
⎝3
g + g⊥ ⎟ β H + A h M I +
⎠
( − ) βe H + B h M I ⎤⎥ 3 2
θ − 1) (9.63)
3 ⎣3 ⎦
If there is a fast tumbling of molecules in a system (samples in solution), the anisotropic contribution aver-
ages to zero and only the isotropic (contact) interaction is left, i.e. Eq. (9.62) reduces to Eq. (9.27). On the other
hand, in a powdered sample, the crystals are randomly oriented, and a wide range of angles are present. Conse-
quently, a poorly resolved asymmetric broad line spectrum results as compared to the spectrum in the solution
phase, e.g. the spectra of di-t-butyl nitroxide in liquid ethanol at 292 K [Fig. 9.17(a)] and in solid gas at 77 K
H μN(z)
θ
μe
Fig. 9.16 Interaction of nuclear magnetic dipole mN and electron magnetic

moment me. Magnetic flux lines associated with mN are indicated.
[Fig. 9.17(b)] show that the lines are broadened and poorly (a)
resolved due to anisotropic effects in the latter as compared 15 G
to the former. The lines are also shifted in position in the solid
glass spectrum as compared to the spectrum in the solution
phase. It is to be noted that an asymmetry of the absorption
lines in the solid spectra is caused by the anisotropy of the
g-values. The anisotropic hyperfine interaction parameters (b)
determined from measurements on oriented single crystals
of di-t-butyl nitroxide radical, with the magnetic field suc-
cessively applied along each of the three principal axes of the
molecule are listed in Table 9.6. The parameters ax, ay, and az,
determined on oriented crystals contain both isotropic and
anisotropic contributions to the hyperfine splitting. The value 30 G
of az reveals that the unpaired electron spends most of its
Fig. 9.17 ESR spectrum of di-t-butyl nitroxide radical (a) in
time in the 2pp orbital that lies along the Z-axis.
liquid ethanol (at 292 K), and (b) in solid glass (at 77 K).
The experimental values of g and a measured in solution
are in good agreement with that calculated from gx, gy, gz
and ax, ay, az respectively measured on the radical in oriented crystals. In general, ESR spectra of systems in a
gas phase are complicated since rotational transitions also occur in the microwave region of electromagnetic
radiation. However, the spectra of halogen atoms, i.e. F, Cl and Br are simple. These atoms exist in a 2 P3/ 3 2 (L = 1,
S = 1/2, J = 3/2) state and give g = 4/3. The degeneracy of this state is (2J + 1) and the possible number of transi-
tions corresponding to (2J + 1) will be 2J. Thus the number of lines expected in the spectra of halogen atoms is
2J(2I + 1), i.e. six for fluorine atom and twelve each for chlorine and bromine atoms. This has been experimentally
verified in the respective halogen atoms. These effects are not observed in the ESR spectra of solutions and solids
since rotational levels are nonquantised and orbital contributions are largely quenched.
Table 9.6 ESR parameters for the di-t-butyl nitroxide.
gx 2.0089
gy 2.0061
gz 2.0027
1
( x y z ) 2.0059
3
ge (solution) 2.0060 ± 0.0002
ax 7.1 ± 0.5
ay 5.6 ± 0.5
az 32.0 ± 1.5
1
( x y z ) 14.9 ± 0.8
3
a (solution) 15.1 ± 0.5
‘a’ values are in gauss.
9.9 ZERO-FIELD SPLITTING (FINE STRUCTURE TERMS)

AND KRAMER’S DEGENERACY
There are certain systems which contain more than one paramagnetic electron (even or odd 1) rather than a
single unpaired electron. When the number of electrons is even, the two electrons may be present on the same
atom or one electron each on the two atoms of a dimer. The degeneracy of spin states of electrons in such atoms
is resolved by the internal magnetic field (dH) of the electrons even in the absence of an external magnetic field.
This phenomenon of removal of degeneracy of spin states by the internal magnetic field of paramagnetic electrons
is termed as zero-field splitting and is also called fine splitting rather than hyperfine splitting and occurs mainly
in crystals [Fig. 9.18(b)]. Such a resolution of spin states is encountered in systems having noncubical crystal
fields, e.g. trigonal, tetragonal and rhombic, spin-orbit and spin-spin interactions. When the system contains an
odd number of electrons, the spin degeneracy of energy level (and no MS = 0 state) remains doubly degenerate;
this is known as Kramer’s degeneracy which can be removed only by the application of an external magnetic field.
For even number of paramagnetic electrons, the spin degeneracy may be resolved by the crystal field.
For a system with two unpaired electrons on the same atom ⎛⎜ S 1 ⎞
1 1
S2 = , S d M S +1 1, 0, 1⎟ , the
⎝ 2 2 ⎠
state S is made up of two unpaired electrons, i.e. S1 and S2. Each of the electrons generates a small internal mag-
netic field (dH) which contributes to the external field. When the spins of the electrons are parallel to the applied
field (H), the field experienced by the system will be H + dH while H − dH when the spins are anti-parallel to
the applied field direction, i.e. spin state MS = +1 will sense a magnetic field H + dHr, and spin state MS = −1 will
sense a magnetic field H − dH, both the states shift upward by a factor D [Fig. 9.18(b)] called the dipolar shift
or zero-field shift which is independent of the applied field. Since at a fixed frequency, the resonance condition
is hv = ge be H, the energy for the transition MS = 0 → 1 will be large as compared to the transition MS = −1 → 0
[Fig. 9.18(b)]. Thus in the absence of internal magnetic field, the transitions (by selection rule, ΔMS = ± 1) among
the degenerate levels corresponding to MS = ±1, 0 will exhibit a single line spectrum [Fig. 9.18(a)] whereas a two-
line spectrum is expected in the presence of internal magnetic field [Fig. 9.18(b)]; e.g. ground state of d8Ni(II), i.e.
3
A2g in octrahedral field when spin-orbital coupling is large exhibits zero-field splitting. It is to be noted that when
dH is small, it is likely that both the MS = ±1 states will be above the MS = 0 state. Conventionally, the MS = ± 1
states are shown as degenerate states above the MS = 0 state.
Further, if one paramagnetic electron is located on each of the two atoms of a dimer (electron of each atom
influences the electron of the other atom), then in addition to the ΔMS = ±1 transitions, one can also observe
resonance lines corresponding to ΔMS = +2 transition. Consequently, a three-line spectrum results (Fig. 9.19).
Since ΔMS = +2 transition is forbidden according to the ESR rules, it appears as a weak signal. So such a transi-
tion is best observed at low temperature on a highly sensitive instrument. These transitions (ΔMS = +2) occur at
about half the resonant magnetic field to the ΔMS = ±1 transitions and, therefore, are also called half-resonance or
transitions. Thus, the ESR spectrum of such a system should give a value of ge ~2 for ΔMS = ±1 transitions and
a value of ge ~ 4 for the ΔMS = +2 transitions since the resonance condition is ΔE = hv = gebeH, e.g. dimeric 3d9
Cu(II) or dimeric 3d1 oxovanadium (IV) complexes. The influence of zero-field splitting parameters D and E on
the ESR spectrum is shown in Fig. 9.19. ESR spectra, of (C6H5)2C: and (C6H5)2N: prepared by the photo excitation
of (C6H5)2CN2 and (C6H5)2N3 respectively reveal that the two paramagnetic electrons mainly reside on the carbon
and nitrogen atoms in the respective systems. Further, we know that H varies with q (the angle between Z-axis
and applied field direction) and for a rigid solution containing randomly oriented triplet molecules, q can take
(a) (b) Ms = +1
Ms = +1
ΔMs = +1
ΔMs = +1
Ms = ±1.0 Ms = 0 Ms = 0
ΔMs = −1
ΔMs = −1
Ms = −1
D
Ms = −1
δH ≠ 0
δH = 0
H
Fig. 9.18 The energy-level diagram and spectra of a molecule or ion with S = 1 (a) zero internal
field (dH = 0), and (b) with non-zero internal field (dH ≠ 0).
Ms = +1
2E
Ms = +1
ΔMs = +1
Ms = −1 ΔMs = +2
D (D − E) Ms = 0
Ms = 0
ΔMs = −1
Zero Field
Leve (δH ≠ 0)
Applied Field Ms = −1
H
Fig. 9.19 The energy-level diagram and spectra for a dimeric molecule or ion in the presence of zero-field splitting.
on any value and the absorption spectrum will then be a series of overlapping lines (arising from the ΔMS = ±1
transitions) of varying separations, the overall signal will be too broad to be detected. This problem can be tackled
by preparing a diluted single crystal of triplet state—a molecule in some other molecule with geometry identical
to the triplet molecule. In such a situation, all the triplet molecules will be aligned in the same manner and there is
only one value of q, e.g. ESR spectrum of a dilute single crystal of triplet naphthalene in durene (at 77 K) gives a
g-value of 2.003. The value of g reveals that Kramer’s degeneracy is not due to crystal field effect on spin-orbital
coupling but because of s-s coupling between the two p electrons in the triplet state. For triplets (S = 1) the s-s
interaction is often solely responsible for zero-field splitting. On the other hand, the most important feature of
the half-resonance lines is that the anisotropic line broadening is reduced and it is often possible to study these
transitions in randomly oriented molecules, e.g. ESR spectrum of triplet state naphthalene produced by irradiating
it with UV light and stabilised at low temperature. In general, ESR studies of triplet molecules are difficult and
hence limited since their lifetimes are too short at high temperature and anisotropic line broadening effect is too
great. Further, in a triplet molecule, the energy due to electron–nucleus contact interaction, which is of primary
importance in the doublet state in solution is smaller than the dipolar interaction. Therefore, ESR spectra of triplet
molecules rarely show any resolved hyperfine structure due to this contact interaction.
Let us now extend the foregoing treatment to systems with odd number of electrons other than one: Such systems
exhibit Kramer’s degeneracy, e.g. d3Cr (III), high spin d5Mn Ms = +5/2
(II). The energy level diagram for the S6 state of Mn (II) (d5)
is displayed in Fig. 9.20. Zero-field splitting generates pairs
5 3 1
of degenerate spin states, M S = ± , ± , ± (Kramer’s
2 2 2 Ms = +3/2
degeneracy). These pairs have slightly different energies from
each other because of zero field already present. Under the
influence of the applied field, each of these doublets split into ±5/2 Ms = +1/2
singlets resulting into six levels. The allowed ESR transitions d5 ±3/2
(by ΔMS = ±1) between these levels result in fine line ESR
spectrum. Further, each of these spectral lines interacts with ±1/2 Ms = −1/2
5 ⎛ 5 ⎞
spin of magnitudes, I = to give ⎜ 2 × + 1⎟ = 6 hyperfine
2 ⎝ 2 ⎠ Ms = −3/2
components. Components of hyperfine splitting have vary-

Ms = −5/2
ing intensity: the intensity is greatest for the central lines and
smallest for the outermost lines. In simple cases the separa-
tion between lines varies as (3 cos2q − 1) where q is the angle Zero-Field
Levels
Applied
Field
Nuclear
Splittings
between applied field and Z-axis. Fig. 9.20 The energy-level diagram for Mn(II) (d6).
In certain cases where the magnitude of zero-field splitting exceeds the energies of ESR resonances, it is
possible to observe weak ESR transitions corresponding to ΔMS = ±2 but not corresponding to ΔMs = ±1, e.g. in
⎛ 7 51 ⎞ ⎛ 7 ⎞
V3 V ⎟ , the line corresponding to ΔMS = ±2, further splits into ⎜ 2 × + 1⎟ = 8 components.
⎝ 2 ⎠ ⎝ 2 ⎠
9.10 NUCLEAR QUADRUPOLE EFFECTS IN ESR SPECTRA OF SOLIDS

We know that ESR theory is the simplest for liquid solution and it is in these systems most of its chemical
applications are found. Let us see how quadrupole effects operate in the ESR spectra of solids. To start with, the
Hamiltonian operator for the paramagnetic system possessing both the magnetic and quadrupole nucleus can be
expressed as
Hˆ FS = Hˆ M + Hˆ Q
μN
Hˆ a( )− ( ) + A( 2
) (9.64)
I
where the terms and symbols have their usual meanings and the terms within the braces are vectorial in charac-
ter. The first term stands for the interaction between nuclear and electron spins, the second terms stands for the
direct interaction between nuclear spin and external magnetic field and the third term stands for the quadrupole
interaction.
The first term may arise due to the following two mechanisms. (i) In the first mechanism, the magnetic field at
the nucleus is sum total of external magnetic field (H) and the average magnetic field produced by the unpaired
electron. Since the electrons move faster than the nuclei, so the field produced by the electron can be replaced by
one or more magnetic dipoles residing at a particular points on the molecular frame work. Since the nuclear dipole
and the equivalent electron dipole are always parallel to the external filed, the field produced by the electron at the
nucleus depends on the angle(s) between the vectors joining the nucleus to the equivalent dipole(s). The average
of this interaction over all directions in space is zero. Hence this term does not contribute to electron resonance
spectra of substances in the gas or liquid phases. (ii) The second mechanism of electron-nuclear magnetic interac-
tion arises if there is a spin imbalance at the nucleus in question, i.e. if the wave function describing the motion of
unpaired electron has a finite value at the nucleus. This term is independent of orientation and is responsible for
the hyperfine structure of spectra of gaseous or liquid substances. We know that the factor representing the sum
of these two effects is a tensor quantity ‘a’ and is uniquely given by its two components a and a⊥ , parallel and
perpendicular to the symmetry axis respectively. In the absence of quadrupole coupling the eigen values of the
fine structure Hamiltonian HFS, are given, to first order by
μ N HM I a g a g⊥
E FS a SMI −
aM [ cos 2 θ + in 2 θ ] (9.65)
I ag ag
a2 g 2 a2 g ⊥2
where a = cos 2
θ + i 2θ
sin
g2 g2
The quadrupole effects will manifest themselves in the ESR spectra provided the nuclear-electronic magnetic
moments are coupled, i.e. the tensor a ≠ 0. There is no direct interaction between a nuclear quadrupole and an
electronic magnetic dipole so that if a quadrupole nucleus in a molecule is at a very large distant apart from the
paramagnetic site, the electron-spin resonance spectrum will reveal nothing about the quadrupole coupling con-
stant.
Further, if we assume that a quadrupole interaction energy is much smaller than magnetic interaction of the
nucleus with the electron spin and external magnetic field, i.e. the axis of precession of nuclear spin remains
unchanged, then the quadrupole interaction introduces a further term into the expression for the eigen values of
HFS given by
eq 2Q ⎡ 3M M I2 − I (I + 1) ⎤ ⎡ 3a2 g 2 cos 2 θ ⎤
EQ = ⎢ ⎥⎢ − 1⎥
4I (I − 1) ⎣ 2 ⎦⎣ a2 g 2 ⎦
e 2qQ
= − [3M I2 − I (I + 1)]]f ′(( ) (9.66)
4I (I − 1)
Selection Rules
(i) ΔMI = 0, the limit of first-order theory, the quadrupole coupling does not produce any effect on the ESR
spectrum.
(ii) ΔMI = ±2, the quadrupole transitions become allowed for directions of the magnetic field at angle to the
symmetry axis.
(iii) ΔMI = ±1, becomes also allowed when the magnetic coupling tensor “a” is highly anisotropic and the
magnetic field is in directions neither parallel nor perpendicular to the symmetry axis.
Let us now describe the effect of quadrupole coupling for a nucleus with I = 3/2. The hyperfine energy levels are
given by
3a 3 μ N H 3
E 3 =± − f A f ′(θ )
+
2 4 2 I 2
1a 1 μ N H 3
E 1 =± − f A f ′(θ )
+
2 4 2 I 2
1a 1 μ N H 3
E 1 =∓ + f A f ′(θ ) (9.67)
−
2 4 2 I 2
3a 3 μ N H 3
E 3 =∓ + f A f ′(θ )
−
2 4 2 I 2
1
1 2
Here, f ( ) [ (I t 2 θ ]2
) tan
2
e 2qQ
where the angle q has its usual meanings and A = . The upper signs in the formula give the position of
4I (I − 1)
1 1
the energy levels relative to g β H for M S = + while the lower signs give the corresponding position of the
2 2
1 1
energy levels relative to − g β H for M S = − . The energy-level diagram is given in Fig. 9.21 (a) and the cor-
2 2
responding spectrum for the various transition in Fig. 9.21(b) given below.
Electron-
Nuclear Nucleus–
Nuclear
Quadrupole Magneticfield
Magnetic
Interaction Interaction
Interaction
MI
+3/2
+1/2
MS = +1/2
−1/2
−3/2
−3/2
−1/2
MS = −1/2
+1/2
+3/2
(a)
(b)
Fig. 9.21 (a) Energy levels, and (b) spectrum of a paramagnetic molecule, S = 1/2, I = 3/2. Heavily shaded full lines are transitions for
ΔM = 0; light shaded full lines are for transitions ΔM = ± 1; dotted lines are ΔM = ± 2.
Figure 9.21 and the equations reveal the immediately that the relative sign of ‘a’ and e2qQ can be obtained from
ESR measurements. It is often probable to obtain probable absolute sign of ‘a’ on the theoretical grounds so that
the absolute sign of e2qQ is then revealed.
When the asymmetry parameter is also taken into account so that as a consequence thereof all three compo-
nents of ‘a’ are different. The directions of axes and value of η are given directly by the spectrum. The general case
in which none of the directions of the principal axes of g, ‘a’ and A coincide has not been treated theoretically but
in any case an analytical solution to this problem would not be possible.
In summary and also recall from Chapter 7 on NMR and Chapter 8 on NQR respectively that nuclei with spin
I ≥ 1 such as deuterium 14N, 35Cl, 37Cl, 63Cu, 65Cu [m1 (Cu65) > mI (Cu63)], 67Zn, 127I, etc., besides nuclear magnetic
moment (which makes them NMR active) possess electric quadrupole moment ‘eQ’ also. The electric quadrupole
moment interacts with the local electric field gradient (EFG), which is because of the electrons at the site of the
probe nucleus. This type of interaction is called the nuclear electric quadrupole interaction and causes a splitting
of the Zeeman energy levels. In most of the complexes with paramagnetic electron, the quadrupole interaction is
small or not observed.
However, such an effect has been observed in some Cu2+ complexes. Thus, in paramagnetic systems, the unpaired
electron experiences both the magnetic dipole moment and the nuclear quadrupole interaction. Consequently,
additional signals which are not permissible by the selection rule ΔMI = 0 appear in the ESR spectrum. Some-
times the signals permissible by both the selection rules ΔMI = ±1 and ΔMI = ±2 are also observed, e.g. the ESR
spectrum of K2Cu(SO4)2 6H2O shown in Fig. 9.22 exhibits four major peaks due to the interactions of Cu63(I = 3/2)
with the unpaired electrons. Cu65 with higher gyromagnetic ratio should also exhibit four peaks, of which only
two peaks, at the extremes of the spectrum, are resolved [Fig. 9.22(a)]. The four minor peaks, which are also split
into two components because of the different gyromagnetic ratios of Cu63 and Cu65 nuclei are permissible by the
selection rule ΔMI = ±1. An expanded view of two of these peaks has been shown in Fig. 9.22(b).
ΔM1 = ±1
ΔM1 = ±1
(a) Cu65 Cu65
Cu63
Cu63
(b)
Fig. 9.22 ESR spectrum of K2Cu(SO4)2· 6H2O: (a) normal view of the spectrum,
and (b) expanded view of the spectrum.
9.11 RULES FOR THE PREDICTION OF NUMBER OF HYPERFINE

LINES AND THEIR RELATIVE INTENSITIES
In general, the number of hyperfine lines in the ESR spectrum of the free radical system having a set of n equiva-
lent nuclei with nuclear spin, I is given by the quantity (2nI + 1). However, if the electron is delocalised over a
set of n equivalent nuclei with nuclear spin I1 and another set of m equivalent nuclei with nuclear spin I2, then the
maximum number of possible hyperfine lines is given by the quantity (2nI1 +1) (2mI2 + 1), i.e. ∏i (2niIi + 1) where
i = 1, 2. Let us now apply these rules to free radicals with different structures.
⎛ 1⎞ ⎛ 1 ⎞
The interaction of the unpaired electron with n equivalent protons ⎜ I = ⎟ results in ⎜ 2 × × n + 1⎟ = n + 1
⎝ 2⎠ ⎝ 2 ⎠
lines whose relative intensities are proportional to the coefficients of the binomial expansion of (1 + x) (Fig. 9.23).
n
• i
Thus, the maximum number of lines expected in the ESR spectra of 3 and C6 H 6 (benzene anion radical)
would be four and seven respectively and same has indeed been observed experimentally [Figs 9.24(a) and (b)].
The intensity ratios of the quartet and septet also conform to the expected ratios of 1:3:3:1 and 1:6:15:20:15:6:1
respectively. It is important to note that the overall shape of the ESR spectra is influenced by the ratio of splitting
i
to line width (Fig. 9.25). In naphthalene anion radical ( ) /( N ), prepared by the reduction of naphthalene
with alkali metals in ether, there are two different sets of four equivalent protons, so the expected number of lines
⎛ 1 ⎞⎛ 1 ⎞
in the spectrum is ⎜ 2 4 × + 1 2 × 4 × + 1 25. However, the number of lines in the observed experimental
⎝ 2 ⎠⎝ 2 ⎠
spectrum [Fig. 9.24(c)] is 21. The difference may be due to overlapping of lines in the observed spectrum.
Electron 1
Proton
1 1 1
2 1 2 1
3 1 3 3 1
4 1 4 6 4 1
5 1 5 10 10 5 1
6 1 6 15 20 15 6 1
7 1 7 21 35 35 21 7 1
8 1 8 28 56 70 56 28 8 1
Fig. 9.23 Pascal’s triangle. The horizontal numbers are the coefficients of binomial expansion
for different number of equivalent protons interacting with an unpaired electron.
3G
3.8 G 1.8 G
4.9 G
(a) (c)
100 G
(b)
15.2 G
(d)
(e)
• •
Fig. 9.24 ESR spectrum of (a) C H3 (b) C6 H6 (c) C10 H8 (d) (CH3 )3C N OC (CH3)3, and (e) Mn2+ ion in water.
If the unpaired electron couples with non-equivalent protons then each proton will have its own coupling constant,
e.g. as shown below. In general, n non-equivalent protons will produce a spectrum with 2n hyperfine lines. The number
of lines predicted by the formula for equivalent protons, i.e.
(2nI + 1) will be less than the number of lines expected for 0.4G
non-equivalent protons, i.e. 2n because several of the possible
arrangements of the nuclear spins degenerate.•
The number CH3 C H CH3 CH2 C H
of hyperfine lines in the spectrum of H C — OH should be 27G 22G 33G 22G
22 = 4 with intensity ratio 1:1:1:1. The relative intensities of ESR
1
lines of free radicals in which electron is delocalised over nuclei having I >
are listed in Table 9.7. Three and five
2
peaks are expected for an electron delocalised over one and two (equivalent) nitrogens respectively. The ESR spec-
trum of (CH 3 )3 C N OC (CH 3 )3 exhibits three lines of equal intensity as expected (Fig. 9.24(a)]. ESR spectrum of
•
i
NH 3 formed by x-ray irradiation of ammonium perchlorate crystals is shown in Fig 9.26(a). The twelve-line pattern
could arise due to coupling of electron with nitrogen nucleus (2 I + 1) = (2 × 1 + 1) = 3 lines and coupling with three
⎛ 1 ⎞
equivalent protons, ⎜ 2 3 × + 1 4 lines. Thus, the three equivalent protons split each of the major three lines
⎝ 2 ⎠
into a quartet of equally spaced lines with intensity ratios of 1:3:3:1. Based on the intensity of lines and the separation
between them, the spectrum is illustrated in Fig. 9.26(b).
Number Ratio of component separation to component width

of protons 0.76 0.90 1.0 1.3 1.6
1
Intensity ratios 1-1
2
Intensity ratios 1-2-1
Intensity ratios 1-3-3-1
4
Intensity ratios 1-4-6-4-1
Intensity ratios 1-5-10-10-5-1
Intensity ratios 1-6-15-20-15-6-1

Fig. 9.25 Theoretical hyperfine structure curves for Gaussian derivative curves.
Table 9.7 Relative intensities of ESR lines for n equivalent nuclei with nuclear spin I.
I=½
n=1 1 1
n=2 1 2 1
n=3 1 3 3 1
I=1
n=1 1 1 1
n=2 1 2 3 2 1
n=3 1 3 6 7 6 3 1
I = 3/2
n=1 1 1 1 1
n=2 1 2 3 4 3 2 1
n=3 1 3 6 10 12 12 10 6 3 1
I=2
n=1 1 1 1 1 1
n=2 1 2 3 4 5 4 3 2 1
n=3 1 3 6 10 15 18 19 18 15 10 6 3 1
The first quartet involves lines 1, 3, 6 and 9; the second quartet consists of lines 2, 5, 8 and 11; and the lines 4,
7, 10 and 12 constitute the third quartet. The individual members of the three-line pattern have been located from
the centre of each quartet. The observed values of hyperfine splitting constants for the triplet (1:1:1) and each of
the three quartets are 18.1 and 25 G respectively. In case of transition metal ions, the hyperfine splitting arises due
1
to the coupling of metal nucleus having spin I > with the unpaired electron of its ion. Thus, the ESR spectrum
2 ⎛ 5 ⎞
of Mn2+ ions in water exhibits six lines of equal intensity [Fig. 9.24(e)] as expected, i.e. ⎜ 2 × + 1 6 whereas
⎝ 2 ⎠
. . ⎛ 3 ⎞
the expected number of lines in CuCl2 2H2O and CuSO4 5H2O is ⎜ 2 × + 1 4 with intensity ratios 1:1:1:1.
⎝ 2 ⎠
The dimer of Cu(II) acetatedmonohydrate exhibits seven lines with intensity ratios 1 : 2 : 3 : 4 : 3 : 2 : 1 in ESR
⎛ 3 ⎞
spectrum as expected, i.e. ⎜ 2 2 × + 1 7. Spectrum reveals the formation of metal–metal bond in the dimer
⎝ 2 ⎠
and the same has been confirmed by x-ray studies.
(a)
3 5 6 7 8 10
1 2 4 9 11 12
25 G (b)
18.1 G
•
Fig. 9.26 ESR spectrum of N H3 formed by x-ray irradiation of ammonium perchlorate crystals.
The ESR spectrum of bis-salicylaldimine Cu(II) as illustrated here consists of four major
signals and each of the four major signals consists of eleven lines of intensity ratio 1 : 2 : 3 : O
Cu
4 : 5 : 6 : 5 : 4 : 3 : 2 : 1. The four major signals in the spectrum are due to the interaction of 2
⎛ 3 ⎞ N C
unpaired electron of Cu with the nuclear spin I = 3/2 of Cu nucleus, i.e. ⎜ 2 × + 1 4.
2+ 63
⎝ 2 ⎠ HA HB
Now each of these four main signals undergoes splitting by two equivalent nitrogen atoms 2
⎛ 1 ⎞
(I =1) and two hydrogen (HA) atoms to yield (2 × 2 × 1 + 1) ⎜ 2 2 × + 1 15 lines.
⎝ 2 ⎠
HB hydrogen being away does not contribute to splitting. The overlapping between the expected lines results into
eleven lines in the observed experimental spectrum. It is to be noted that the splitting of main ESR signal into a
number of lines by the donor atoms of the ligand in the coordination complexes is termed super-hyperfine splitting.
Let us now examine the ESR spectrum of distorted tetrahedral dichloro (orthophenanthroline) Cu(II) in isomor-
phous dimagnetic dichloro (orthophenanthroline) Zn (II) which consists of four main groups and each of the main
groups contains five lines of intensity ratio 1 : 2 : 3 : 2 : 1. The expected number of super-hyperfine lines in each of
the four major groups which arises by the interaction of two nitrogen atoms (I = 1) of donor orthophenanthroline
3
is also (2 × 2 × 1 + 1) = 5. Although 35Cl has nuclear spin I = , the chloride hyperfine interaction has not been
2
observed as the chloride splitting constant is small in a Cu(II) complex. Further, the presence of super-hyperfine
lines in the ESR spectrum of a coordinate ion complex is an evidence to the delocalisation of the unpaired electron
of the metal ion to the ligand atom orbitals which is contrary to the basic assumption of crystal field theory, i.e.
‘the metal and the ligand orbitals do not overlap in coordination complexes.’
Problem 9.3: Predict the number of ESR lines for the triphenyl methyl radical, i.e (C6 H5 )3 C .
•
Solution In this radical, three-proton spin systems correspond to the ortho-, para- and meta-protons. Each
ortho- and meta-system contains six protons, the para-system contains three protons. Thus, the total number of
2
ESR hyperfine spectral lines is : ⎛⎜ 2 6 × + 1⎞ ⎛ 2 × 3 × + 1⎞⎟ = 196. However, in the observed spectrum about
1 1
100 lines are resolved. ⎝ 2 ⎠ ⎝ 2 ⎠
Problem 9.4: Predict the number of lines in the ESR spectrum of ethyl free radical, i.e. C H3 CH 2 .
Solution CH3 and CH2 are three- and two-proton systems in this radical, Consequently, the total number of
hyperfine ESR lines will be
⎛ 1 ⎞⎛ 1 ⎞
⎜⎝ 2 3 × + 1⎠ ⎝ 2 × 2 × + 1⎠ 12
2 2
(Number of observed lines is also twelve.)
Problem 9.5: Analyse the hypothetical spectrum shown in Fig. 9.27 (a) for the radical H.X+. H − X + ↔ H − X +
(I = 1 for X).
⎛ 1 ⎞
Solution The hypothetical radical H·X+ exhibits ⎜ 2 × × 1 1⎟ = 2 major signals of equal intensities
⎝ 2 ⎠
⎛ 1⎞
[Fig. 9.27(b)] due to coupling of unpaired electron with hydrogen atom ⎜ I = ⎟ . Each of these two signals splits
⎝ 2⎠
into (2 × 1 + 1) = 3 lines due to the interaction of X atom (I = 1). Thus, the expected number of lines for the hypo-
thetical radical H.X+ is 3 × 2 = 6. However, there are only five peaks in the hypothetical spectrum of the radical
[Fig. 9.27(a)]. Most probably, the two innermost lines are not well resolved [Fig. 9.27(c)], and consequently a five
line spectrum results.
(a)
(b)
(c)
Fig. 9.27 Hypothetical ESR spectrum of H . X+ (I = 1 for X) radical.
Problem 9.6: Analyse the ESR spectrum shown in Fig. 9.28 for Mo(V) complex (4d1), i.e. K3 [Mo(CN)8] in
water at room temperature. The natural abundance of different isotopes of molybdenum is 94Mo and 96Mo (I = 0) =
⎛ 5⎞ ⎛ 5⎞
75 percent, 95Mo ⎜ I = ⎟ = 15.78 per cent, Mo97 ⎜ I = ⎟ = 9.60 per cent. At a high gain of ESR spectrometer, each
⎝ 2⎠ ⎝ 2⎠
of the six weak lines (marked with arrows) exhibits two small peaks similar to the ones on either side of the central
line.
Solution The nonmagnetic isotopes (94Mo, 96Mo) with I = 0 will give (2 × 0 + 1) 1 ESR line. On the other hand,
5 ⎛ 5 ⎞
each of the magnetic isotopes 95Mo and 97Mo (with I = ) will give ⎜ 2 × + 1 6 ESR lines. However, the
2 ⎝ 2 ⎠
lines due to 95Mo and 97Mo overlap to give rise to only six broad lines in the spectrum. These lines are symmetri-
cally distributed on either side of the central line due to nonmagnetic isotopes, i.e. 94Mo and 96Mo. The intensity of
ESR line for an isotope is directly proportional to the natural abundance and inversely to the number of lines, i.e.
75 6
(2I + I) it exhibits. Thus, the intensity of the line corresponding to nonmagnetic isotopes (94Mo, 96Mo) is = × = 18
25 1
times more than those of the magnetic isotopes. The two small peaks on either side of the central line are due to
⎛ 1 ⎞
the hyperfine interaction of 13 C ⎜ 2 × + 1 2⎟ of CN−. The splitting of each of the six broad lines into two peaks
⎝ 2 ⎠
is also due to the hyperfine interaction of 13C of CN−. This splitting pattern shows the presence of Mo—CN rather
than Mo—NC bond in the molecule.
9.12 BASIC PRINCIPLE OF AN ESR SPECTROMETER

The block diagram of a spectrometer is shown in Fig. 9.29, The main seven components of the spectrometer are
klystron power supplies, klystron, magnet, cavity system, modulator, phase sensitive detector, and recorder. As
shown in Fig. 9.29, klystron is a source of monochromatic microwave radiation. The frequency of radiation is
Klystron
power
supplies
Klystron
Cavity
g0
Magnet
Modulator Oscilloscope
A0 H
Phase
Chart
sensitive Amplifier
recorder
detector
Sweep Computer
500 G supply
Fig. 9.28 ESR spectrum of K3 [Mo(CN8)] Fig. 9.29 Block diagram of an ESR spectrometer.
in water at room temperature.
determined by the voltage applied to the klystron from the klystron power supplies. Power in klystron used is usu-
ally a few hundred milliwatts. The frequency of microwave radiation is stabilised by an automatic frequency con-
trol (AFC) system (not shown in Fig. 9.29) which works on the voltage. Heat produced by the klystron is removed
by circulating water. The lifetime of modern klystrons is about 7000 hours. The frequency range of klystrons is
as under.
Band S X K Q E
Approximate frequency (in GHz) 3 9 24 35 70

Approximate field (in kG) 1.1 3.3 8.5 12.5 25
There are differences between the European and American names for the bands, so the frequencies should be
reported also. Commonly used frequencies are around 9.5 GHz (X-band) and 35 GHz (Q-band). The radiation
is transmitted with the help of a wave guide through the sample contained in the sample tube placed in a cavity
located between the poles of a magnet. A wave guide consists of copper or brass tube (2.2 cm by 10 cm) with
silver or gold plating inside to produce a highly conducting flat surface and has the dimensions close to the wave-
length of radiation. When a sample and sample holder are introduced into the cavity, the quality factor, Q of the
cavity, which is a measure of its power to store energy, should not change too much. Q is given by
2 ( maximum microwave energy in the cavity )
Q= (9.68)
energy lost per cycle
Rectangular cavity is used for large samples, especially for liquid samples. A cylindrical cavity is used for
gaseous systems and liquid samples in capillaries. A tubing of 3−5 mm internal diameter with a sample volume
0.15−0.5 ml can be used with samples which do not possess a high dielectric constant. For samples with a high
dielectric constant, flat cells with a thickness of about 0.25 mm and sample volume of 0.05 mL, are often used.
Rotatable cavities are used for studying anisotropic effects in single crystals and in solid samples. The microwave
radiation enters and leaves the sample cavity by the same hole, called the iris. An electromagnet is a source of
static magnetic field and it should have a large field region (500−5000 G is required to handle samples whose
g-factor ranges from 1.5−6), a wide air gap and good field homogeneity, it should be about 1 part in 106 for solu-
tion studies. The line widths for paramagnetic ions and for free radicals in solid matrices are rarely less than
a few gauss, so that homogeneity and stability might be as low as 1 part in 103 before serious line broadening
takes place. The stability of a magnet is obtained by regulation with a Hall crystal (Indium arsenide; In As) in the
magnet gap. The crystal gives voltage proportional to the field and any change can be corrected with a feedback
circuit to the current. The modulation system consists of small Helmholtz coils, one on each side of the cavity,
along the axis of the magnetic field. The strength of the magnetic field may be determined from the current flow-
ing through the Helmholtz coils.
32π n
H= ⋅I (9.69)
10 125 rc
where, n is the number of turns in each coil, rc is the radius of coils, I the current (in amp) flowing through the
coils.
The microwave radiation absorbed by the sample is detected by a phase-sensitive detector which is usually the
semiconducting silica crystal. A superheterodyne electronic system is also used, especially for lines which are
quite narrow. The absorption signal is amplified and the absorption spectrum is displayed on an oscilloscope or
recorded on paper chart. The phase detected signal is usually represented as the first derivative. The absorption
curve and the first and the second derivatives are shown in Figs 9.2 and 9.3. Modern ESR spectrometers are coupled
with computers for storing and analysing data. A computer is especially convenient for storing time-dependent
signals which change too fast to be monitored accurately by mechanical recorders.
A good spectrometer should have maximal sensitivity. The sensitivity at room temperature of X-band (9.5 GHz)
ESR spectrometers is given by
ΔH PP
N min = 1 × 1011 (9.70)
τ
where, Nmin is the minimum number of detectable spins per gauss, ΔH the width between the deflection points on
the derivative absorption curve and t is the time constant of the detecting system which is inversely proportional
to the bandwidth of the detection circuit. The sensitivity of ESR spectrometers in the X-band is 5 × 1010 spin G−1
for 100 kHz modulation and in the Q-band 6 × 109 spin G−1 for 100 kHz modulation. It is to be noted that for a
sample of very small dielectric loss, the minimum sensitivity is about 6 × 1014 spin G−1 to barely see the line.
Further, in order to measure line widths and hyperfine splittings, the accurate calibration of the field is car-
ried out with standard sample, in which the line positions can be carefully determined. The standard sample can
be introduced in the dual cavity and the spectra recorded simultaneously. The standards used are Mn2+ in SrO
powder characterised by 6 lines, 1.6 G wide, totally over 420 G, splittings 84 ± 0.2G, peroxylaminedisulphate ion
in solution characterised by 3 lines,
– 250 mG wide, outer splitting 26.18 G, splittings 13.1 ± 0.004G; tetracyano-
ethylene anion radical (TCNE)• in solution characterised 9 lines, 100 mG wide, outer splitting 1.6 G, splitting
1.57 G, Wurster’s blue perchlorate characterised by 39 lines listed for half spectrum, total spectrum 86G, accuracy
± 0.01G.
9.13 DETERMINATION OF g-VALUE

The position of resonance of ESR line of an unknown sample can be determined if operating microwave fre-
quency (v) of the instrument and resonance magnetic field (Hr) of the sample is known.
⎛ hv ⎞ v( )
g un = ⎜ ⎟ = 714.44 (9.71)
⎝ βe H r ⎠ Hr ( )
If the microwave frequency (v) is not accurately known, then
⎛ ΔH ⎞
g g stdtd 1 − (9.72)
⎝ H r ⎟⎠
where gstd is the g-value of the standard free radical which is used to calibrate the ESR spectrum. Most commonly
used standard, α, α'-diphenyl-b-picrylhydrazyl (DPPM) contains 1.53 × 1021 unpaired spins per gram. Line width
and ge values for DPPH are listed in Tables 9.8 and 9.9 respectively. ESR spectra of DPPH in different media are
displayed in Fig. 9.30. ΔH = Hr(un) − Hr(std) is negative if Hr(un) < Hr(std) and positive if Hr(un) > Hr(std). From Eq. (9.68)
we note that for negative ΔH, g > 2 whereas g < 2 for positive ΔH. Substandards can be prepared by dilution with
carbon black.
In addition to DPPH, the other primary standards are (i) fresh, deep blue CuSO4 . 5H2O crystal; (ii) fresh MnSO4 . H2O;
(iii) K2NO(SO3)2 in aqueous solution; (iv) O2 gas; and (v) solutions of peroxylaminedisulphonate.
Some secondary standards which are more stable than primary standards may also be used to calibrate the
ESR spectrum, but before use they should be calibrated against primary standards with spin concentration that
are determined by independent methods. Various secondary standards in use are (i) synthetic ruby crystal; which
can be placed in the cavity permitting simultaneous observation of standards and unknown sample. It shows a
strong resonance (g = 1.4); (ii) charred dextrose, which does not change properties between 4 K and 500 K; and
(iii) powdered coal (pitch) diluted with KCl. Varian supplies two such standards (peak-to-peak line width 1.7
G, ge = 2.0028) of different concentrations (1 × 1013 and 3 × 105 spin cm−1 ) with each spectrometer; (iv) Mn (II)
in MgO, CaO and CaCO3.
Table 9.8 Line width in DPPH samples crystallised from various solvents.
Solvent ΔH1/2 (G)
v = 300 MHz v = 9400 MHz
295 K 90 K 295 K
Benzene 6.8 4.6 4.7
Toluene 2.9 2.6 2.6
Xylene (mixture) 2.5 2.2 2.3
Pyridine 5.3 5.0 5.0
Bromoform 2.2 2.5 2.5
Carbon tetrachloride 1.9 2.7 2.3
Chloroform 1.7 2.1 2.0
Carbon disulphide 1.3 1.3 1.5
Table 9.9 Variation in ge-values of DPPH at room temperature and 77K.
Sample ge-value
Single Crystal* 2.0027−2.0039
Single Crystal 2.0035−2.0041
2.0028−2.0O38
2.0030−2.0040
Powder* 2.0037
Polycrystalline 2.0036 (g|| = 2.0028 and g⊥ = 2.0039)
*Difference in the ge-value for single crystals may occur, because the number of spins in crystal
samples depends on the solvent used for the growing of the crystal. The ESR lines of DPPH in the
whole temperature range from 293 K−1.7 K have Lorentzian shape.
(a) (b)
1.9 G
(d)
(c)
8.9 G
20 G
Fig. 9.30 ESR spectra of α,α'-diphenyl-b-picrylhydrazyl (DPPH) in different media: (a) solid powdered sample, (b) saturated solution
of DPPH in tetrahydrofuran, (c) saturated solution diluted by a factor of 103, and (d) in benzene solution (5 × 103 mole lit−1).
9.14 FOURIER TRANSFORM ESR SPECTROSCOPY (FTESRS)

The principle of FTESR is similar to FTNMR except that in FTESR, a microwave pulse is used instead of
a radio frequency pulse. FTESR technique is faster than FTNMR because the relaxation times are much
faster in ESR. ESR requires 10−20 ns pulses. The main problem encountered in FTESR is the dead time of
the pulsed ESR instrument that prevents observation of FID during the first 50−100 ns after the pulse. It is,
therefore, difficult to measure broad resonance lines employing FTESR. However, with FTESR method, the
ESR spectrum can be recorded within a few microseconds. This enables the detection of transient paramag-
netic species.
9.15 ELECTRON NUCLEAR DOUBLE RESONANCE (ENDOR)

ENDOR is a method to detect changes in the amplitude of a particular monitored ESR signal due to the induc-
tion of a nuclear transition of nuclei coupled to the paramagnetic centre. This method is used to improve the
resolution of ESR spectrum. The resolution and specificity increase are because of smaller number of ENDOR
transitions and selectivity of ESR line. This method is used when hyperfine lines are not well resolved in the ESR
spectrum and when the identity of the interacting nucleus is to be established by measuring the nuclear g-factor,
1 1
i.e. gN. A spin S = system coupled to n inequivalent nuclei of nuclear spin I = yields only 2n lines while an
2 2 ⎛ 1⎞
ESR spectrum would give 2n lines, A schematic diagram of ENDOR transitions for an electron ⎜ S = ⎟ coupled
⎛ 1⎞ ⎝ 2⎠
to a nucleus ⎜ I = ⎟ is shown in Fig. 9.31. Remember that the down-spin (b ) of an electron is more stable than
⎝ 2⎠
the up-spin (α) because the electron is negatively charged 4 (αβ)
NMR (a)
whereas the reverse is true for nucleus because of its positive 3 (αα)
charge. Because of the mutual interaction of energy states,
f
the states with paired spins (αb, bα) tend to be stabilised or f
lowered in energy while parallel spins (αα, bb) are raised ESR
in energy. Thus, two ENDOR lines would appear at the two (a)
ESR
nuclear transitions, say vn1 and vn2, as shown in Fig. 9.32. (a)
Further, when the unpaired electron interacts with more than
one proton or with more nuclear spin systems, the ENDOR
lines appear at frequencies corresponding to the hyperfine 2 (ββ) NMR (a)
coupling constants, no matter how many equivalent spins are 1 (αβ)
present in each spin system. ΔE (1–2) > ΔE (3–4), ΔE (1–3) > ΔE (2–4)
Fig. 9.31 Schematic representation of ENDOR transitions
In ENDOR method, the radio frequency (rf) generator
for an electron (S = 1/2) coupled to a nucleus (I = 1/2):
having power (200 W) and NMR radio frequency (a) Allowed transitions (f) Forbidden transitions.
(2−30 MHz), and large power output is connected to
ferrite-core transformers to the side of the coils of the spe-
cial designed cavity. The sample is subjected simultane- A
ously to a microwave frequency (MW) suitable for electron
resonance and a radio frequency suitable for nuclear reso-
signal amplitude
Change in ESR
nance. The radio frequency is slowly altered while observing

the one point of ESR spectrum under conditions of micro-
wave saturation or partial saturation. A change occurs in the
spin population of the electronic levels at the values cor- νn1 ν0 νn2
νrf
responding to the resonance of a given type of nuclei. This
change in population alters the amplitude of the original Fig. 9.32 ENDOR lines for a system with S = 1/2 and I = 1/2.
ESR signal. The magnitude of the difference in amplitudes
is measured as a function of radio frequency. This process is known as ENDOR. The ENDOR displays an ESR
signal amplitude as a function of the swept nuclear radio frequency. At two frequencies of the rf generator (v1
and v2), two separate peaks will be observed (Fig. 9.32). The difference v2−v1 is numerically equal to the hyper-
fine coupling constants (A). For systems with short nuclear relaxation times, such as free radicals in solution,
or with small nuclear magnetic moments, large rf fields are required. However, for systems with long relaxation
times, low power (5W) rf generator would suffice. The pulsed ENDOR sequences due to Davies and Mims have
been displayed here.
9.16 ELECTRON DOUBLE RESONANCE (ELDOR)

ELDOR is a method to study the changes in the amplitude of an ESR signal of the sample at some point of the
spectrum induced by the second microwave resonance frequency. It leads to higher resolution of hyperfine struc-
ture. This method is used to study relaxation processes including chemical and spin exchange and to separate
overlapping multiradical spectra. In this method, the sample is subjected simultaneously to two microwave fre-
quencies. One of these is used to observe an ESR signal at some point of the spectrum, while the other is swept
through other parts of the spectrum to display the amplitude of ESR signal as a function of the difference of the
two microwave frequencies.
9.17 APPLICATIONS OF ESR SPECTROSCOPY

Like all other spectroscopic tools, viz., NMR, IR, UV/V, MW, etc the applications of ESR spectroscopy are based
on the four properties of ESR lines, i.e. their intensity, width, position (g-factor) and multiplet structure and may
be broadly divided into two categories. The first of these exploits the specific features of ESR spectrum to obtain
structural and physical information characteristic of the paramagnetic sample under investigation. The second
approach exploits the ability of ESR to provide quantitative information on the concentration of a paramagnetic
centre available in the sample.
9.17.1 Determination of Concentration of Free Radicals

The concentration of free radicals or number of spins in the sample (per g, per mL or per mm of the sample) is
proportional to the total area under the absorption curve. The area is obtained by integrating the derivative signal.
The integration is made graphically, electronically or digitally with a computer. The accuracy of all the methods
decreases when the signals have long wings or signal-to-noise ratio is more and the baseline is affected by a back-
ground signal. These problems are overcome by using a digital computer.
In the graphical method, the spectrum is divided into constant intervals and the amplitude is measured in each
interval. These intervals are cumulatively added and each partial sum is obtained. The last value of the partial sum
should be zero. If it is nonzero, it is divided by the number of intervals, and the parameter so obtained is cumu-
latively subtracted from the partial sums. The second integral is then obtained by adding all the partial sums. The
accuracy of the graphical method depends on the number of intervals. Let us demonstrate this method by dividing
the first derivative absorption curve into eight equal intervals as shown in Fig. 9.33. If suffixed, (x) and A denote
the interval and the amplitude respectively. We may write
x0 = A0 (= 0)
x1 = A0 + A1
x2 = A0 + A1 +A2
x3 = A0 + A1 +A2 + A3 (9.73)
Davies
π π/2 π
t τ τ
MW
π
rf
Mims
π/2 π/2 π/2
τ t τ
MW
π
rf
Fig. 9.33 Pulsed EN DOR sequences due to Davies and Mims.

x4 = A0 + A1 +A2 + A3 + A4 A
…
x8 = A0 + A1 +A2 + A3 +A4 + A5 +A6 + A7 +A8
If the value of the last interval is zero, i.e. x8 = A0 + A1 + A2 + A3 + ... + X
5 6 7 8
A8 = 0, then the value of integral, i.e. the area under the absorption 0 1 2 3 4
curve
i =8
= ∑ xi where i = 0, 1, 2,..., 8. (9.74)
i =0 Fig. 9.34 Representation of graphical method
to determine the area under the curve; X stands
If x8 ≠ 0 = λ (say), for the interval and A stands for the amplitude.
λ
then, Error = =ϕ (9.75)
8 ( no. of intervals)
Applying the correction, we write
x0 A 0 ( = 0) − ϕ
x1 A 0 + A1 − ϕ
x2 A 0 + A1 A 2 − ϕ
x3 A 0 + A1 A 3 − ϕ (9.76)
x8 A 0 + A1 + A 2 + A 3 A 4 + A 5 A6 + A7 A8 −ϕ
Thus x 8′ = 0 and the area under the curve
i =8
= ∑ x i′ where i = 0, 1, 2, 3,..., 8. (9.77)
i =0
Since many experimental factors such as (i) the overall spectrometer gain, (ii) the microwave frequency, (iii) the
g-factor of the sample, (iv) the sample temperature, (v) the transition probability, (vi) the concentration of free
radicals in the sample, (vii) the filling factor of the sample, and (viii) the modulation, amplitude at the sample,
affect the intensity of the ESR signal, a number of corrections need be applied for accurate measurements of the
concentration of free radicals in the sample.
The most common method used for the determination of concentration of free radicals in an unknown sample
is to compare its ESR signal with that of some standard sample containing a known amount of free radicals. The
standard sample should have the following properties: (i) line width and line shape of its ESR spectrum should be
identical to that of the unknown sample, (ii) number of spins, the line width, and the g-value should be constant,
i.e. independent of temperature and time, (iii) concentration should be similar to that of the unknown sample,
(iv) the s-l relaxation time should be short otherwise saturation of the signal will take place easily, and (v) physical
shape and dielectric loss of the standard sample should be similar to that of the unknown sample.
Vanadium over the range 0.1 to 50 mg/mL can be determined in petroleum products. Mn (II) ion in aqueous
solutions can be determined. The sensitivity of determination lies in the range 10−6 M−0.1 M. Cu (II), Cr (III),
Fe (III), Ti (III) and Gd (III) ions can also be determined quantitatively by ESR method. Polynuclear hydrocar-
bons such as anthracene, perylene and naphthalene have been determined after conversion to radical cations and
adsorption on to the surface of an activated silica–alumina catalyst.
9.17.2 Study of Electron Transfer Reactions

The ESR technique is best suited for the study of free radical reactions. The appearance of an ESR signal during
the course of a reaction indicates that the reaction proceeds via some free radical mechanism. However, the low
concentration of free radicals formed at one or another stage of a reaction due to their short lifetimes pose a
problem for their detection by ESR measurements. This problem has been overcome with the help of some spe-
cially designed cells. One such cell is due to Dixon and Norman (Fig. 9.35). It consists of two tubes with a flat
cell on the top. The inner tube contains a spray head, from which one solution (e.g. the reductant) is injected into
the second solution (e.g. the oxidant). The mixed solution then rapidly flows through the cavity under hydraulic
pressure. The rate of flow is kept faster than the rate of production of radicals. A steady concentration of the radi-
cals can be attained with proper adjustment of pressure. Free radicals with lifetimes of the order of 10−2 s can be
Detection ESR
cavity
Mixing part
Flow 1
Flow 2 10 G
Fig. 9.35 Rapid flow apparatus/Dixon-Norman Fig. 9.36 ESR spectrum of isobutylvinyl ether-iodine system
type mixing cell for ESR studies of free radicals. measured at 77 K.
detected by this technique. Figure 9.36 shows the ESR spectrum 2 I2 I+ • I−3
of radical cation generated by mixing iodine solution with isobu- I+ • I− + H C CH (Complex)
3 2
tylvinyl ether at 77 K, The radical cation is formed according to
the following scheme. The radical cation (a) is stabilised by the
OR
isobutoxy group. The observed ge value of 2.0057 is larger than that +
for a free electron value of 2.0023. This may be attributed to con- R = CH2CH(CH3)2 I•+ H2C• CH3 I3−
siderable spin density on oxygen due to conjugation, which makes

the radical cation reluctant to recombine. Unstable aliphatic and OR
(a)
aromatic free radicals formed in liquid phase reactions are detected
by this method. The ESR spectrum of TiCl3 (acidified solution)—CH3OH + H2O2 (acidified solution) system when
recorded by the rapid flow•
method exhibits a three-line multiplet with intensity ratio 1:2:1,
The triplet is due to C H 2 OH radical formed according to the following scheme:
•
Ti3+ + H 2 O 2 → Ti 4+ + O H + OH −
Ti3+ + O H → Ti 4+ + OH −
• •
O H + H 2O → H 2O + O2 H
• •
CH 3OH + O H → C H 2 OH + H 2 O
When aqueous solution of TiCl3 and (CH3)3COOH at pH < 2 are mixed in the resonance cavity of the ESR spec-
trometer,
•
a quartet spectrum of narrow lines with a separation of about 23G has been observed and is attributed
to C H 3 radical produced in a two-step reaction.
Ti3+ + (CH 3 )3 COOH = Ti 4+ + OH − + (CH 3 )3 C O

•
(CH3 )3 C- O
• •
3 -CO CH 3 + C H 3
-CO-CH
9.17.3 Determination of Rotational Correlation Time

The stable nitroxide radicals has tertiary carbon bonded to the nitrogen, e.g. di-tert-butylnitroxide (a), 2,2-
dimethyl-oxazoline derivatives, (b) pyrroline, (c) piperidine, and (d) nitroxides. They are used as both paramag-
netic markers and reactants.
The ESR spectra of these radicals consists of (2 × 1 + 1) = 3 lines of equal intensities. These radicals in solution
(organic solvents, e.g. C6H6) of low viscosity are in rapid Brownian motion. In viscous medium, the rotational
motion is hindered and tumbling rate of molecules decreases. Consequently, the broadening of spectra occurs and
hyperfine coupling constants increase (Fig. 9.37). The rotational motion of these radicals in solution is charac-
terised by a parameter called the ‘rotational correlation time (t). It is defined as the time required for complete
rotation of a nitroxide radical about its axis and is 10−11 < τ < 10−6 s. It may be defined as the time required
for the nitroxide radical to rotate through an angle of one O

CH3
radian. It can be calculated from one of the following (CH3)3C C(CH3)3
equations: N N CH3
⎡ 1 1
⎤ O O
⎛I 0⎞ ⎛I 0⎞ 2 2 • •
τ = ΔH PP ⎢⎢⎜ PP ⎟ − ⎜ PP ⎟ ⎥C
⎥
(9.78) (a) (b)
⎝ I +1 ⎠ ⎝ I PP −1 ⎠
⎢⎣ PP ⎥⎦ R′
⎡ 1 1
⎤ R
⎛ I 0 ⎞
2 ⎛ I 0 ⎞
2
τ = ΔH PP ⎢⎢⎜ PP ⎟ + ⎜ PP ⎟ ⎥⎥ − 2C ′ (9.79) H3C CH3 H3C CH3
⎝ I PP+1 ⎠ ⎝ I PP −1 ⎠
⎢⎣ ⎥⎦ H3C N CH3 H3C N CH3
where ΔHPP is peak-to-peak width, I PP 1 , I PP0 , I PP+1 are O
• •O
the measured peak-to-peak amplitudes of three lines of (c) (d)
ESR spectrum of nitroxide radical (Fig. 9.38). C and C'
R⬘ = O, C6H5COO—, —NH2, –OH, CH3COO−
are the characteristic parameters for the nitroxide and the
particular experimental conditions employed. Because of
the rotational behaviour of organic nitroxide radicals in solution of low viscosity, they have been used as spin
probes to study their interactions with host polymers in which they are dissolved.
2A||
g|| ΔHpp
77 K
g⊥
A⊥
142 K Ipp0 Ipp +1

Ipp −1
293 K
20 G H
Fig. 9.37 ESR spectrum of di-tert-butylnitroxide Fig. 9.38 Notation of the symbols used for the determination of the
radical at 293, 142 and 77 K. rotational correlation time (t) from ESR spectrum.
Some of the radicals easily diffuse into various polymers and they have been found to reside in the amorphous
phase of the solid. Since the glass transition temperature (Tg) corresponds to the onset of liquidlike transla-
tional motions of the long segments of molecules in amorphous phase, and the rotations of nitroxyl radicals
are sensitive to changes of geometry of their surroundings, the rota-
tional correlation times of spin probe radicals in polymers at tempera- 143 K
tures near Tg can be measured, e.g. the well-resolved triplet line ESR
spectra of polyethylene (at different temperatures, 90−333 K) and poly
(vinylchloride) (93−423 K) above the glass transition temperature (Tg)
containing 2,2,6,6-tetramethyl-4-hydroxy-piperidine-1-oxyl benzo-
ate shows rotational motion of radicals in the amorphous part of the 223 K
polymers. The broad line shape of the spectra in the low-temperature
range indicates that the polymer matrix is in the rigid state. When the
temperature is raised, a small splitting takes place because a part of the
chain in the probe begins to rotate to cause change in the hyperfine pat-
tern. ESR spectra of 2,2,6,6-tetramethyl-4-hydroxy-piperidine 1-oxyl 333 K
obtained as grown polyethylene crystals at 348.5 K in the temperature
range, 143−333 K also show similar behaviour (Fig. 9.39). From these
spectra, the rotational correlation time of free radical can be computed. 50 G
The rotational correlation time (t) obeys the Arrhenius equation, Fig. 9.39 ESR spectra of 2,2,6,6-4 tetramethyl−
4−hydroxy-piperidine-1-oxyl obtained as grown
polyethylene crystals at 348.5 K in the tempera-
τ τ 0e Ea / RT valid for T >Tg (9.80) ture range 143−333 K.
Table 9.10 Rotational activation energy of some spin-probe nitroxide radicals in linear polymers.
Polymer Piperidine Nitroxide with R Ea kcal mole−1

Polyethylene (0.918 g/cm3) H 10.7
Polyethylene (0.918 g/cm )
3
C6H5COO 11.0
3
H 10.4
3
C6H5COO 10.0
Polystyrene H 18.2
Polystyrene C6H5COO 18.3
Ea is the rotational activation energy of the free radical. T is the absolute temperature (K) and R is gas constant.
1
Ea can be calculated from the slope (Ea/R) of the straight-line plot of log t against . The rotational activation
energies of some spin probe radicals in linear polymers are listed in Table 9.10. T
The rotational activation energy of free radicals is independent of the size of the radical but dependent only on
the properties of the polymer matrix. Thus, the radical rotational frequency is determined by the mobility of the
polymer segments.
The rotational correlational time (t) depends on molecular volume (Vm) and obeys the relation
= A e − kV g
τ = BE − kV1 (9.81)
where Vg and V1, are the molar volumes (cm3 mole−1) when T < Tg and T > Tg respectively, k ≈ 1, A and B are con-
stants.
The interaction of polymers with plasticisers can be measured by using nitroxide radicals as spin probes. The
rotational correlation time (τ) of probe radicals in plasticised polyvinylchloride at 298 K, decreases strongly when
the amount of plasticiser is increased.
9.17.4 Study of Ion Pairs

ESR spectrum of alkali salts of naphthalenide ion in tetrahydropyran shows that the hyperfine lines of naphtha-
lenide are split into quadruplets by the presence of Na23. This indicates the presence of sodium naphthalenide ion
pair, i.e. NNa + in tetrahydropyran. The character of the ion pair depends on the solvent and temperature of the
solution. The magnitude of splitting (aNa) varies from about 1.4 G at 320 K to 1.0 G at 236 K, when a small amount
of tetra-adenate ether (tetraglyme) (~0.24 mole lit−1) is added, the splitting constant aNa decreases. The splitting is
about 0.4 G at 300 K. On lowering the temperature, aNa decreases further and at 236 K, the four lines merge into
one. Thus, the reduction in aNa on addition of tetraglyme to solution of alkali salt of naphthalenide ion in tetrahy-
drofuran is attributed to the formation of glynated pair, N(G)Na + . On the other hand, a comparison of ESR spectra
of alkali salt of naphthalenide ion in tetrahydrofuran at 300 K in the absence of tetraglyme [Fig. 9.40(a)] and in
the presence of intermediate concentrations of glyme [Fig. 9.40(b)] shows that the two sets of lines in the latter
spectrum are characterstic of the tight ion pair NNa + and glymated pair N(G)Na + , respectively. The lines for the
glymated pair are marked in Fig. 9.40(b) by arrows. The equilibrium in this system is presented as
(a) (b)
1G
Fig. 9.40 Low field wing of ESR spectrum of NNa+ in tetrahydrofuran at 300 K: (a) In the absence of tetraglyme, and
(b) in the presence of 0.006 mole lit−1 tetraglyme.
k1
NNa + + G (G)Na +
N(G)N
k2
From the width of the lines in Fig. 9.40(b), the values of k1 and k2 have been found to be ~108(mole/lit)−1−s−1
and 106 (mole/lit)−1 s−1, respectively.
9.17.5 Kinetics of Electron-Exchange Reactions

The electron-exchange reactions between radical ion N and the corresponding neutral molecule (N) in the solu-
tion phase are studied by the rapid-flow technique, e.g. when naphthalene is added to the solution of naphthalenide
radical ion an electron-exchange reaction occurs N N N+N which causes a broadening of the hyperfine
components of the ESR line. The additional width of resonance line ΔHl/2(G), i.e. the difference between the line
widths of a particular line in the absence and presence of electron transfer, is given by the equation
1
Δ τ=
1/ 2 (9.82)
2π
where t is the mean jump time for an electron. If M is the molarity of naphthalene, the pseudo first-order rate
constant (k) may be expressed as
1
k [M ] = = 2π ΔH 1/ 2 (9.83)
τ
It is not required to know the concentration of the radical as long as its concentration is kept low. The rate constant
of electron transfer can be determined at slow and at fast exchange limits from the plots between ΔH1/2(G) and
8
7
0.4
6
0.3 5
ΔH1/2 (G)
ΔH1/2 (G)
4
0.2
3
0.1 2
1
0
0 2 4 6 8 10 12
1 2 3 4 5 6 7 8
[M] × 103 (mole)
1/[M]
Fig. 9.41 Dependence of ESR line width ΔH1/2 on naphthalene
Fig. 9.42 Dependence of ESR line width ΔH1/2 on 1 in the fast
concentration [M] in the slow exchange N + N N + N in [M]
M
hexamethylphosphoroamide. exchange N + N N + N in hexamethylphosphoroamide.
high concentration (Fig. 9.41) and low concentration (Fig. 9.42) of naphthalene in hexamethylphosphoroamide
solvent, respectively.
Similarly, the rate of electron transfer reaction between a tetracyanoethylene anion radical ( TCNE ) and its
neutral molecule (TCNE) can be studied. TCN E is formed by the reduction of TCNE with alkali metal in ether.
TCN E exhibits (2 × 4 + 1) = 9 lines ESR spectrum. As already discussed, the broadening of lines occurs. When
the concentration of TCNE is increased and at very high concentration of TCNE the electron jumps from one mol-
ecule to another so fast that the hyperfine structure is lost and only one sharp line in the spectrum is observed. From
the addition line width, the rate constant for the electron transfer reaction, TCN E + TCNE TCNE + TCN E is
determined from the equation
1 5 × 10 7 ⋅ ΔH
H 1/ 2
k= (9.84)
[M ]
Here, [M] is the concentration of neutral species added, i.e. naphthalene.
When the solution of hydroquinone is mixed with oxygen in the observation area of the ESR spectrometer by
the rapid-flow method, an ESR spectrum consisting of five lines multiplet with intensiy ratio of 1:4:6:4:1 results.
The multiplet is attributed to the formation of semibenzoquinone since the five-line pattern corresponds to the
magnetic interaction between odd electron and four protons of the ring. Results of kinetic study of the formation
and decay of semiquinone according to the scheme indicate that the half life of the formation reaction is about
0.15 second.
1 1
O2 •
O2
2 2
HO(C6 H 4 )OH HO(C6 H 4 ) O O(C6 H 4 )O
Semibenzoquinone
m
v = 9500 MHz
H = 3400 G
9.17.6 Determination of Exchange Integral

In systems with two unpaired electrons, one electron located on each of the two atoms, the interaction between
the two electrons gives rise to a spin singlet state (S = 0) and spin triplet state (S = 1) separated by the exchange
integral J (Fig. 9.43), e.g. dimeric copper (II) acetatemonohydrate, dimeric oxavanadium (IV) tartrates. The
population in the two spin states depends on the magnitude of J and temperature. In a ferromagnetic complex
(J > 0), the paramagnetic triplet state (S = 1) is the ground state whereas the diamagnetic singlet state (S = 0) is
the excited state. With decreasing temperature, the population in the triplet state grows at the cost of the singlet
state; consequently, the intensity of the ESR line grows. On the other hand, in an anti-ferromagnetic complex
(J < 0), with the decreasing temperature, the population in the ground diamagnetic singlet state grows at the
cost of the excited paramagnetic triplet state. As a result, the intensity of the ESR line will decrease. Thus, the
magnitude of J can be determined by measuring the intensity of ESR lines at different temperatures. The relative
intensity (IR) of ESR signal is related to J by the expression
⎡ −J ⎤
exp ⎢ ⎥
IR =
1 ⎣κT ⎦ (9.85)
T ⎡ −J ⎤
1 + 3 exp ⎢ ⎥
⎣κT ⎦
The intensity is measured relative to the signal from a small amount of DPPH mixed with the sample.
The appearance of an ESR line in the spectrum of a sample corresponding to ΔMS = +2 transition, indicates the
presence of the triplet state in the complex involved in the magnetic exchange interaction. It is to be noted that
a weak magnetic exchange interaction (J or −J ≤ 30 cm−1) which cannot be detected by magnetic susceptibility
measurements at room temperature or even single crystal, x-ray studies, can be detected through ESR measure-
ments. It is possible to detect the presence of magnetic exchange in the magnetically concentrated coordination
compounds in which the single crystal x-ray studies have shown the M−M distance to be greater than 4 Å.
S=1
S=0 (Triplet State)
(Singlet State)
S=1 S=0
(Triplet State) (Singlet State)
Ferromagnetic Antiferromagnetic
(J > 0) (J < 0)
Fig. 9.43 Coupling of two S = 1/2 spins of two interacting metal ions.
9.17.7 Determination of Unpaired Electron Spin Density

and the Molecular Shape of Free Radical
The ESR technique, like NMR, does not give information about the arrangement of functional groups in the mol-
ecule. However, it allows the determination of the distribution of unpaired electron spin density over the atoms in
the molecule and in some cases the shape of the free radicals.
The main parameter which influences the magnitude of electron-nucleus interaction is the amount of time
which the electron spends in the neighbourhood of the coupled nucleus or in other words the electron spin density
at the nucleus.
Quantum mechanical considerations on the electron nucleus interaction show that
8π
g e βe g N β N ψ ( 0 )
2
a (9.86)
3
3 2
θ −1
and τ β g N βN 3
(9.87)
r
1 Z3
= where the symbols have their conventional meanings, Z is atomic number, a0, (Bohr
r3 3 3 ⎛ 1⎞
n a0 l l + ⎟ (l + 1)
⎝ 2⎠
radius for hydrogen atom) = 0.53 Å, n, the principle quantum number, and l, the orbital quantum number. The
importance of these isotropic-a and anisotropic-t, coupling constants results from the fact that these parameters
are directly related to the electronic configuration of the radical.
As already discussed, Eq. (9.86) represents the Fermi interaction. |ψ(0)|2 is the probability of finding the electron
in contact with the nucleus giving rise to isotropic coupling. |ψ(0)|2 ≠ 0, when the electron has some s character
C S2 . Equation (9.87) represents the dipolar interaction. The s orbital makes no contribution to this anisotropic
interaction, because in this case the spherical symmetry results in an averaging of (3cos2 q − 1) to zero. If the
unpaired electron is in a pz orbital then
4 1
τ βe g N β N 3 = 2 B 0 ( ) (9.88)
5 r
−2 1
and τx τy = ge β βN = −B0 ( ) (9.89)
5 r3
1
Hartree Fock calculations have made available the theoretical values of |ψ(0)|2 and for elements of the
r3
Periodic Table. Electron spin density and isotropic coupling constant of some representative elements are listed
in Table 9.11.
Table 9.11 Isotropic hyperfine coupling constant (aN ) and the density of valence s orbital
of atom N evaluated at the nucleus |yN(0)|2.
Nucleus aN (G) |ψN(0)|2 (au)−3

H
1
539.86 0.338
13
C 820.10 2.042
14
N 379.34 3.292
17
O −888.68 41.082
19
F 44829.20 29.840
Thus, it is possible to calculate the isotropic coupling constant Aiso and anisotropic coupling constants, 2B0,
−B0, −B0, associated respectively with the ns and np orbital of each atom. The principal aim in the analysis of ESR
spectra of solid samples is to obtain the experimental parameters which make it possible, after comparison with
a τ
the atomic constants, to estimate the s character, C S2 = , and the p character, C P2 = z of the atomic orbital
A iso 2B 0
containing the unpaired electron. When the values of C S2 C P2 are determined, one can calculate the hybridiza-
⎛C2 ⎞
tion ratio λ 2 = ⎜ P2 ⎟ , and thus estimate certain structural parameters, e.g. if C3v symmetry is assumed, it can be
⎝ CS ⎠
•
shown that for R3 A radical
⎛ 1.5 1⎞
ϕ = cos −1 ⎜ 2 − ⎟ (9.90)
⎝ 2λ + 3 2 ⎠
1
and cos 2 (1 + 2 cos ϕ )
(1 (9.91)
3
where, j is the interbond angle RAR and q is the angle between the symmetry axis and an AR bond.
It is apparent that the isotropic coupling constant can be obtained from both liquid phase (direct measurement
of aiso) and solid state
(ax + τ x ) + (ay + τ y ) + (az + τ z )
aiso =
3
since for dipolar interaction tx + ty + tz = 0] ESR studies, In contrast, the anisotropic components are available
only from solid-phase measurements.
Let us now examine how these concepts could be applied to determine the distribution of unpaired electron
density in the entire molecule. For a free atom, the dependence of aN on rN is expressed as
aN R N ρN (9.92)
where, aN is the hyperfine constant in an atom N, rN is the fraction of unpaired electrons in the valence state, RN
is the hyperfine coupling constant for a single electron in the valence state. For hydrogen atom RN = 506 G (Calc:
539.86), rN = 1, since the electron spends its time in 1s orbital only. Thus, for hydrogen atom aN = RN.
•
(a) H3 Radical The ESR spectrum of trimethyl radical in solution exhibits a quartet with intensity ratio of
1
1:3:3:1 due to splitting by the protons (I = , natural abundance 99.985 per cent). Splitting due to carbon is not
2 ⎛ 1⎞
readily observed, as principal isotope C (natural abundance 98.9 per cent) is nonmagnetic and 13C ⎜ I = ⎟ has
12
•
⎝ 2⎠
a natural abundance of only 1.1 per cent. The observed value of the C H 3 hyperfine coupling constant is −23 G.
• −23
Applying Eq. (9.92) to C H 3 radical with aH = −23 G, we get ρH = = −0.045. It follows that the electron den-
506 •
sity at the carbon atom is (1 − 3 × 0.045) = 0.865. Thus, we know the electron density at each atom of the C H 3
radical: each hydrogen, −0.045; carbon, 0.865. This implies that the unpaired electron spends 4.5 per cent of its
time in the 1s orbital of each hydrogen atom and 86.5 per cent in the vicinity of the carbon atom.
(b) p-spin Density In the planar p-type hydrocarbon radicals there is zero overlap of the s orbitals of the
hydrogen and carbon atoms with the p-orbitals (pz) perpendicular to the molecular plane. The s atomic orbitals (AOs)
cannot contribute at all to the half-filled molecular orbital (MO), which is composed of pz AOs only. Although no
unpaired electron spin density can enter the AOs directly, the spin polarisation leads to a small spin density in both
the carbon and the hydrogens AO’s, the former being of the same sign as that of the unpaired electron (α or positive
spin) and the latter being ‘opposite in sign (b, or negative spin). Furthermore, the magnitude of spin density on the
ring proton is roughly 5 per cent as large as the pz spin density on the adjacent ring carbon. The splitting constant aH,m
of proton on ring carbon m is related to the p-spin density rm on carbon m by the McConnell relation
H , μ = Qρμ (9.93)
Here, Q or (QCH) is the proportionality constant, also called the spin polarisation parameter and roughly con-
stant for p-radicals and is usually taken to be about −27 G (or −25 G (or −25G)), approximately 5 per cent of the
coupling constant of hydrogen atom (540 G) but of opposite sign. The constant Q is the proton hyperfine splitting
constant per unit spin density on the sp2 carbon containing CH bond. The value of Q has also been estimated from
the spectrum of benzene negative ion which consists of seven peaks with ‘a’ value of 3.75 G. The probability of
the unpaired electron being at any one carbon is 1/6, so from Eq. (9.93), we obtain
|Q| = 3.75 × 6 = 22.5 G.
The value of Q = −22.5 G being based on experimental data is preferred over the value of Q = −27 G from theo-
retical considerations.
• •
(c) Shape of C H3 Radical The geometry of C H 3 radical could either be planar (C atom is sp2 hybridised) or
tetrahedral (C atom is sp3 hybridised). We know that for planar hydrocarbon p-radicals
a ,μ Q ρμ (Eq. 9.89)
•
The proton hyperfine coupling constant for C H 3 is −23 G. Using aH = −23 G and rm = 1, Q •= −23 G which is
close to the value of Q for most of the aromatic radicals, i.e. QCH = −22.5 G. This implies that C H 3 radical has •
a
planar structure. However, the observed splitting could not be fixed as the sole criterion for the planarity of C H 3
radical
•
since it is not possible to estimate the proton splitting coupling from theoretical calculations
•
for non planar
C H 3 . On the other hand, the 13C coupling constant can be estimated in planar and tetrahedral 13 C H 3 radical. The
•
13
observed and calculated isotropic coupling constants for 13C and C H 3 have been found to be (+) 38.34 and 45
G, respectively. For methyl radical to be tetrahedral, the unpaired electron should reside in one of the four sp3
hybrid orbitals each of which has 25 per cent s character, Thus, the interaction of unpaired electron with carbon
nucleus should give rise to coupling constant of (75 x 4) = 300 G. The very small value of observed 13C splitting
constant suggests that the unpaired electron has a very small amount of s character and it resides mostly in pp-type
•
orbital, which implies that C H 3 has planar structure.

•
(d) p-electron Density and Molecular Conformation of Triphenylmethyl Radical; (C6H5)C The ESR spec-
1
trum of (C6H5)3 C radical in solution exhibits hyperfine splitting due to ring protons (I = , natural abundance
•
2
99.985 per cent). Splitting due to carbon atom which is essential for the estimation of p-electron density and
determination of conformation of (C6H5)3 C is not observed since 12C is nonmagnetic (I = 0, natural abundance
•
⎛ 1⎞
98.9 per cent) and 13C ⎜ I = ⎟ has natural abundance of only 1.1 per cent. On the other hand, the radical tris-
⎝ 2⎠
•
(3.5-di-t-butylphenyl)-methyl with isotopic enrichment at C−1 has a simpler spectrum than (C6H5)3 C and gives
13
C-splitting constants for all the five types of positions. Thus, it is used as a model radical to obtain p-electron
•
spin density and molecular conformation of (C6H5)3 C . The proton and 13C splitting constant, aH and a13 C for the
model radical have been listed in Table 9.12. The signs of the aH and a13C values have not been determined experi-
mentally, but are assumed on the basis of MO calculations.
Recalling McConnell equation
a ,μ Q ρμ (Eq. 9.93)
•
and using Q = −22.5 G, we obtain rortho = +0.114 and rpara = + 0.124 for radical (C6H5)3 C . Spin densities at C−1,
C−2, C−2′ and C−2′′, cannot be estimated by this method since these positions do not have attached proton. MO
calculations reveal that there is induced negative p-electron spin density at the metacarbons, resulting in a positive
aH (meta) and a negative a13C (meta). Because the phenyl rings are twisted, Eq. (9.93) is not applicable. However,
rmeta can be calculated from the modified McConnell relation of the form
a 3,μ
Q CH ρμ (9.94)
where, aCH3 is the methyl proton splitting of m-tolyldiphenyl methyl radical, aCH3 = −0.84 G and Q CH3 = +29 G.
−0 84
Thus, ρ meta = = −0.029.
+29
Further, the total spin density must be unity,
ρ + 3ρ 2 6 ρ + 6 ρ meta 3ρpara = 1
Substituting the values of rortho, rmeta and rpara we get,

rl + 3r2 = 1 − 0.882 = 0.118 (9.95)
The 13C coupling constants a13C,i of a planar p-radical are related to the p-electron spin density ri of carbon i and
on the neighbouring carbons j by the Karplus−Fraenkel expression
a13C,i = Q C ρ i + ∑ Q C C ρj (9.96)
j
Table 9.12 The proton and l3C splitting constants —(the a13C values are those measured for
the model radical tris−(3, 5-di-t-butylphenyl)-methyl.
Position aH(G) a13C (G)
1 — +23.5
2 — −11.3
ortho −2.57 +6.5
meta +1.13 −3.2
para −2.81 +4.1
where, QC and Qc′c are proportionality constants. QC = 30.5 and QC′C = −13.9 G. For C−2 of triphenyl methyl radi-
cal, we use the observed value of a13C = −11.3 G , ρ ortho 0.114.
Thus, Eq. (9.96) becomes
−11.3 = 30.5 r2 − 13.9 ( r1 + 2 × 0.114) (9.97)

−11.3 = 30.5 r2 − 13.9 ( r1 + 0.228)
Solving Eqs (9.95) and (9.97) for r1 and r2 we get, r1 = +0.387 and r2 = −0.089.
Thus, C−1, ortho-and para-positions have large positive n-spin densities whereas intervening C−2 and meta
positions have smaller induced negative spin densities.
The deviation from planarity can have a pronounced affect on the a13C of carbon centred radicals because the
2s AO directly contributes to the half filled MO. As the 2s spin density from orbital mixing and that from spin
polarisation are both positive, nonpolarity results in an enhanced positive a13C. The a13C values for planar (or nearly
• •
planer) C H 3 and the pyramidal C F3 are + 38.34 and + 271.6 G respectively.

•
In case of (C6H5)3 C , the angle (q′) between the C−l : C−2 bond and the plane defined by C−2, C−2′ and C−2′′
can be calculated from the relation
a13C (calc) a13C (obs) = 2400 tan 2 θ ′ (9.98)
The experimental value of a13C for C−1 of (C6H5)3 C is + 23.6 G and the pure spin polarisation value, a13C (calc)
is 30.5 G [from Eq. 9.96)]. These parameters give q′ = 3° and a hybrid orbital containing less than 1 per cent s
character. The higher spin density of +0.525 and −0.078 at C−1 and C−2 respectively, a q′ value of less than 1°,
and less than 0.1 per cent s character in the hybrid AO has also been reported in the literature. Thus, it appears on
•
the basis of ESR studies of (C6H5)3 C in solution that C−1 is sp2-hybridised and is virtually coplanar with its three
nearest neighbours, i.e. C2, C′2 and C″2 .
The angle of twist ϕ′, of the phenyl rings, i.e. the angle between the 2p orbitals of the ring and central carbon
2p orbital can be obtained from the observed and calculated values a13C (ortho). Both these parameters are related
to each other through Eq. (9.99).
a C( h ) ρ( ) i 2ϕ ′
Q sin (9.99)
Use of a13C (ortho) = 6.5 G, Q = 355 G and r(ortho) = +0.095 yields ϕ′ = 26°.
The hybridisation of the central carbon atom C−1 as well as the spin density on it can also be determined
directly from the anisotropic components of the hyperfine splitting constants measured from the ESR spectrum
of the radical in the solid state. ESR spectrum of a single crystal of (C6H5)3 C gives a13C = 25 G and t⎥⎥ =35 G for
•
the central carbon atom. The calculated values of the isotropic coupling constant, Aiso and the anisotropic cou-
pling constant 2B0 associated respectively with 2s and 2p orbitals of central carbon atom C−1 are Aiso = 1100 G
and 2B0 = 65 G. These data yields CS2 = 0 02 and C2P = 0.53. The high value of C2P in combination with a near
zero value for CS2 indicates that the radical centre C−1 is planar. The small CS2 value is due to spin polarisation
mechanism.
In conclusion, complementary chemical information can be obtained from the solution spectra and from
single-crystal ESR spectra.
9.17.8 Absolute Dating

ESR measures the electrons captured in bone or shell sample up to two million years. ESR testing on human tooth
enamel dates some of the earliest anatomically modern humans to about 100,000 (=105) years.

ESR was first introduced into biology as a tool to probe the participation of free radicals in oxidation–reduction
reactions and is a major tool in the characterisation of metallo proteins. Iron is the element which provides by
far and wide the richest variety of ESR spectra with diagnostic characteristics established by high spin, low
Table 9.13 ESR parameters of metal ions in biochemical systems.

Metal Ion Electronic Spin Biological g-values Nuclear Temperature
Configuration State System Spin (K)
Fe(III) 3d5 1/2 Cytochrome 3.8−0.5 0 <100

Fe(III) 3d5 5/2 Hemoproteins 8−1.8 0 <100
F(III) 3d5 5/2 Non-heme iron 10−0.5 0 <100
Fe(III), 3d5, 3d6 1/2 Spinach 1.7−2.1 0 <50
F(II) pair Ferredoxin
3Fe(III), (3d5)3 3d6 1/2 HiPiP 2−2.2 0 < 50
Fe(II) tetrad
Fe(III), 3d5(3d6)3 1/2 Bacterial 1.7−2.1 0 <50
3Fc(II) tetrad Ferredoxin
Mn(II) 3d5 5/2 Concanavalin A 2−6 5/2 <300
Co(II) 3d7 1/2 Cobalamin (B12r) 2.0−2.3 7/2 < 120

Cobalt substituted
Co(II) 3d7 3/2 1.8−6 7/2 <40
proteins
Cu(II) 3d9 1/2 Plastocyanin 2−2.4 3/2 <100
Mo(V) 4d1
1/2 Xanthine oxidase 1.95−2.0 5/2 <100
The ranges of g-values shown encompass both g-anisotropy in a given system and the variation among proteins.’
spin and polynuclear iron species. The unique contributions of ESR to the chemistry of iron sulphur proteins
are an example of the power of the method. The technique is also of major value in the characterisation of
heme proteins. The characteristic parameters of metal ions found in some biological systems have been listed
in Table 9.13.
Biological molecules like hemins, chlorophillinis, polypeptides, amino acids are studied by substitution or
introduction of different transition metal ions. By following the changes of the ESR spectrum of the metal ion in
the free and bound state, one can deduce information about the environment close to the binding site or about the
biomolecule as a whole.
ESR also provides a quantitative information on the concentration of a paramagnetic centre available in a given
biological sample, e.g. it has been used to study the kinetics of reactivity of molybedenum in xanthine oxidase,
the intensity of the ESR signal at g ~ 1.95 quantify the amount of Mo5+, present at any time during the turnover
of enzyme.
The enzyme glutamine synthetase is an oligomer which can bind 2Mn2+ per sub unit. It catalyses the reaction
glutamate NH +4 ATP glutamine + ADP + Pi
and thus one can assume that there is a site for ATP on each sub-unit. On addition of Cr3+-ATP, a paramagnetic
analog of nucleotide to the enzyme-glutamate complex, the amplitude of low-field hyperfine line due to Mn2+ (g⎥⎥
= 2.0) in the ESR spectrum of enzyme-glutamate complex recorded at 35 GHz decreases without any increase in
linewidth and without any change in the shape of the ESR spectrum. The decrease in amplitude of ESR signal with
concentration of CrATP is hyperbolic which implies that a single dissociation step is operating (Kd = 280 mM).
From the total decrease in the ESR signal amplitude, distance between two paramagnetic metal ions, i.e. Mn—Cr
has been found to be 5.2 ± 0.5 Å. The metal–metal distance in Mn−CrATP is 7.1 Å. This increase in metal-metal,
i.e. Mn−Cr distance in the absence of substrate provides an evidence for the conformational change on substrate
binding to this enzyme. The metal–metal distance in the presence of a number of substrate and substrate analogs,
have been listed in Table 9.14.
Table 9.14 Distance between Mn2+− and Cr3+− or Co2+ nucleotides for various
complexes of glutamine synthetase.
Enzyme Complex Distance (Å)

Mn−Cr ATP 7.1
Mn−Cr ATP-glutamate 5.2
+
Mn−Cr ATP-glutamate- 4 5.9
Mn−Cr ATP-glutamine 5.9
Mn−Cr ATP-methionine-SR-sulfoximine 6.8
Mn−Cr ADP 5.9
Mn−Cr ADP-Pi 5.2
Mn−Cr ADP-glutamine 6.2
Mn−Cr ADP-glutamine-Pi 4.8
Mn−Cr ADP-methionine-SR-sulfoximine 5.8
Mn−Cr ADP-methionine-SR-sulfoximine-Pi 7.6
Mn−Co-methionine-SR-sulfoximine 7.3
Mn−Co-ADP-methionine-SR-sulfoximine 6.5
Mn−Co-ADP-methionine- SR-sulfoximine-Pi 4.9
Mn−Co-ATP-methionine-SR-sulfoximine 5.2
g = 4 for Co 2+
Erythrocuprein is a copper-transporting protein. The ESR signals of oxidised bovine erythrocuprein are similar to
those obtained with human erythrocuprein, i.e. g⎢⎢ = 2.263, g⊥ = 2.062 and A⎢⎢ = 0.014 cm−1. At 7.5 pH, four hyper-
fine components are centred at the g⎢⎢ position and the superhyperfine splitting is approximated 14 G [Fig 9.44(a)].
Exposure of the protein to 6M guanidine hydrochloride causes no measurable changes in the ESR spectrum. Thus,
the detection of the number of transferable electrons is possible. Aqueous dithionite effectively reduced erythrocu-
prein. The plot of height of the g⊥ signal against dithionite concentration is straight line [Fig. 9.44(b)]. One mole of
erythrocuprein required one mole of dithionite indicating the transfer of two electrons per mole of erythrocuprein.
ESR measurements of cyanide-, cyanate-, thiocyanate- and azide-treated erythrocuprein revealed
changes in the ESR parameters (Table 9.15). Data revealed that there are at least four anion-binding
sites of different nature on the protein. These can be classified into two portions. One of these portions
strongly binds two moles of azide causing a substantial change in the ESR spectrum. These two binding
sites are thought to be the Zn2+ of the protein. The second portion of binding sites is attributed to the Cu2+
which binds the cyanide. The evidence of the proximity of Zn2+ and Cu 2+, i. e. Cu2+ and Zn 2+ form a ligand
bridged by metal ligand complex, is obtained by using erythrocuprein with the Zn 2+ partially replaced by
Co2+. This substitution caused a marked reduction of the Cu-ESR signal indicating magnetic interaction
between the two paramagnetic ions.
40
Erythrocuprein (Moles)
Pure erythrocuprein
30
20
Increasing
concentration
of dithionite 10
0 10 20 30 40
2600 2800 3000 3200 3400 (G) Dithionite (Moles)
(a) (b)
Fig. 9.44 (a) ESR spectra of bovine erythrocuprein (0.305 mM) in absence and presence of
aqueous (sodium or potassium) dithionite (7.04 mM) under argon for 2.5 min at 20°C before
freezing to 77 K (pH = 7.5, microwave frequency 9.175 GHz) (b) Behaviour of erythrocuprein
reduction with concentration of dithionite.
Table 9.15 ESR parameters of anion-erythrocuprein complexes.
Complex g g⊥ A⎢⎢(G)
⎢⎢
Native erythrocuprein in water 2.27 2.09 137

Azide
2 per mole 2.263 2.074 175
3 per mole 2.260 2.072 148
4 per mole 2.254 2.069 153
8 per mole 2.253 2.066 155
200 per mole 2.242 2.054 158
2 azide and 2 cyanide 2.22 180
Excess cyanide 2.21 188
Excess cyanate 2.268 2.068 160
Excess thiocyanate 2.27 2.076 148
Chemical, structural and kinetic properties of ESR inactive biological molecules such as the enzymes
(lysozyme, ribonuclease, DNA polymerase, citrate synthase), lipids, peptides and amino acids can be obtained
from ESR studies of synthetic organic free radicals which are chemically attached to these molecules. Such
radicals are•
called spin label radicals (i.e., paramagnetic markers). They are generally nitroxide radicals of the
type R 2 N O . They are very stable and inert and have characteristic, intense and simple ESR spectra that are
sensitive to molecular environment. Specific sites in a molecule can be tagged by choosing suitable spin label
and reaction conditions. Thus, by following the changes of the ESR spectrum, i.e. line splitting and g-factor
parameters of the spin label radical in the free and bound state, one can deduce information regarding the envi-
ronment close to the binding site or about the biomolecule as a whole. For example the spin label radical: N-(1-oxyl-2,2,5,
5-tetramethylpyrrolidinyl)-maleimide (I) is attached to bovine serum albumin (BSA). This particular radical is chosen
because N-ethyl maleimide (II) reacts specifically with the
—SH groups of BSA. Thus, in order to study the envi-
ronment close to the —SH site we need a radical which O O
contains a functional group to react with the sulphur atom. H H
C N O C
This radical can also be attached to an amino site in the C C
lysine group in BSA. The spin-labeled lysine is prepared by N N C2H5
C C
blocking the —SH group of BSA with N-ethylmaleimide C C
H H
and then reacting the product with nitroxide radical. The
O O
ESR spectra of the nitroxide radical when it is attached to
(I) (II)
the —SH group and when it is attached to the lysine group
of BSA is shown in Fig. 9.45. The arrows in Fig. 9.45(a) shows that the spectra is a superposition of two spectra of the
radicals because of two different environments. The similarity between the central spectrum of BSA with spin labelled
—SH group [Fig. 9.45(a)] and central spectrum of BSA with
spin labelled lysine group [Fig. 9.45(b)] suggests relative free (a)
motion of the spin-label radical. The broad line spectrum of
BSA with spin labelled —SH group indicates that the spin
label cannot rotate freely. The rotational time estimated from
peak-to-peak width and the measured peak-to-peak amplitudes
of three lines of ESR spectrum of spin label is about 10−8 s. At
pH 2.1, the reversible denaturation of BSA occurs. In such a
situation, the intensity of the ESR signal of protein with spin- (b)
labelled —SH group becomes zero, whereas becomes double
in case of protein with spin-labelled lysine. It is to be noted that
even in the mobile spin label case, the intensity of the three
lines is not equal as expected thereby leading to suggest that
the spin label does not have complete rotational freedom and 20 G
asymmetry of lines is caused by anisotropic effects. A sche- Fig. 9.45 ESR spectrum of bovine serum aibumln (BSA)
matic diagram for the native and spin-labelled protein is shown spin labelled with the nitroxide maleimlde. (a) BSA with spin
in Fig. 9.46. In conclusion, the ESR studies reveal that (i) the labelled—SH group, and (b) BSA spin labelled lysine group.
2.001
2.081
(a)
(a) Native protein
2.076
SH
(b)
(b) Spin labelled SH

2.121
S
(c)
(c) Spin labelled lysine
SH
•
Fig. 9.47 ESR signals of O 2 obtained both enzymi-
cally and non-enzymically obtained at −170°C (a)
Xanthine-Xanthine oxidase reaction. The sloping
Fig. 9.46 Schematic diagram of (a) native protein, (b) baseline is attributed to an iron signal. Note the
protein with spin-labelled −−SH group yielding an im- molybdenum signals (over modulated). (b) and (c):
•
mobile spin, and (c) protein with spin-labelled lysine group Non-enzymically induced O 2 from H2O2and NaIO4 at
yielding mobile spin. pH 9.9 and 13.2 respectively.
protein structure is sufficiently open for the entry of spin label for reaction with —SH group, and (ii) the protein
structure in the neighbourhood of —SH group is sufficiently close for the immobility of the spin label, i.e. the spin
label cannot rotate freely. Thus the protein has an elastic character.
The conformational changes of biomolecules in the solution phase can be studied using biradicals, i.e. mol-
ecules containing two radicals in a single molecule connected by long chains. Spin exchange occurs between two
radicals. The conformational changes of biomolecules cause dynamic changes in the exchange coupling between
the radicals. This results in a time-dependent ESR spectrum.
Many biological reactions take place via free radical mechanism. The enzymatic catalysis occurs through
free radical intermediates. The study of oxygen, superoxide and hydroxyl radicals are of importance in cancer
•
research. The most established source of superoxide ( O 2) in enzymic systems proved to be flavoproteins. The
actual presence of ( O 2 ) is determined by ESR studies using the enzymic reactions of flavoproteins. The ESR
•
signals of O 2 obtained both enzymically and non-enzymically recorded at −170°C, by rapid freezing technique
is displayed in Fig. 9.47(a−c).
Nitric oxide is a relaxant of blood vessels. These radicals may be studied in vivo by CW and FTESR meth-
ods to determine their activity in different organs and tumours. A special technique called the spin-trap is
employed for their identification and detection. A spin-trap is an ESR inactive molecule by itself but forms
an adduct by extracting a free radical thereby becoming ESR active. Nitrogen and proton, etc., hyperfine
splittings of the spin adducts are diagnostic parameters for the identification of the radical spin adduct
formed.
Several narrow line spin probes capable of targeting the organs of living animals have been developed. ESR
imaging on animals is done to study the action of brain, heart, liver and tumour growth. In vivo ESR imaging of
drug distribution and oxygen and nitric oxide mapping have been reported. ESR images of nitric oxide have been
measured at 700 MHz in the head of a rat. Anatomic FTESR images of mice have been measured at an operating
frequency of 280 MHz.
PROBLEMS
1. At what temperature, will the intensity of ESR signal 9. Predict the number of lines in the ESR spectrum of 2,5-
reduce to one-fourth of its value at 77 K? methyl-1,4-benzosemiquinone and what would be the
[Ans: 308 K] number of different types of coupling constant?
2. Prove that [Ans: 21, 2]
⎛ gf l t ⎞ • Hint: Protons in two different environments are
a(G) = 0. a(MHz)
⎝ gradical ⎠ involved. The large septet with an intensity ratio of
1 : 6 : 15 : 20 : 15 : 6 : 1 arises from six equivalent
3. For an ESR instrument that operates at a frequency of 70 protons. Each of these lines is then split into triplet
GHz, calculate the static magnetic field required and the by weak coupling of two protons (2 × 1/2 × 2 + 1)
value of ΔE. [Ans: H = 25 kG; ΔE = 4.63 × 10−16 erg] = 3 Distance between triplet centres corresponds to
4. Predict the number of ESR lines in (a) p-semibenzoquinone, larger coupling constant while the distance between
(b) monochloro-, (c) 2,3-dichloro-, (d) trichloro-, and (e) the hyperfine pattern of each triplet corresponds to
tetrachloro semibenzoquinone, (f) NH2 , (g) AB3 (I for A =0 the smaller coupling constant.
and B = 3/2), (h) atomic oxygen, (i) NF F2 ,( j) carbon black. 10. Predict• the number of lines in the ESR spectrum of
How does the splitting of each spectrum arise? O3 )2 NO2 − . The natural abundance of N14 (I = 1) is
[Ans: (a) 5, (b) 4, (c) 3, (d) 2, (e) 1, (f) 9 (Three major 99.635 per cent and of N15 (I = 1/2) is 0.365 per cent,
lines with intensity ratio 1:1:1 are due to coupling of Also calculate the relative intensity of ESR lines.
electron with nitrogen nucleus. Each of these lines then [Ans: 3(14N) + 2 (15N) = 5; I(14N)/I(l5N) = 182]
again split into triplet by weak coupling of two equivalent 11. Calculate the p-spin density rm on carbon atom m in the
protons), (g) 4, (h) 6, (i) 9, (j) 1] following planar ring system.
5. Predict the number of ESR lines in the spectrum of
K3CrO8. The natural abundance of nonmagnetic iso- System μ aH, μ (G)
topes of chromium (I = 0), i.e. 50Cr, 52Cr, 54Cr = 90.5 per
⎛ 3⎞ Anthracene • 1 (−) 2.74
cent and of magnetic isotope 53Cr ⎜ I = ⎟ is 9.5 per cent
⎝ 2⎠ 2 (−) 1.51
Also compute the relative intensity of ESR lines. 9 (−) 5.34
[Ans: 1 (nonmagnetic isotopes) + 4 Phenanthrene •
1 (−) 3.60
(magnetic isotope) = 5.
2 (+) 0.72
I( )
= 38]
I( ) 3 (−) 2.88
4 (+) 0.32
6. The measured ‘a’ values for the two non-equivalent 1
and 2 sets of protons of naphthalenide ion are (−) 4.9 9 (−) 4.32
and (−)1.83 G respectively. What would be the electron
•
density at each of the 1 and 2 carbon atoms? [Ans: Anthracene : r1 = +0.122;Phenanthrene • :
[Ans: 0.21 and 0.08] r9 = +0.192]
7. (a) How many lines are expected in the spectrum of free 12. Show that the Zeeman interaction for electron is very
radical p-benzosemiquinone and what would be the large as compared to that of proton at 3.3 kG.
relative intensities of these? •
13. The hyperfine splitting constant for 1H in 3 radi-

(b) The magnitude of coupling between each ring hydro- cal is 25 G. Draw an energy-level diagram to show
gen and the paramagnetic electron is about −2.4 G. the expected number of lines and intensities and
Calculate the electron density at each atom of the hence construct a theoretical stick diagram. Extend
molecule. the problem to tetracyanoethylene anion radical,
[Ans: (b) each H = − 0.005, each C = − 0.085,
a(14N) in TCNE = 1.56 G.
each O (by difference) = − 0.24]
14. Calculate the population difference between the two
•
•
Hint: Electron density on carbon atom of CH3 radical states of the electron at T = 300 K and H = 3.3 kG pro-
is − 0.85 [a(•CH3) = −23.7 G]. vided the total number of unpaired electrons in both
populations is 105. What would be the effect on popu-
8. Predict the number of lines in the ESR spectrum of
lation if (i) applied field is doubled and temperature is
2-methyl-1-4-benzo-semiquinone radical ion and what
reduced to one half, (ii) temperature is doubled and field
would be the relative intensity of these lines?
is kept constant, and (iii) temperature is doubled and
[Ans: 7, 1 : 6 : 15 : 20 : 15: 6 : 1] field is reduced to one half?
• Hint: If a double bond or a system of double bonds 15. The ratio of the two unpaired electron populations in
is followed by a single bond, e.g. methyl substituted thermal equilibrium is governed by the Boltzmann dis-
aromatic radicals, the double-bond character may tribution law. Determine the ratio between two popula-
be partially transferred to the single bond. The spin tions at T = 300 K and H = 3.3 kG if the total number
density is the same at the methyl protons as at the of unpaired electrons in both populations is 105. What
ring protons, i.e. the methyl protons are equivalent to would be the effect on the population ratio if (i) applied
ring protons. field is doubled and temperature is reduced to one half,
(ii) temperature is doubled and field is kept constant, (a)

and (iii) temperature is doubled and field is reduced to
one half? Draw a graph between population ratio and
applied field at 300 K and also between population ratio
1
and at H = 3.3 kG.
T
16. The ESR spectrum of some hypothetical free radical
consists of five lines with intensity ratio of 1:3:5:3:1.
Draw absorption curve from the data and convert it to
the first derivative curve and vice versa. H
17. From the experimental and theoretical ESR data given
below in the tabular form, calculate the values of
angles q and j.
Variation Radical Nucleus Hyperfine Constants (G)

Number
Observed Calculated (b)
a t || Aiso (2s) 2Bo(2p)
•
1 (C6H5)3 Si 29
Si 80 40 1600 70
•
2 (C6H5)3 73
Ge 83 25 830 31
•
3 (C6H5)3 119
Sn 1866 469 15550 642
[Ans: 1: q, j =116.2°, 101.9°; 2: 114.9°, 103.5° 3:

113.5°, 105.1°]
18. The coupling constants for the ortho-, meta- and para H
•
protons in (C6H5)3 C are (−) 0.93, + (0.46) and (−) 0.93G

Fig. 9.48 (a) First-derivative ESR spec-
respectively. Calculate the p-spin densities for the trum of a free radical, and (b) absorption
respective ring protons. Use Q = −23.96. spectrum of some free radical.
[Ans: r(o) = + 0.039; r(m) = − 0.019; r(p) = + 0.039]
19. Consider an unpaired electron that interacts with two 22. (a) When solutions of equimolar compounds of p-benzo-
inequivalent protons ( ). Sketch the elec- quinone and hydroquinone are mixed, slightly soluble,
tron-spin transitions that occur. How many lines does the nearly black crystals of quinhydrone begin to sepa-
EPR spectrum have? Prove the statement in the text that, rate out. Quinhydrone is molecular compound consist-
in general, the number of lines is given by 2n for an elec- ing of these two substances, in which the molecules
tron that interacts with n inequivalent protons. are united through hydrogen bonds and, besides, by
charge transfer from hydroquinone to the quinone;
[Ans: Four lines of equal intensity]
• Hint: For two equivalent protons, number of lines = H
(2nI + 1) = (2 × 1/2 × 2 + 1) = 3.
20. By making use of Table 9.5, prove that in a square planar O O
Cu (II) complex, when the electron resides in the d x 2 y 2
and d z2 orbitals respectively, the values of g⊥ are given
by
2λ
gxx g⊥ = 2 00 −
Ed 2 Edxz ,dyz O O
x y 2
and H
Quinhydrone
6λ
gyy g⊥ = 2.00 −
Ed 2 Edxz, yz
z
21. (a) Convert the first-derivative spectrum of a free radical

shown in Fig. 9.48 (a), into its absorption spectrum. (b) If we label, say benzoquinone, by introducing deute-
(b) Sketched in Fig. 9.48 (b) is the absorption spec- rium into it, it will turn out that the two halves of the
trum of some free radical, obtain its first derivative quinhydrone molecule do not inter convert, as might
spectrum. be expected; and
(c) But if quinhydrone is dissolved in alkali, i.e. if the to study the phenomena occurring in the above
bonding hydrogen atoms (protons) are removed, cases?
the hydroquinone anion will donate one electron [Ans: (a) IR, NMR, electronic spectroscopy
to the quinnone, and both compounds will be con- (b) IR, NMR, electronic spectroscopy (c) EPR.]
verted into two identical molecules of semiquinone.
What spectroscopic techniques might be employed
•
O O O O
+ 2 2
O O O O
•
CHAPTER 10 MÖSSBAUER
SPECTROSCOPY
The book of nature is written in a mathematical language whose letters are triangles, circles, and other geometri-
cal figures.
––Galileo
10.1 INTRODUCTION
Mössbauer spectroscopy (MBS), also known as g -ray nuclear resonance fluorescence –, g -ray recoil free and zero
phonon spectroscopy, deals with the study of recoilless nuclear transitions arising due to the absorption or fluorescence
of g -rays by the sample called target. The absorption of g -rays by the sample depends on the electron density around
the nucleus and the number of lines obtained in the Mössbauer spectrum are related to the symmetry of the target
under investigation. MBS is of fundamental importance in that it provides a means of measuring comparatively weak
interactions between nucleus and surrounding electrons. Unlike nuclear magnetic resonance and nuclear quadrupole
resonance, which deal with only the ground-state properties of the nucleus, MBS deals with both the excited as well
as the ground-state properties of the nucleus. It was discovered in 1958 by Rudolph L Mössbauer, a German scientist
working for his PhD degree at Heidelberg University and in 1961, the Nobel prize was awarded to him for researches
concerning the resonance absorption of gamma radiation.
10.2 PRE- AND POST-MÖSSBAUER ERA

Before going into the details of the theoretical and practical aspects of MBS, Excited state Excited state
let us first review the pre-Mössbauer time during which it was realised on the
basis of g -ray resonance fluorescence experiments that a small overlap between g -ray photon
the emission and absorption transitions, as expected, theoretically exists. The
phenomenon of g -ray resonance may occur between two nuclei shown schemati-
Ground state Ground state
cally in the ground and excited states in Fig. 10.1. It is pertinent to note that in Fig. 10.1 Schematic diagram for the
g -ray nuclear fluorescence, usually the low-lying nuclear excited states having nuclear transition between excited
lifetimes in the range of 10−11–10−4 second are involved. Now when the g -rays with and ground states in two nuclei.
energy Eg is emitted by the nucleus (source), it experiences a recoil Eg such that
E0 − ER = Eg (10.1)
E0 − Eg = ER
where, E0 is the nuclear transition energy.
Similarly, for the absorption of g -ray by the nucleus within the absorber (target), we can write,
E0 + ER = Eg (10.2)
Eg − E0 = ER
Accordingly, the separation between the emission and absorption will be 2ER and is shown in Fig. 10.2(a).
Here, no overlapping will occur between the absorption and emission lines.
Now, if M0 is the rest mass of the nucleus it will acquire a velocity vR by the emission of g -ray such that
ER
1
M v R2 =
(M 0 v R ) = PR2
2
(10.3)
2 2M 0 2M 0
Here, pR is the momentum of the recoiling nucleus.
Macroscopic model Energy relationship Atomistic model

(a)
Boat
g
Island ER
Pre-Mössbauer time
EO − ER EO EO + ER
(b)
Emission Absorption
ER ER g
ER
EO − ER EO EO + ER
Mössbauer-and post
Boat frozen in ice (c)
Mössbauer time
g
EO E
Fig. 10.2 Energy relationship relevant to Mössbauer effect: (a) 2ER > Γ (b) 2ER < Γ (c) 2ER = Γ and macroscopic and atomistic models of the
pre-Mössbauer era, MB- and post-Mössbauer era.
We know that the momentum of g -ray is given by,

Eγ
Pγ = (10.4)
c
According to law of conservation of momentum, pR and pg are equal in magnitude and opposite in sign. There-
fore, for small recoils (ER << M0 c2), we may write
Pγ2 Eγ2 E 02
ER = = =
2M 0 2M 0c 2 2M 0c 2
E 02
or 2E R = (10.5)
M 0c 2
For a nucleus with E0 = Eg = 104 eV, M0 = 50 amu and lifespan of the excited nuclear state = 10−7 second,
ER = 10−3 eV
or 2ER = 2 × 10−3 eV.
The line width G of the emitted line is defined as
h 1
Γ= ⋅ (10.6)
2π τ
where, t is the mean lifespan of the excited state of the nucleus, t is related to half life of the state by,
1 2 0.693
= = (10.7)
τ t1/1 2 t1/ 2
If t is given in electron volts (1 eV = 1.60219 × 10−19 J = 96.49 kJ mole−1) and t in seconds, then Eq. (10.6)
reduces to
6 58 × 10 −16
Γ= eV (10.8)
τ
Thus for the typical nucleus,
6 58 × 10 −16
Γ= = 6.58 × 10 −9 eV
10 −7
This small value of G (10−9 eV) as compared to 2ER (10−3 eV) suggests that there is no possibility of absorption
of g -photon emitted by the nucleus at rest. Thus, it follows from equations (10.5) and (10.6) that resonance absorp-
tion will not occur if 2ER > G, For resonance absorption to take place, 2ER ≈ G , i.e. resonance absorption will take
place only if an overlapping happens between the emission and absorption lines as shown in Fig. 10.2(b).
Mössbauer Spectroscopy 703
This is the picture when the nucleus is at rest. Let us now consider the emission of g -rays from a nucleus
moving (due to thermal agitation) with velocity vN along the direction of g -rays. The nucleus will now recoil with
1
a new velocity (vR + vN) and a total kinetic energy M 0 ( R 2
N ) .
2
By law of conservation of energy, we can also write,
1 1
E0 M 0 v 2R = Eγ M 0( R + v N )2 (10.9)
2 2
Further, if relative velocity vS is imposed on the source, the velocity of nucleus, i.e. vN in Eq. (10.9) should be
replaced by (vN + vS). Consequently, Eq. (10.9) modifies to,
1 1
E0 M 0 v 2R = Eγ M 0( R + v N + vS ) 2 (10.10)
2 2
⎡ Eγ E 0 ⎤
⎢Since M 0 v s = c = c ⎥
⎢ ⎥ (10.11)
⎢ E ⎥
⎢Therefore, vS = 0 ⎥
⎣ M 0c ⎦
Eq (10.10) yields
1
Eγ E0 − M 0 ( N S )2 M 0 vR ( N + S ) (10.12)
2
If vS is ignored for the time being, then Eq. (10.12) reduces to
1
Eγ E 0 − M 0 v 2N M 0 vR vN (10.13)
2
or Eg = E0 − ER − ED
1
where, ER M 0 v 2N and E D M 0 vR vN
2
It follows from Eq. (10.13) that the energy of the emitted g -photon is deficient by a recoil kinetic energy
⎛ 1 ⎞
⎜⎝ E R M 0 v 2N ⎟ which is independent of the initial recoil velocity vR of the nucleus and the thermal Doppler
2 ⎠
energy (ED = M0 vN vR ) which is vR dependent and can therefore be positive or negative.
Thus, in general we may write
Eg = E0 − ER ± ED (emission) (10.14)
Eg = E0 + ER ± ED (absorption)
If q is the angle between the directions of propagation of g -rays and moving nucleus respectively then
vN
ED = Eγ cos θ (10.15)
c
Applying the law of conservation of momentum to the emission process, we obtain,
Eγ
M0 R M 0 (vR + vN ) + (10.16)
c
Eγ is the momentum of g -photon and hence the recoil momentum is Eγ . Thus the recoil energy is expressed as
−
c 2
c
Eγ
ER = (10.17)
2M o c 2
Eγ2 −4
or E R = 5 37 × 10 eV
Mo
where Eg is in keV, M0 is in amu.
It follows from Eq. (10.17) that the recoil energy ER is directly proportional to the square of the energy of g -rays
and inversely to the mass of the nucleus. ED depends upon the thermal motion of the nucleus. Nuclear recoil (ER)
and Doppler (ED) energies are of the order of 10−4−10−2 eV and are compared with some nuclear and chemical
interactions in Table 10.1. For nuclei moving isotropically, the energy distribution due to linear Doppler effect to
Eg , i.e. line positioned at Eg is governed by Maxwell velocity distribution law. The Doppler broadening of energy
distribution estimated from the theory of ideal gas is given by
DT 2 Ek ER (10.18)
1 M
where, E k = κT = 2 is the mean kinetic translation energy per degree freedom per atom, κ is the Boltzmann
2 2vN
2
constant and vN is the mean square velocity in the direction of observation. Eq. (10.18) reveals that DT is pro-
portional to the square root of the thermal energy (Ek) and recoil energy (ER). DT for a gas made up of atoms with
M0 = 100 is about 103 eV for 1 keV g -ray transitions and about 1 eV for 1 MeV transitions. The shift in g -ray energy
due to linear Doppler effect manifests itself in line broadening. This suggests that energy loss because of recoil
can be compensated by the linear Doppler effect. Consequently, a small overlap between the emission and absorp-
tion distributions takes place as shown in Fig. 10.2(b). It is to be noted that in ultraviolet and optical spectroscopy,
the emission and absorption bands overlap to a great extent due to the small recoil energy involved.
This was the situation in the pre-Mössbauer time. During this period, g -ray resonance fluorescence experiments
had already been performed. It was realised that the small overlap between emission and absorption areas, i.e.
energy distributions as shown in Fig. 10.2(b) made these experiments possible. The energy loss in the recoil
process was compensated by ultracentrifugation, increasing the temperature of the nucleus and (p, g ) nuclear
reactions with photon being emitted by the excited nucleus in flight; the Doppler contribution in this case as
already mentioned is given the expression:
E
E D = γ VN cos θ (Eq. 10.15)
c
Hence, it is possible to find an angle q at which g -rays are resonantly absorbed. The compensation of the
recoil energy loss by Doppler effect was used in 1951 by P B Moon to obtain resonance scattering of the 411 eV
transition of 198Hg by ultracentrifugation. Using a rotor to move the 198Au source, which decays to 198Hg and a
mercury absorber, he observed that the intensity of the resonantly scattered radiation grows with the velocity of
the rotor. Maximum intensity was observed at ≈700 m/s in agreement with the relative velocity vS imposed on the
source, i.e. 670 m/s in this case.
Table 10.1 Energies of some nuclear and chemical interactions and energy equivalence.
Interactions Energy range (kJ mole-1/eV)

Mössbauer γ-ray energy (Eγ ) 106 − 107/104 − 105
Chemical bonds and lattice energies 102 − 103/1 − 10
Electronic transitions 50 − 500/0.5 − 5
Molecular vibrations 5 − 50/0.05 − 0.5
Lattice vibrations 0.5 − 5/0.005 − 0.05
Nuclear recoil(ER) and Doppler(ED) energies 10−2− 1/10−4 − 10−2
Nuclear quadrupole coupling constants <10−3/ < 10−5
Nuclear Zeeman splittings <10−3/ < 10−5
Heisenberg natural line widths 10−7 − l0−4/l0−9 − 10−6
Energy equivalence
1 atomic mass unit = 931 MeV = 1.49 × 10−3 erg = 3.56 × 10−11 cal
1 electron mass = 0.510 MeV
1 MeV = 1.07 × 10−3 amu = 1.602 × 10−6 erg
= 3.827 × 10−14 cal
1 eV/molecule = 23.06 kcal/mole
1 mm/second = 4.807 × 10−8 eV = 11.62 MHz
Mössbauer in 1958 while studying the resonance absorption of the 129 keV transition of 191Ir observed that
fluorescence becomes stronger at lower temperatures [Fig. 10.2(c)] instead at higher temperatures as expected.
The result was interpreted using Lamb’s theory for resonance capture of thermal neutrons in crystals. The marked
increase in resonance scattering may be because of the emission and absorption of recoil free g -rays since at
low temperatures, the nuclei are fixed. To summarize, Figs 10.2(a) and (b) represent the pre-Mössbauer time and
10.2(c) characterise the Mössbauer discovery, i.e. a recoil-free line exists at E0. This line is called the Mössbauer
line. The pre-MB era and post-MB era may also be demonstrated by the macroscopic model based on the g -jumps
from a boat with an effort of energy E0 and atomistic model (Fig. 10.2).
So far we have treated the nucleus to be free at rest or in motion. If the nucleus is bound in a solid, the interaction
arising because of the exchange of individual recoil energy among the atoms in the lattice need be considered. The
effect of lattice coupling in a recoil nucleus is expressed in terms of a dynamical effective mass Meff given by
Eγ2
ER = (10.19)
2M efff c 2
For large values of Meff ; ER << G . The binding energy of an atom in a lattice is about several eV while the recoil
energy of the free atom for a 100 keV g -emission is about hundred times less. This energy is dissipated in alter-
ing the vibrational state of the lattice, i.e. by phonon propagation through the crystal lattice. Thus, if phonons are
created in the nucleus in the recoil process of it, the energy of the emitted g -ray will be deficient by
hω Ei
∑i 2π
or ∑ hν Ei and
i
hω
E γ E 0 − ∑ Ei (10.20)
i 2π
or Eγ E 0 − ∑ hν Ei
i
hω Ei
where ω Ei /n Ei corresponds to the frequency of the phonons. When the factor
i
∑
2π
> Γ, the g -ray cannot be
resonantly absorbed. But we know that the recoil energy is negligible in a crystal, therefore if no energy is lost by
phonon creation, all available energy is emitted in the form of g -ray. Since the zero phonon process was known
from x-ray scattering long before Mössbauer effect and the low energy g -rays (5−150 keV) and x-rays are basi-
cally of the same electromagnetic nature in the same energy region, so the zero phonon process is also possible in
Mössbauer effect and hence the effect is also known as zero phonon effect. It is to be noted that in a zero phonon
emission, the whole crystal rather than a single nucleus recoils.
Before we proceed further, the characteristic parameters of the MB line, namely position, intensity and
bandwidth, and the factors governing these parameters need be defined since they play an important role in the
development of theoretical and experimental aspects of MB spectroscopy and are of great value in the study of
chemical problems.
10.3 LINE POSITION

It is the most concerned observable to the experimenter and corresponds to the energy difference between the levels
of source and absorber. A–MB spectrum may be a single lined, doublet, a six-fingered or a complex one according
to the shifting and or splitting of the nuclear levels in the source/absorber excited and ground states due to electric
monopole, quadrupole and magnetic interactions. The presence of different lines may also be because of more than
one valence state or inequivalent sites. The line position obtained will give the other well-known Mössbauer param-
eters: isomeric shift quadrupole splitting, Zeeman splitting, temperature shift, pressure shift and or gravitational
red shift. The definition, origin and information which one can have from their (of the first three only) numerical
estimates will be given in detail when we will discuss the hyperfine interactions.
10.4 INTENSITY OF MÖSSBAUER LINE AND PARAMETERS AFFECTING IT

The relative intensity of MB resonance line, i.e. the probability of recoilless events (emission or absorption) is expressed
in terms of recoil-free fraction ‘f’ (called Lamb–Mössbauer factor) of g -rays which is related to the vibrational proper-
ties of the crystal lattice. The expression relating ‘f’ with the vibrational properties of lattice can be deduced quantum
mechanically either from the simple Einstein model or Debye model of the lattice vibrations of the solid.
10.4.1 Einstein Model

In the Einstein model of the solid, the infrequent occurrence of multiphonon processes is ignored and only the
mean recoil energy of a single phonon process is considered, i.e. ER = (1−f) hnE. When the recoil energy of g -rays
emitted from an atom bound in a solid is insufficient to displace the atom, a thermal spike is created. Consequently,
many high levels in the oscillator will be populated by the thermal excitation. A thermal equilibrium is reached
very quickly between this hot spot and the surrounding lattice. When the recoil energy is larger than the separation
between the vibrational levels (i.e. ER >> hvE; vE is the frequency of vibrational oscillations of the atoms) of the
Einstein solid, the displacement of atoms occur and a thermal spike is created. Low-energy g -rays (≈ 5−150 keV)
are important for our considerations. The condition ER < hvE suggests that zero phonon process takes place in the
lattice, i.e. emission and absorption processes occur without excitations of phonons in the lattice. Specifically, we
2
select a recoil energy ER = hvE = hqE; where qE is the Einstein temperature of the solid.
3
Accordingly, f, i.e. the number of recoil free g -ray events divided by the total of g -ray events is given by
⎡ − Eγ2 〈 x 2 〉 ⎤
f = exp ⎢ 2⎥
(10.21)
⎣ ( hc / 2π ) ⎦
h
〈 2
〉= for the lowest einstein level.
8π Mv
2
Here, <x2> is the mean-square vibrational amplitude of the resonating nucleus in the X-direction which is the
direction of observation of <x2>, i.e. g -ray. It is assumed to be an isotropic function and is linearly proportional
to temperature. It follows from Eq. (10.21) that (i) the recoil free fraction ‘f’ decreases as the energy of g -rays
increases. Due to this very reason, the highest g -ray energy so for used in the MB effect is the 187 keV transition
in 190Os, and (ii) ‘f’ will be large for tightly bound atom with small mean square amplitude of vibration.
10.4.2 Debye Model

In this model, the multiphonon processes though occur infrequently but are not ignored, i.e. the large number of
vibrational levels in the solid and their energy distribution is taken into account. Each of the levels available has a
certain probability to be excited by the recoil energy. The mean energy is transferred to the lattice. ER and a recoil-
free line appears at E0. The recoil-free fraction ‘f’ according to Debye model of the solid is given by
⎡ −3E ⎧ ⎛T ⎞ D
2 θ /T
x ⎫⎪ ⎤
f = exp ⎢ ⎨1 + 4 ⎜ ⎟ ∫ x dx ⎬ ⎥
R
(10.22)
⎢⎣ 2κθ D ⎩⎪ ⎝ θ D ⎠ 0 e − 1 ⎭⎪ ⎥
⎦
Here, qD is the Debye temperature and is defined as hvD = k q D; hvD represents the separation between energy
levels of Debye solid. This expression is valid for a monoatomic lattice of identical atoms.
hv m
θD = , v m is limiting vibrational frequency of a solid.
κ
/
For θ D >> T or as T → 0, so we can replace θ D T by ∞. In such a case
⎡⎛ − E ⎞ ⎧ 3 π 2T 2 ⎫ ⎤
f = exp ⎢⎜ R ⎟ ⎨ + 2 ⎬ ⎥
⎣⎝ κθ D ⎠ ⎩ 2 θD ⎭⎦
⎛ −3E R ⎞
or f = exp ⎜ (10.23)
⎝ 2κθ D ⎟⎠
For T > θ D ,
⎛ −6 E R T ⎞
f = exp ⎜ (10.24)
⎝ κθ D2 ⎟⎠
It follows from Eqs (10.23) and (10.24) that (i) ‘f’ increases as the temperature decreases. This was what
observed by Mössbauer with 191Ir, (ii) the higher the value of qD of the solid, the larger the value of ‘f’ and (iii) the
MB effect is limited to relatively low energy g -ray emission since ‘f’ decreases rapidly with the increase in energy.
The effect has been observed up to 155 keV g -radiation. Moreover, it is also known from Eq. (10.21) that recoil
free processes in a solid is solely determined by <x2>. No further specification regarding the solid is necessary.
Therefore, the MB spectroscopy is not only applicable to crystalline materials only but also to liquid, amorphous
and gaseous materials.
10.5 LINE WIDTH OF MÖSSBAUER LINE AND FACTORS AFFECTING IT

It is well known that the ground state of nucleus is well defined, i.e. it has an infinite lifetime and there is no uncer-
tainty in its energy. However, the excited state is not well defined. The energy ‘E’ of the level is not sharp, but is
spread over a certain range. The width of this range can be obtained from Heisenberg–Bohr uncertainty principle
h
ΔE Δt ≥
2π
Here, ΔE is the uncertainty in energy and Δt is the time interval available to measure the energy. The uncertainty
in the lifetime of the excited state is expressed in terms of its mean life τ, and the uncertainty in its energy is given
by the width of the Mössbauer line at half height, i.e. G as indicated in Fig. 10.3.
Excited State I(E)

E
Io
E Io Γ
2
E
Ground State Eo
Fig. 10.3 Natural line width for the transition in which energy of the excited state level is spread over a certain energy range.
Thus, h
Γτ≥
2π (Eq. 10.6)
t is related to half-life of the state by
1 In2
= (Eq. 10.7)
τ t1/ 2
⎛ h ⎞ ⎛ h ⎞ ln 2 ⎛ h ⎞ 0.693
Thus Γ=⎜ ⎟ =⎜ ⎟ =⎜ ⎟
⎝ 2π ⎠ τ ⎝ 2π ⎠ t 1 ⎝ 2π ⎠ t 1
2 2
Γ is called the natural/inherent/Heisenberg line width of the source emission or absorption line. For a nuclear
excited state with t1/2 =10−7 second, Γ is of 10−9 eV. The ratio of natural line width and energy of the photon is a mea-
sure of the resolution of the emitted g -ray. Thus if energy of the excited state is 45.62 keV, the value of Γ /Eg = 10−13,
i.e. emitted g -ray will have natural resolution of 1 part in 1013. The maximum resolution obtained in atomic line
spectra is only about 1 in 108.
Further, it is to be noted that in the g -ray emission process, the de-excitation of the system may occur through
several mechanisms, namely radiative transition, internal conversion (IC), isomeric transition, i.e. change in the
valence or oxidation state (IT) of the atom,’ etc. All these processes take place independent of each other and
decrease the lifetime of the excited state. Consequently, the g -ray line broadens. Now if the excited state decays in
several different ways, the total energy spread of the excited level, G is equal to the sum of all the partial widths
Γ i . Thus as the number of modes of decay increases, the lifetime of the excited state also decreases.
Mathematically,
1 1
=∑ (10.25)
τ τi
h 1
or Γ ∑Γ = ∑
2π τi
The curve between excitation probability (i.e. observable width) W (E )

W(E) and energy of resonance radiation is Lorentzian type as shown in
Fig. 10.4 and is described by the Breit−Wigner relation
Γ2 / 4
W (E ) = (10.26) Γ
(E E )2 + Γ 2 / 4
If in a resonance experiment, an emission line of natural width overlaps
the absorption line of the natural width, Γ , the sum of both resonance line
widths, 2Γ, will be obtained. The Lorentzain line of such an absorption
experiment is given by the total absorption cross-section.
Γ2 E
σ abs σ0 (10.27) ER
Γ + 4 (E − E0 )
2 2
Fig. 10.4 A plot between excitation probability
and energy of resonance radiation.
With maximum cross-section
λ2 2 e +1 1
σ0 = ⋅ (10.28)
π 2( 2I g + 1) (α t + 1)
where, λ is the wavelength of the g -ray.

Similarly, the total scattering cross-section is given by
Γ2
σ scatt σ0 (10.29)
Γ 2 + 4 (E − E0 )
2
With the maximum cross-section
λ2 2 e +1 1
σ0 = ⋅ (10.30)
π 2( 2I g + 1) (α t + 1)2
Units of s0 are barn (1 barn = 10−24 cm2 = 10−28 m2). For s0 to be large, both E0 and at should be small. Ie and Ig
are the spins of the excited and ground states of the transition respectively, λ is the resonant wavelength, at is the
total of g -ray internal conversion coefficient and takes into account the competing mode of transition. Note that
not all nuclear g -transitions produce a physically detectable g -ray, a proportion eject electrons from the atomic
orbitals, giving x-rays and these internal conversion electrons instead. The internal conversion coefficient is thus
defined as the ratio of the number of conversion electrons (atomic s-electron) to the number of g -ray photons
emitted. For the most celebrated Mössbauer isotope 57Fe, emitting 14.4 keV g -ray, at is of the order of 10. For
high value of at it is difficult to observe the MB resonance line, e.g. for 73Ge, emitting 13.5 keV g -ray, at = 3600.
The values of at for 57Fe and 73Ge suggest that in the former s0 is reduced by a factor of 100 whereas in the latter
by a factor of about 1×107 which is very large indeed. The factor 2 in the denominator expresses the multiplicity
caused by two possible independent polarisations of the photons. It is to be noted that the expressions (10.28) and
(10.30) have been derived on the assumptions that (i) the emission line is very sharp, and (ii) only one absorp-
tion or scattering level exists. The characteristic parameters for selected transitions of some Mössbauer nuclei are
The observed line width of the MB line always differs from the natural line width. The deviation could be due
to Doppler broadening of an improper functioning of Mössbauer apparatus (oscillation) or saturation effects in
the absorber or in the source (self-absorption, i.e. resonance). In addition, a number of inherent physical processes
such as diffusion and Brownian motion of the atoms or molecules, relaxation processes, spin-flip process, super
paramagnetism, fluctuations near the critical temperature (magnetic) ferroelectric and other phase transitions,
etc., could cause line broadening.
The Doppler broadening due to the thermal agitation of the emitting and absorbing systems being temperature
dependent is given by
2κT
ΔΓ = v 0 log 2 , ΔΓ > Γ (10.31)
M 0c 2
Table 10.2 Mössbauer Isotopes and relevant parameters for the selected Mössbauer transitions.
Isotope Isotope Nuclear decay at E/E0 t1/2 Ig Ie mg
abundance (%)
me (nm) 2Γ (mm/s) ER (10−3 eV) keV (ns) (Qg barn) (Qe barn) (nm)
s0 (10−20 cm2)
1 3
57
Fe 2.19 57
Co (EC 270d) 8.17 14.4125 97.81 − − 0.0902
2 2
−0.1547 256.6 0.194 1.957 (0) (0.2)
1 5
57
Fe 2.19 57
Co(EC 270d) 0.14 136.46 8.7 − − 0.0902
2 2
(−) 4.3 0.2304 175.4 (0) (−)
3 5
6l
Ni 1.25 61
Co(b−1 99m) 0.12 67.4 5.06 − − −0.7487
2 2
0.47 72.12 0.8021 39.99 (+0.162) (−0.3)
1 3
119
Sn 8.58 119m
Sn (IT 250d) 5.12 23.871 17.75 + + −1.041
2 2
0.67 140.3 0.6456 2.571 (0) (−0.08)
5 7
121
Sb 57.25 121m
Sn (b− 76y) ~10 37.15 3.5 + + 3.359
2 2
2.35 19.7 2.104 6.124 (−0.26) (−0.36)
1 3
125
Te 6.99 l25
I(EC 60d) 12.7 35.46 1.48 + + −0.8872
2 2
0.6 26.56 5.212 5.401 (0) (−0.2)
5 7
127
I 100 127m
Te(b− 109d) 3.7 57.6 1.9 + + 2.809
2 2
2.02 21.37 2.5 14.03 (−0.79) (−0.71)
7 5
129
I nil (radioactive) 129m
Te(b− 33d) 5.3 27.77 16.8 + + 2.617
2 2
2.84 40.32 0.5863 3.21 (−0.55) (−0.68)
1 3
129
Xe 26.44 129
I (b−1.7 × 107y) 11.8 39.58 1.01 + + −0.7769
2 2
0.68 23.31 6.843 6.521 (0) (0.41)
7 5
l49
Sm 13.83 149
Eu (EC 106d) ~12 22.5 7.12 − − −0.665
2 2
−0.620 7.106 1.708 1.824 (≤ 0.06) (0.4)
5 7
+ +
l51
Eu 47.82 151
Gd (EC 120d) 29 21.64 9.7 2 2 3.465
2.587 11.42 1.303 1.665 (+1.16) (+1.51)
5 5
161
Dy 18.88 16l
Tb (b−16.9d) ~2.5 25.65 28.1 + + −0.472
2 2
(Continued)
Table 10.2 Mössbauer Isotopes and relevant parameters for the selected Mössbauer transitions—Cont’d
Isotope Isotope Nuclear decay at E

l /E0 t1/2 Ig Ie mg
0.59 95.34 0.3795 2.194 (+1.35) (+1.36)
1 3
Tm
l69
100 l69
Er (b−19.4d) 220 8.42 3.9 + + −0.231
2 2
0.5 21.17 8.33 0.2253 (0) (−1.30)

182
W 26.41 182
Ta(b−1115d) 3.2 100.102 1.37 0+ 2+ (0)
0.532 25.17 1.995 29.56 (0) (1)
3 5
189
Os 16.1 Ir(EC 13.3d)
189
8.2 69.59 1.64 − − 0.6565
2 2
0.988 8.419 2.397 13.76 (+0.91) (−)

3 1
193
Ir 62.7 193
Os(b−131h) ~6 73.028 6.3 + + 0.1589
2 2
0.47 3.058 0.5946 14.84 (+1.50) (0)
3 1
197
Au 100 197
Pt(b−118h) 4 77.35 1.9 + + 0.1449
2 2
0.419 3.857 1.861 16.31 (+0.56) (0)
5 5
237
Np nil (radioactive) 241
Am(a 458y) 1.06 59.537 68.3 + + 2.8
2 2
1.5 32.55 0.06727 8.031 (+4.1) (+4.1)
EC-electron capture, β−1-beta-decay, IT-isomeric transition, a-alpha decay, Eg /E0-nuclear transition energy, t1/2-half-life period, Ie and Ig-nuclear spin
quantum numbers excited (e) and ground (g) states respectively, mg and me-nuclear magnetic moments, s0-maximum resonance cross-section in barn,
273.8
G -Heisenberg line width, ER-recoil energy. The following derivations can be used for the determination of 2G , ER and s0; 2Γ= mm/sec (Eg in keV,
Eγ t 1/ 2
(E )
2
γ
⎡ 2.446 × 10 −15 ⎤ ⎡ 2 + 1 ⎤ ⎡ 1 ⎤
tl/2 in ns, ER = 5.36942 × 10−4 eV (Eg in keV, M in amu) and s0 = ⎢ ⎥.⎢ e ⎥.⎢ ⎥ cm (Eg in keV)
2
M ⎢⎣ ( ) ⎥⎦ ⎢⎣ 2 g + 1⎦⎥ ⎣1 + α t ⎦
where v0 is the frequency of harmonic vibrator. The MB spectral line is no more Lorentzian but becomes Gaussian.
The effective Doppler scattering cross-section is given by
⎛Γ π⎞ ⎡ ⎛ E E0 ⎞ 2 ⎤
σ scatt
D
σ0 ⎜ ⎟ exp ⎢− ⎜ ⎟ ⎥ , ΔΓ > Γ (10.32)
⎝ 2 ΔΓ ⎠ ⎢⎣ ⎝ ΔΓ ⎠ ⎥⎦
The maximum Doppler scattering cross-section is then expressed as
σ 0Γ 2 1 (10.33)
σ max,scatt
D
= ⋅
ΔΓ (1 + α t ) 2
2
The corresponding maximum Doppler absorption cross-section will be
σ 0Γ 1
σ max,abs
D
= ⋅ (10.34)
ΔΓ (1 + α t )
The line broadening due to diffusive motion of atoms could be explained by the sudden jump and continuous
diffusion models of Singwi and Sjolander. The diffusion in solids would be approximated by the first model while
in liquids by the second model.
According to sudden jump model, the atom or ion does not stick to a particular site in the lattice. This is pos-
sible in two ways. In the first, an electron may hop from the resonant atom (within the Mössbauer lifetime) to
another atom on a neighbouring site and in the second, the resonant nucleus itself changes its lattice site by jump
diffusion. When the jump frequency of the resonant atom matches with the reciprocal of the mean lifetime of the
1
excited state, , line broadening occurs. For either model, the diffusion broadening ΔG is expressed as
τ
ΔΓ
= σ( τ )D
2
(10.35)
Γ
where, D is the diffusion coefficient,
2π Eγ
K= ⋅ , σ = 12 / f r02 K 2 = 0 04
h c
in the sudden jump model and s = 2 in the continuous diffusion model, ‘fc’ is a factor which describes the
correlation in a direction between successive jumps of one and the same atom and r0 is the jump distance.
h
The simple jump model predicts a line broadening of whereas the Lorentzian full width predicted by the
πτ
Singwi and Sjolander continuous diffusion model is given by
h h
ΔΓ = (Γ + 2αβ
β) = ( Γ + 2K 2 D ) (10.36)
2π 2π
β
a << 1, K2 << and
D −1/ 2
⎡⎛ Γ + 2αβ ⎞ 2 ⎤
ΔΓ = (Γ + αβ ) ⎢⎜ ⎟ + 1⎥ ; α <1 (10.37)
⎢⎣⎝ Γ + 2αβ + 2β ⎠ ⎥⎦
6πη
ηa
where, b = , the quantity b−1 is called the characteristic or relaxation time. For atomic motion, b−1 is of
m
the order of 10−12−10−20 second. For particles, b−1 is of the order of lifetime of the nucleus in the excited state,
K 2D
α= ; α is a dimensionless number connecting the g -ray energy with relaxation time of the liquid and the
β
diffusion constant D, ‘a’ is the size of the particle and h is coefficient of viscosity.
The Gaussian full width is then given by
⎛ λ⎞
Γ = 2⎜ ⎟ [ + ]1/ 2 (10.38)
⎝ 2π ⎠
It has been observed that when a >1, (i) the width of resonance line varies as T / η rather than as T/h, and
h
(ii) the line is always narrower than ( 2αβ
β ). However, on the basis of experimental data, it is seen that the
2π
type of linear relationship between the experimental Mössbauer line broadening and T/h depends upon the system
T ⎛T ⎞ T
under diffusion studies in liquids, i.e. (i) ΔΓ ∝ (ii) log ΔΓ ∝ log ⎜ ⎟ with slope <1, and (iii) ΔΓ ∝ .
η ⎝ η⎠ η
h
The broadening of the Mössbauer resonance line due to diffusion of iron and in copper and gold corresponds to
2πτ
h
but not as expected on the basis of simple jump model. Diffusion of different-sized nuclei say 57Fe (a = 0.08 m),
πτ
Sn (a = 0.22 m) and 127I (a = 1.28 m) and b = 0.1 × natural width in a liquid of h = 1P (say machine oil) gives
ll9
ΔΓ ( ) = 2.754 and
ΔΓ ( ) = 16.00
ΔΓ ( ) ΔΓ ( )
on the basis of continuous diffusion model. The respective ratios from Eq. (10.37) are 2.509 and 10.74.
Further, it is interesting to note that resonance lines narrower than the natural line width could also be observed.
Line narrowing could take place due to delay coincidence measurements and thermal spike. In both the cases,
the resonant g -rays come from excited states with apparently longer mean lifetimes which results in narrowing
the line width. In the former case, the foregoing decay to the nuclear excited state can be used as signal in delay
coincidence experiment. In the following transition to the ground state, only g -rays from long-lived excited states
is selected. By this mechanism, the mean lifetime has artificially been increased and line width decreased. In the
latter case, the foregoing decay is associated with the creation of thermal spike because of which even high oscil-
lation levels will be occupied. In the transition from 57Co → 57Fe, more than 3 keV energy is dissipated. The
probability of following recoil-free emission becomes a function of time for which the nucleus has been excited
during the thermalisation process with decrease of <x 2>. Thus the resonant g -rays come from excited states with
apparently longer lifetimes. The Heisenberg principle is not violated in both the situations. The relaxation pro-
cesses cause, a decrease/increase in the lifetime of the excited state, consequently the broadening/narrowing of
MB line occurs.
10.6 HYPERFINE INTERACTIONS

Many of the applications of MB spectroscopy are based on the fact that the line widths encountered are small
compared to the characteristic energy of interaction of nuclei with their surroundings. The interactions arise due
to the coupling of nuclear magnetic dipole moment with an appropriate electronic or atomic property and are
called hyperfine interactions. The hyperfine-coupling mechanisms provide a wealth of information pertaining to
electron and spin-density distribution. The temperature dependence of these interactions also provide valuable
chemical information. The three main hyperfine interactions corresponding to the nuclear moments determining
the nuclear levels are electric monopole interaction (eo)–nuclear isomeric/chemical shift; magnetic dipole
interaction (m1)–magnetic hyperfine interaction/nuclear Zeeman effect and electric-quadrupole interaction
(e2)–nuclear quadrupole coupling.
The theoretical treatment of these interactions is based on the following three assumptions: (i) The nuclear
dimensions are infinite and the electrostatic potential sensed by the electrons is different inside and outside the
nucleus, (ii) The nuclear charge distribution is nonspherical and therefore the nucleus possesses an electric qua-
drupole eQ. Q is a measure of the deviation of the nuclear charge from spherical symmetry. (iii) All the three
effects may occur together, but only the magnetic and quadrupole interactions are directional and thus have a
complicated inter-relationship. The behaviour of isomer shift is independent of the other two interactions.
10.6.1 Centre Shift

The shift of the transition energy of the nucleus in an absorber (source) relative to the source (absorber) is known
as centre shift and is denoted by dc. This energy difference actually consists of two parts one of which is the physi-
cal, i.e. second-order Doppler shift whereas the other is the chemical, i.e. isomeric shift. The former arises due to
the difference between the mean square velocity of the Mössbauer probe in the absorber and in the source. This
is expressed as
hν
δ SOD
O =
⎡ v abs
2
− v s2o ⎤⎦ (10.39)
2c 2 ⎣
Here, v is the apparent frequency of the emitted photon and v is the apparent velocity of the absorber and the
source. The latter, i.e. isomeric shift results, because of the difference in the electrostatic interaction between
the positively charged nucleus density and the s-electron density at the nucleus in the source and the absorber
respectively. This is given by
δ IS
2
5
π Z e2 ( − )| ψ 0 2
Chemical
| ψ s (0)so |2
a term
(10.40)
Nuclear term
Here, Re and Rg are the radii of the nuclei in the excited and the ground states respectively, (–e) |ys(0)abs|2 and
(–e) |ys(0)so|2 are the s-electron densities at the nucleus in the absorber and the source respectively, Z is the atomic
number of the nucleus. The values of 1s, 2s, 3s Hartree–Fock wave functions and s-electron densities at the iron
nucleus have been recorded in Table 10.3(a) and (b) respectively.
Table 10.3(a) Values of 1s, 2s, 3s restricted Hartree–Fock wave functions and
s-electron densities (in atomic units; electrons per cubic Bohr radius) at the iron
nucleus (at r = 0) for various 3d electron configurations of the iron atom.
Configuration 1s 2s 3s
Fe (d )
I 8
259.9656 −78.78302 +29.10285
Fe (d )
II 7
259.9656 −78.77200 +29.16182
Fe (d )
III 6
259.9572 −78.77200 +29.26054
FeIV (d5) 259.9572 −78.77020 +29.49279
FeV (d4) 259.9504 −78.77805 +29.49729
Fe (d )
VI 3
259.9423 −78.79529 +30.040612
Table 10.3(b) Electron densities at r = 0 for various configurations.
( ) r =0
2
|y1s(0)|2 |y2s(0)|2 |y3s(0)|2 |y4s(0)|2 2 ∑ ψ ns 0
3d8 5378.005 493.953 67.524 11878.9

+
3d (Fe )
7
5377.973 493.873 67.764 11879.2
3d (Fe )
6 2+
5377.840 493.796 68.274 11879.8
3d5(Fe3+) 5377.625 493.793 69.433 11881.7
3d 4s (free atom)
6 2
5377.873 493.968 68.028 3.024 11885.8
2 1 r
Partial electron densities | ψ ns |r = 0 = ∑ | ns ( r ) |r = 0 for the configuration 3dn (n = 5, 6, 7, 8) and 3d 64s2.
4π n
The centre shift may thus be written as
δc δ SOD + δ IS (10.41)
When the source and the target are at the same temperature, i.e. room temperature, the centre shift becomes
equal to isomer shift since under this condition, the second order Doppler shift, i.e. dSOD is zero.
10.6.2 Second-Order Doppler Shift (dSOD)

The emitting or absorbing nucleus is not at rest but is vibrating on its lattice site. The frequency of vibration
is smaller than the reciprocal of the MB lifetime, so the average displacement and velocity are effectively zero
but the mean-square values of the velocity, <v2> are finite. <v2> depends on temperature, and pressure, Debye
temperature of the surrounding matrix and lattice defects, etc.
The relativistic expression for the Doppler effect on the apparent frequency v of the emitted photon as recorded
at the absorbing nucleus is
−1 / 2
⎛ v ⎞⎛ v ⎞
2
ν ν 0 1 − cos θ ⎟ ⎜1 − 2 ⎟ (10.42)
⎝ c ⎠⎝ c ⎠
Here, v0 is the frequency of the nucleus at rest, v is the apparent relative velocity of the emitting nucleus along
the direction making an q with the direction of propagation of g -ray.
The mean value of <v> is zero for a vibratory motion but the second-order term <v2> being finite can influence
the Mössbauer resonance.
⎛ v2 ⎞
ν = ν 0 ⎜1 + 2 ⎟ (10.43)
⎝ 2c ⎠
This gives rise to shift in MB line.
ν ν 0 Δν
= =− 2 =
δE ( )
ν0 ν0 2c Eγ
v2
or δE = − ⋅ Eγ (10.44)
2c 2
It is obvious that <v2> grows with temperature. Accordingly, the MB line moves to more negative velocity as
the temperature is increased. The second-order Doppler shift is smallest at absolute zero but will have finite value
because of zero point motion of the nucleus.
In the high-temperature region,
Δv 3κT ⎡ 3κT ⎤
= , ⎢Since v = m ⎥
2
(10.45)
v 0 2m i c 2 ⎣ i ⎦
where, mi is the mass of the ith atom.

Differentiating (10.45) with respect to temperature, we obtain
∂( Δ ) 3v 0κ
= (10.46)
∂T 2m i c 2
Since, NA κ = R
∂( Δ ) 3R 3Rv 0
therefore, = v0 ⋅ = (10.47)
∂T 2N A m i c 2
2M 0c 2
Δv
The behaviour of fractional shift, with temperature is shown in Fig. 10.5.
v0
θ
The second-order Doppler shift in the temperature limit T > D is given by,
3
δ SOD ( ) κT ⎛ θ D2 ⎞
= 1 + (10.48)
Eγ 2M 0c 2 ⎜⎝ 20T 2 ⎟⎠
The shift of the resonance line by a temperature-governed second-order

Doppler effect is termed a thermal red shift. At temperatures, T >> qD, the
value of dS0D(T) for 57Fe estimated from Dulong Petit’s law is 7 × 10−4 mm
(second K). The second-order Doppler shift enables one to calculate the iso- Dn
no
meric shift accurately and is being discussed under temperature effect on
isomer shift.
10.6.3 Nuclear Isomer Shift and Factors Affecting It Temperature

The nuclear isomer shift arises because of the electrostatic interaction Fig. 10.5 A plot between fractional shift
between the nuclear charge (Ze) distribution over a finite radius R and the (Δν/ν0 ) and temperature.
s-electron charge density at the nucleus (−e) |ys(0)|2. The energy shift due to
the charge of potential inside a sphere of radius R is given by
2π 2 2
Z e R ψ s (0 )
2
δE (10.49)
5
Here, Ψs(0) is the wave function at r = 0. This relation is based on the assumption that the radial distance r, i.e.
the distance between the positive and negative charges, is greater than R, i.e. r >> R. Since the size of the nucleus
in the ground and excited states is different, so the electrostatic shift dE will also be different in each nuclear state.
Thus in a transition from a nucleus in the excited to the ground state, the change in energy of g -ray due to this
‘volume’ effect will be
ΔE = (dE)e – (dE)g
2π 2
= Ze | ψ s (0) |2 ( R e2 − R g2 ) (10.50)
5
It is frequently written as
4π 2 2 δ R
ΔE = Ze R | ψ s (0) |2 (10.51)
5 R
The nuclear radius R (cm) = 1.4 × 10−13 (mass number)1/3.
δR
The value of is characteristic of each transition and is of the order of 10−4. It can either be positive or nega-
R
tive. A positive sign signifies that the nucleus shrinks on de-excitation.
The Mössbauer experiment compares the difference in the energy of g -ray between the nuclear transitions in
the source and the absorber (Fig. 10.6). In such a situation, the chemical isomer shift is given by
(d E)e, so (d E)e, abs

Excited
state
Eso Eabs
Eo Eo
Ground (d E)g, so (d E)g, abs

state Absorber
Source
(a)
Relative counts
δIS
− ve O + ve
Doppler velocities
(b)
Fig. 10.6 (a) Shift of the nuclear levels due to the monopole coulombic interaction between
the nuclear and elecctronic charges inside the nucleus. (b) Theoretical MB spectrum of
Fig.10.6 (a) above. (δIS = Eabs – Eso).
δ IS =
4π 2 2 δ R
5
Ze R ⋅
R
ψ ( ψ
Chemical ter
erm
so ) (10.52)
Nuclear term
The theoretical shift dIS is related to the measured shift in the Doppler velocity units v (mm/s) by
⎛ c ⎞
v IS = ⎜ ⎟ δ IS (10.53a)
⎝ Eγ ⎠
⎛ c ⎞
v=⎜ ⎟⎠ Γ cm/s (10.53b)
⎝ Eγ
Thus, from Eqs (10.52) and (10.55), we write
4π 2 2
δR
v IS = ⋅ (| ψ (0) |2 − ψ s (0)so |2 ) (10.54)
5E γ R
In Eq. (10.54) appears two interesting quantities, the radius of the nucleus and the total electron density at the
nucleus. The first information is useful for testing of different nuclear models whereas the second is of interest for
δR
the molecular structure and magnetic properties of solids. If is positive, dIS is positive, i.e. s-electron density
R δR
at the absorber is greater than at the source. For negative value of , the reverse is true. The chemical isomer
R δR
shifts of two different isotopes in the same series of compounds could be explained by the difference in , e.g.
R
127
I and 129I spectra of Na3H2 IO6 (Fig. 10.7). The source matrix is zinc telluride in each case but contains 127Te and
129
Te respectively. These isotopes undergo b decay to populate the 57.6 keV state of 127I and the 27.7 keV state of
δR
129
I. The absorption occurs at positive and negative velocities because is negative in 127I and positive in 129I
R
whereas the s-electron density at the nucleus is greater in the source than in the absorber. The relative magnitude
⎡ δ 127 ⎤ ⎡ δ R 129 ⎤
of the shift also differs because ⎢
⎣ R
I
⎦ ⎣ R
( ) ⎦
( )
I ⎥ = −0.65 .
The line widths are also different for two isotopes, i.e. 2G (127I) = 2.54 and 2G (129I) = 0.5863 mm/s. We know
that all the occupied s-electron orbitals in the atom contribute to |ΨS(0)|2. In addition, it is also affected by the
changes which occur in the outer valence shells. Although the values of |Ψs(0)|2 for p-, d-, and f-electron are zero,
nevertheless, these orbitals indirectly interact with the nucleus via interpenetration shielding of the s-electrons,
e.g. 3d 54s1 configuration have larger value of |ΨS(0)|2 as compared to 3d64s1 since in the latter case the extra
d-electrons shields the 4s electron from the nucleus. The difference in the electron density at the nucleus in differ-
ent chemical environments influences the mean lifetime of the radioactive decay by the relation
1
Δ
τ = α s ⋅ Δ ψ s (0 )
2
(10.55)
1 α t + 1 ψ s (0 ) 2
τ
The internal coefficients as are proportional to the s-electron densities of the various corresponding shells. The
1
chemically induced change in the mean lifetime, i.e. ‘ Δ ’ can be used as a tool for calibrating 57Fe and other
τ
isomer shifts and in principle, the Mössbauer line width should indicate the effect according to Eq. (10.6).
129I
(a)
1.00
Source : Zn129mTe
0.95 Absorber : Na3H2129IO6
Transmission
0.90
d1.s 127I
1.00
(b)
0.98 Source : Zn127mTe

Absorber : Na3H2127IO6
d 1.s
−10 −5 0.0 +5 +10

Velocity (mm /sec)
Fig. 10.7 Chemical isomer shifts of (a) 129l and (b) 127l in the Mössbauer spectra of Na3H2129IO6 and Na3-
H1227IO6 respectively.
It should be kept in mind that the MB spectra of a particular material measured in different sources under
constant conditions will have different isomer shifts because |ΨS(0)|2 changes with chemical environment of the
MB nucleus. Therefore, isomer shift data must be reported with respect to some standard material. The standard
can be the Mössbauer source used in a particular experiment or any absorber material, e.g. in 57Fe MB spectros-
copy 57Co in metal foils of Pd, Pt, Cr, Cu and standard steels used as source will show different isomer shifts.
The commonly used standards in this spectroscopy are metallic a– 57Fe foil and sodium nitroprusidedihydrate;
Na2[Fe(CN)5NO] 2H2O, (SNP); but the metallic iron spectrum is popular as a reference standard for calibration of
chemical isomer shifts. Metallic iron spectrum has six hyperfine fingers (Fig. 10.8) because of which the linearity
of the MB spectrometer can be checked in a better way as compared to SNP spectrum which has only two peaks
(Fig. 10.9) with large separation. Moreover, the zero chemical shift is defined on the basis of metallic iron spec-
trum. The advantage of adopting SNP as standard is that nearly all the 57Fe chemical isomer shifts are positive in
sign. Chemical isomer shifts and quadrupole splitting for a – 57Fe and SNP standards are:
100
95
Relative counting rate
90
85
80
75
−10 −8 −6 −4 −2 0 2 4 6 8 10
Velocity (mm/sec)
Fig. 10.8 Mössbauer absorption spectrum of 57Fe in metallic iron showing the magnetic hyperfine structure.
k
120
110
100
−1.0 −0.5 0 0.5 1.0 1.5 (mm/sec)
Fig. 10.9 Mössbauer spectra of single crystal of sodium nitroprusside cut parallel to the bc plane [100].
57
Co/ a – 57Fe or SNP; dIS = 0.000 mm/sec. Δ = 0.000 mm/sec. For other standards, i.e. 57Co/Cu : SNP: dIS = −0.4844
± 0.0010 mm/sec; Δ = 1.7048 ± 0.0025 mm/sec. a – 57Fe is preferred by physicists whereas SNP is adopted by
chemists.
The reference standard material for Sn is SnTe and BaSnO3, I is CsI, Ir is Ir and Zn is ZnS.
Let us now see how the observed chemical shift is influenced by ligands and coordination number and by
temperature and pressure and what do these effects do?
(a) Effect of Ligands and Coordination Number With high-spin complexes, s-electron density increases
for L → Mσ electron transfer while in low-spin complexes it increases for the M → Lπ-back transfer of electron.
Both these processes helps to interpret isomer shift changes. The other two processes σM → L and πM → L do not
play significant role neither in bonding nor in isomer shift changes. The isomer shift decreases as the covalent
character of the ligand increases, i.e. −H2O> −Cl– > −O22–> −S2– > −CN–, etc. The isomer shifts of some inorganic
compounds have been listed in Table 10.4.
Table 10.4 Effect of ligands and coordination number on Isomer shift (in mm/sec) of Some inorganic compounds.
Measured at 77 K, relative to iron metal at 290 K.
Ion Spin Coordination Compound Ligands Isomer shift (mm/s)
Fe 2 + 2 Octahedral FeSiF6˙ 6H2O 6H2O 1.42

( High spin )
–
FeCl2˙ 4H2O 4H2O, 2Cl 1.34
–
FeCl2˙ 2H2O 2H2O, 4Cl 1.24
–
FeCl2 6Cl 1.20
–
Tetrahedral (NMe4)2FeCl4 4Cl 1.05
FeBaSi4O10 4O2−
0.87
Fe 3+ 5 –
( High spin ) Octahedral FeCl3 6Cl 0.53
2
–
Fe2O3 6O2 0.50
–
Fe-tris-dtc 6S2 0.50
–
Tetrahedral (NMe4)FeCl4 4Cl 0.30
Fe 2+ 0 Octahedral K4Fe(CN)6 6CN

–
0.06
( Low pin )
1 –
Fe 3+
( Low pin ) 2 Octahedral K3Fe(CN)6 6CN −0.03
The isomer shifts are also related to the electronegativity of the ligands, i.e.,
Isomer shift ←
Fe(III)Br3 <Fe(III)Cl3<Fe(III)F3
→ Electronegativity
Isomer shift ←
Fe(II)(CN)5NO <Fe(II)(CN)5CO < Fe(II)(CN)6

Electronegativity ←
Sn X4 (where X = F, Cl, Br, I) shows that as electronegativity increases, isomer shift also increases as expected.
ll9
This behaviour is consistent with the fact that Sn is sp3 hybridised. Quadrupole splitting has been observed only in
case of SnF4 which exists as a polymer in which Sn is octahedrally bounded to four bridging and two neighbouring
fluorine atoms.
In [Fe (CN)6]4− complexes p and s bonds exist between metal atom and the ligand. The cyanide group donates
a pair of electrons to iron atom via s-bond and some of the d-electron density is donated back to the ligand via
dp – pp interaction as shown below (X). Thus replacement of one of the cyanide groups by a group which cannot
form p-bonds, leads to higher value of isomer shift. This is due to the increase
in d-electron density on iron and hence, decrease in s-electron density. The metal
− + + −
ligands s-bond has little effect on isomer shift in this case.
Fe σ C N
Thus, as the p-bond strength of ligand decreases, the isomer shift increases, i.e. (X)
− + − − +
p-decrease: ← p bond strength NO > CO > CN > SO3 > Ph3 P > NO 2 > NH3.
+ − 2−
Isomer shift → NO+ is a good p-acceptor ligand.

[Fe(II) (CN)5H2O]3− (0.31), [Fe(II) (CN)5NH3]3− (0.26) ~ [Fe(II) (CN)5NO2]4− (0.26) >[Fe(II) (CN)5 SO3]5− (0.22)
>[Fe(II) (CN)5 (CO)]3− )0.15) >[Fe(II) (CN)5 NO]2− (0.00)
→ p-bond strength
In general, on the basis of molecular orbital theory, we can say that in series of [Fe(II) (CN)5Xn−](3 + n)− com-
2
pounds, both sL→M and pM→L processes increases the Ψs (O ) and since dR/R is negative for 57Fe-dIS becomes
more negative with increasing sL→M +pL→M in the order H2O < NH3 < NO2 < SO32− < CO << NO+. H2O is a
rather weak s-donor ligand with negligible p-acceptor capability. CO and NO+, however, are known to be strong
p-accpetor ligands. For the [Ru(II) (NH3)5Xn−](n-2)− compounds dI.S becomes more positive with increasing (sL→M
+ pM→L) in the order NH3 < SO2 < CO << NO+ (note that dR/R is positive for 99Ru) series. Similarly in the [Ru X5
(NO)]2− series dIS changes in a more positive direction with increasing (sL→M + pM→L) in the order X = Br− < Cl− <
NCS− < CN−. It is to be noted that, there is close correlation with the ranking of these ligands in the spectrochemi-
cal chemical series.
Thus, isomer shift can be fixed as a criterion to study the nature of chemical bond in complexes.
Although the isomer shifts depend upon the oxidation state and degree of covalency of the iron, it is not always
possible to use them to measure the oxidation state in proteins. However in many cases, the information regarding
the oxidation state of iron can be obtained from the data recorded in Table 10.4.
(b) Effect of Temperature As we have already pointed out that the observed line shift is not solely caused
by the chemical isomer shift but by SOD shift also. The isomer shift is not temperature dependent whereas SOD
shift is temperature dependent because <v2> changes with temperature. However, in the high-temperature region,
dIS becomes temperature dependent due to the thermal expansion of the electron charge density at the nucleus.
At such temperatures,
dIS(T) = dIS (0) + aT (10.56)
where, dIS(T) is the isomer shift at T K. The coefficient a can be estimated from the pressure dependence of isomer
shift.
Lumping together Eqs (10.41), (10.48) and (10.56), we obtain
δc ( ) ⎛ κ α ⎞ δ IS ( ) 3κ
κθ
θ D2
+T − = − (10.57)
Eγ ⎝ 2M c
2
Eγ ⎟⎠ Eγ 40 M 0c 2T
cδ c (T ) ⎛ κ αc ⎞ δ ( )c 3κθθ D2
or +T − ⎟ = IS −
Eγ ⎝ 2M c Eγ ⎠ Eγ 40 M 0cT
δ IS ( )
The plot between the left-hand side of Eq. (10.57) against 1/T will be a straight line. From the intercept ,
Eγ
one can determine the value of isomer shift at 0 K and from the slope, the value of qD, the Debye temperature can
be estimated. Once dIS (0) is known, one can determine accurately, the value of chemical shift at any temperature
from Eq. (10.56), dSOD from Eq. (10.48) and dc(T) from Eq. (10.41).
The plots between (i) isomer shift and temperature (°C), and (ii) left-hand side of Eq. (10.57) and 1/T (K–1)
are shown in Figs (10.10) and (10.11) respectively. The sudden change in the isomer shift (Fig. 10.10) and centre
shift (Fig. 10.11) at the Curie temperature (Tc) (ferromagnetic substances become paramagnetic at this temperative).
Tc < Tm (i.e. melting point) may be due to (i) sudden change in volume at Tc or phase transition, i.e. change in
PbTiO3 Sn119 vs BaSnO3
0.10
Isomer shift (cm/s)

0.00
−0.10
−0.20
−0.30
0 120 240 360 480 600 720

Temperature (⬚C)
Fig. 10.10 Isomer shift in PbTiO3: 57Fe Mössbauer spectrum as a function
of temperature, 0–800°C.
0.06
540°C 450°C
(cm/s)
0.05
Eg
αc
−
2M0c
3K
+T
0.04
δ c(T)c
Eg
Tc = 480°C
0.03
0 0.10 0.12 0.14 0.16 0.18 0.20 0.21 × 10−2
I/T (K−1)
δ c ( )c ⎛ κ αc⎞ 1
Fig.10.11 Plot of +T − ⎟ against for PbTiO3 probed with 57Fe.
Eγ ⎝ 2M c Eγ ⎠ T
nature of chemical bond, and (ii) change in the Debye temperature (qD) occurring at the transition or disappear-
ance of Cochrane-type soft mode at Tc. (Transverse optical mode in ferroelectrics is anharmonic in nature and is
called Cochrane type soft mode. Because of this mode, the crystal becomes unstable). Further, it follows from
Fig. (10.11), that the Debye temperature falls from ~ 500 K above Tc to 225 K below Tc. The curve also exhibits
a discontinuous change of about 0.075 mm/s in the vicinity of the Curie temperature. Assuming that the sudden
break in the curve is due to the change in volume at Tc, we may express the behaviour as
∂E γ / E γ ⎛ ΔV ⎞
= (5.37 + 0.07) × 10−12 ⎜⎝1 − V ⎟⎠ (10.58)
∂V / V
ΔV
where, is the relative change in volume. It is estimated that sudden change in isomer shift at transition tem-
V
perature is about 0.01 mm/s. The change in isomer shift due to change in nature of the chemical bond at Tc is also
expected to be very small in view of the pseudo cubic transformation at Tc. Thus the origin of discontinuous in dC
at Tc may be due to some other factors rather than isomer shift. The three factors which can result in a discontinu-
ous change in dc close to Tc are (i) change in Debye temperature on two sides of Tc, (ii) Cochrance-type mode is
temperature dependent, and (iii) the transverse acoustic energy is temperature dependent. It is possible to estimate
the change in dc because of the change in qD. The observed changes in qD below and above Tc give rise to change
in dc of about 0.012 mm/s. These results reveal that the observed sudden change in dc at Tc is due to (i), disappear-
ance of soft mode or, and (ii) the transverse acoustic energy depends upon temperature. However it is difficult to
estimate the contributions due to these two factors.
(c) Effect of Pressure The chemical isomer shift decreases with the growth of pressure. Extreme pressures (up
to 200,000 atmospheres; 1 atm = 1.01325 × 105 Pa = 1.01325 bar) will compress the material (in the context of
57
Fe), as a consequence the overlap between the inner s-shells of iron and the valence orbitals of the ligands will
increase. The induced overlap due to pressure results in increase in s-electron density (i.e. decrease in volume)
at the nucleus and a decrease in the chemical isomer shift. The decrease in isomer shift with increase of pressure
has been observed in a large number of compounds and the effect is fully reversible on release of the pressure.
Theoretically, the effect has been fully confirmed on the basis of wave function calculations in the case of KFeF3.
The behaviour of isomer shift with volume is given by
∂δ IS ( )
= 0.14 cm/s (10.59)
∂ ( /V 0 )
The plot between isomer shift for 57Fe in Fe and pressure is shown in Fig. 10.12 and can be predicted from
Eq. (10.59).
Isomer shift (cm/s)
(V/Vo)/P
Fig. 10.12 Isomer shift of Fe in Fe as a function of volume.

57
It is to be noted that at low pressures, iron has bcc struc-

ture and is ferromagnetic. At about 130 kb occurs a first-order
phase transition to non ferromagnetic hcp structure. However,
Isomer shift change (cm/s)
the behaviour of isomer shift as a function of volume for 57Fe

embedded in nickel, copper and palladium which have fcc struc- (a)
ture and 57Fe in hcp phases in titanium, cobalt and iron cannot
be predicted from Eq. (10.59). The predicted value is compara-
tively more than the expected one as shown in Fig. 10.13. (b)
The most interesting aspect of pressure studies is that iron

in many compounds shows a change in oxidation state or elec-
tronic configuration at high pressures. Only a fraction of total
iron is changed and the effect is fully reversible on the release
V/Vo
of pressure, e.g. high-spin iron (III) in Fe2(SO4)3 and FePO4
Fig. 10.13 The behaviour of isomer shift for 57Fe in nickel,
shows a partial conversion to high-spin iron (II); low-spin iron
copper and palladium with volume, (a) experimental, and
(III) in K4Fe(CN)6 can change to low spin iron (II). The change (b) theoretical.
in oxidation state occurs because of electron transfer from
ligand to metal, i.e. L →
e
M.
The foregoing treatment of chemical shift indicates that it depends on the electronic environment as well as the
parameters, viz., temperature, pressure, etc., influencing the electronic environment around the nucleus. Thus, it
provides very useful information about the bond properties, valency and oxidation state of a Mössbauer atom.
10.6.4 Partial Isomer Shift

The concept of partial isomer shift (p.i.s.) stemmed from the observations that (i) isomer shifts vary linearly with
average electronegativities of surrounding ligands. This has been observed in many iron and tin compounds.
(ii) There exists a characteristic bonding distance between the metal atom and a given ligand in a variety of related
compounds.
The partial isomer shift is an additive property. The isomer shift of a Mössbauer atom in a given compound
is the sum of the p.i.s. values of all coordinated ligands, i: dIS = ∑ .i .s .)i . The concept of p.i.s. holds good for
i
Fe(II) and Sn(IV) compounds. p.i.s. for ligands in low-spin Fe(II) compounds is listed in Table 10.5. With these
values and using the model,
6
dIS = 0.16 + ∑ ( )i (10.60)
i =1
the isomer shift for low-spin Fe(II) compounds can be predicted. This model is partially useful for predicting the
isomer shifts of low-spin Fe(II) compounds which only exist as unstable intermediates. Alike isomer shifts in
Fe(II) compounds, p.i.s. also decrease as the s-bonding and p-back bonding increase.
Table 10.5 Partial Isomer Shift (in mm/s) values (with respect to sodium nitroprusside at 295 K) for
ligands in low-spin Fe(II) compounds.
Ligand p.i.s. Ligand p.i.s. Ligand p.i.s.
NO+ –0.20 CN– 0.01 phen/2* 0.07
H– –0.08 SnCI3–1 0.04 py* 0.07
SiH3– –0.05 Pc/2* 0.05 NH3 0.07
ArNC* 0.00 NCS– 0.05 N3– 0.08
MeNC 0.00 NCO– 0.06 H2O 0.10
EtNC 0.00 depe/2* 0.06 Cl– 0.10
CO 0.00 bipy/2* 0.06 Br– 0.13
I– 0.13
ArNC = p-methoxyphenylisocyanide; pe = phthalocyanine; depe = bis(diethylphosphino) ethane; bipy = 2, 2′-bipyridyl,

phen = 1, 10-phenanthroline; py = pyridine
Problem 10.1: Using the data in Table 10.5 on p.i.s., predict the isomer shifts for cis-FeCl2 (ArNC)4 and,
Na3[Fe(CN)5NH3].H2O and trans-FeCl (SnCl3) (depe)2.
Solution The ligands in cis-FeCl2 (ArNC)4 are Cl– and ArNC. From Table 10.5, we write
Cl– = 0.10 mm/s, ArNC = 0.00 mm/s.
From Eq. (10.60) we obtain
2
dIS = 0.16 + ∑ ( p .i .s .)i = 0.16 + 2 Cl– + 4 × ArNC
i =1
= 0.16 + 2 × 0.10 = 0.36 mm/s (dIS (obs) = + 0.28 mm/s)

Similarly, for Na3[Fe(CN)5NH3]H2O
dIS = 0.16 + 5 × CN– + 1NH3
= 0.16 + 5 × 0.01 + 0.07 = 0.28 mm/s (dIS (obs) = 0.26 mm/s)
and for trans-FeCl(SnCl3) (depe)2
dIS = 0.16 + Cl–1 + SnCl–3 + 2 depe
= 0.16 + 0.10 + 0.04 + 2 × 0.12 = 0.54 mm/s (dIS (obs) = 0.55 mm/s)
10.7 INTERNAL OR EFFECTIVE MAGNETIC FIELD

The effective magnetic field acting on the nucleus arises from the atom’s own electrons and is termed as inter-
nal magnetic field, Hint. The main contributions to the internal magnetic field are (i) the interaction between the
nucleus and electron density of the unpaired s-electrons, i.e. Fermi contact interaction, e.g. an unpaired electron in
the atomic environment generates a local magnetic field of about 2T, and (ii) orbital magnetic moment on the reso-
nant atom. In the high-spin Fe2+ compounds the contribution because of this factor varies from +20T to + 60T and
is opposite in sign to Fermi contact interaction contribution. (iii) Doppler interaction of the electron spin moment
of the atom with the nucleus. In high-spin Fe2+ iron, the contribution due to this interaction is in the range 0–8T
and is zero in cubic symmetry. The relative contributions of these interactions depend on the electron configura-
tion of the atom. The most important contribution is because of Fermi contact interaction. It is to be noted that
unpaired spin density appears even in s-shells of transition metal ions. This is so because the exchange interaction
between the spin-up polarized d shell and a spin-up s-electron is attractive whereas the interaction between the
d-shell and a spin down s-electron is repulsive. Consequently, the radial parts of the two s-electron wave functions
are distorted, one being pushed closer to the nucleus and the other being pulled further out.
10.7.1 Effect of Temperature on Internal Magnetic Field

The internal magnetic field is influenced significantly by temperature. The behaviour of internal magnetic field
with temperature in case of FeF2 is shown in Fig. 10.14. There is a steep growth of field at first with the decreas-
ing temperature and a saturation value is attained at absolute zero. In the neighbourhood of Neel temperature, the
hyperfine effective magnetic field is given by the usual function, HT = H0D(l – T/TN)b. For FeF2, D = 1.36 ± 0.03,
b = 0.325 ± 0.005 and the hyperfine field at absolute zero H0 = 329 ± 2 kG. TN = 78.12 K.
10.7.2 Effect of Pressure on Internal Magnetic Field

The total spin-density imbalance is also affected by change in volume, so that there is a small effect on an internal
magnetic field. The behaviour of internal magnetic field of 57Fe in iron, cobalt and nickel as a function of pressure
350
300
250
200
Hint (kG)
150
100
50
0
0 10 20 30 40 50 60 70 80 90
Temperature (K)
Fig. 10.14 The internal magnetic field in FeF2 as a function of temperature. Note that the MB spectrum of FeF2 exhibit more than six
Zeeman lines. This implies that the Zeeman levels are no longer pure MI > eigen-states, so the ΔMI = ±2 transitions are weakly allowed.
At 4.2 K where the magnetic field is considerably greater than the quadrupole coupling, the spectrum approximates to a six line pattern.
Co
Hint
H
Ni
Fe
0 50 2.50
Pressure
Fig. 10.15 Internal magnetic field of 57Fe in Fe, Ni and Co as a function of pressure.
is shown in Fig. 10.15. The decrease in internal magnetic field with pressure for iron agrees well with NMR
measurements. In case of nickel, the decrease is at a smaller rate as compared to iron. In case of Co, the internal
field first increases and then decreases with rise of pressure.
10.7.3 Effect of External Magnetic Field

The field at the nucleus is parallel (+) or anti-parallel (−) to the direction of external magnetic field. Thus, the total
field experienced by the nucleus in the former case will be H + Hint and Hint − H in the latter case.
10.8 MAGNETIC HYPERFINE INTERACTION

The splitting of the nuclear state with spin I (I > 0) into (2I + 1) sublevels because of the interaction of nuclear
magnetic dipole moment m with a magnetic field (internal or external) at the site of the nucleus is called the
nuclear Zeeman effect.
The hyperfine Hamiltonian operator for a magnetic interaction of nuclear level with spin angular momentum
I and magnetic field H is given by
Iˆ ˆ
Hm = −g β IH (10.61)
⎡ μ ⎤ eh
Here, gN is the nuclear Lande splitting factor ⎢ g N = I ⎥ ; bN is the nuclear Bohr magneton ( = 5.04929
⎣ Î β N ⎦ 4π mp
× 10–27 Am2 or J T–1), H is the strength of magnetic field. If the magnetic field is acting along Z-axis, it becomes
Iˆ ˆ z
Hm = −g β IH (10.62)
The resulting Hamiltonian matrix is diagonal with (2I + 1) eigenvalues, given by

E = −g b H . M (or –m HM /I)
m N N I I I
(10.63)
Here, MI is the magnetic quantum number and can take on values; I = I − 1, I − 2, ...., −I. Em corresponds to
minimum value of mH which is negative.
1
It can readily be shown that the sublevels are equally spaced with a separation of gN bNH. In case of 57Fe, I g =
2
3
for the ground state and I e = for the 14.4 keV first excited state. For this isotope mg = + 0.0902 nm and me =
2
μe
−0.1547 nm. Thus = −1.7145.
μg
The magnetic splitting in mm/s of 57Fe ground state and first excited state are gg = 3.9098, ge = 2.2342 at 298 K
1 3
and gg = 4.0117, ge = 2.2931 at 4.3 K. For 119Sn, I g = , I e = , mg = −1.047 nm and me = +0.67 nm. The ground
2 2
state splits into two ⎛⎜ 2 × + 1 2⎞⎟ sublevels separated by an amount b = ΔEm(g) = gg bNH and the excited state
1
⎝ 2 ⎠
⎛ 3 ⎞
splits into four ⎜ 2 × + 1 4⎟ sublevels with an equal separation of a = ΔEm(e) = ge bNH. According to the dipole
⎝ 2 ⎠
radiation selection rule, ΔMI = 0, ±1, the number of permissible transitions are restricted to six at the most, and
the energy of each transition is listed in Table 10.6.
Table 10.6 Relative energy positions and intensities for various allowed transitions in 57Fe.
Transition ΔMI Relative energy with Total Angular dependence

respect to E0 intensity
1 3 –1 1 3 9 2
− →− − ( β + 3α ) (1 cos θ )
2 2 2 4
1 1 0 1 2 3 2
− →− − (β + α ) sin θ
2 2 2 2
1 1 +1 1 1 3 2
− →+ − (β − α ) (1 cos θ )
2 2 2 4
1 1 –1 1 1 3
+ →− + (β − α )
2
(1 cos θ )
2 2 2 4
1 1 0 1 2 3 2
+ →+ + (β + α ) s in θ
2 2
2 2
1 3 +1 1 3 9
+ →+ + ( β + 3α ) 2
(1 cos θ )
2 2 2 4
The magnetic hyperfine splitting and the resultant spectrum is shown in Fig. 10.16. The sepration between the
outermost two lines, 3 a + b or 4.715 a is found to be 10.64 mm/s, which corresponds to an effective magnetic
field of 330 kG. The selection rule for magnetic quantum number (i.e. ΔMI = 0, ±1), if written in the following
fashion, constitutes the polarization values for Zeeman lines.
(a) View Perpendicular to the Field ΔMI = ±1, plane polarised parallel to the magnetic field direction; p-
components: ΔMI = 0, plane polarized, perpendicular to the magnetic field direction; s-components.
(b) View Parallel to the Field ΔMI = ±1, circularly polarised, p-components.
(c) Right Circularly Polarised Spin of g -ray lies in the direction of g -emission.
(d) Left Circularly Polarised Spin of g -ray lies in the opposite direction of g -emission. ΔMI = 0; forbidden,
s components.
These conventions pertaining to polarization in MB spectroscopy are opposite to the one in optical
spectroscopy.
Accordingly, for random orientation of H with respect to the g -ray direction, the radiation is unpolarized. For
a perpendicular magnetic field H(q = 90°), the six lines are observed with linear polarization ||, ⊥, ||, ||, ⊥, ||. For
axial field (i.e. q = 0), the radiation is circularly polarized with ΔMI = 0 transition missing and the remaining four
lines are observed with polarization left, right, left, right.
The nuclear magnetic moment of the ground and excited states of 57Fe are of opposite signs, i.e. positive for
the ground state and negative for the excited state. This change in sign of magnetic moment results in relative
inversion of the multiplets. The lines are not of equal intensity but have 3:2:1:1:2:3 ratio. However, the spec-
trum is symmetrical about the centroid. Such spectra have been observed in polycrystalline samples of 57Fe
H not = 0, Uzz = 0 M1
+ 3/2
H=0 α = ΔEm(e)
+ 1/2
1 = 3/2 − 1/2
− 3/2
− 1/2
b = ΔEm(g)
1 = 1/2
+ 1/2
Isomer shift 1 2 3 4 5 6
Magnetic dipole splitting
(a)
Isomer shift
1 2 3 4 5 6
It
E
Velocity (mm/s)
(b)
Fig. 10.16 Magnetic dipole splitting (H ≠ 0, Uzz = 0) without electric quadrupole
(a), and the resultant theoretical spectrum (b) of Fe57.
and 119Sn. Combined magnetic and quadrupole splittings are also possible in some samples. However to negate
the impact of quadrupole splitting, the source material is selected to give a single-line spectra, e.g. when 57Co
is embedded in metallic chromium, no net electric field gradient and magnetic fields are present at the cobalt
nucleus.
10.8.1 Intensity of Hyperfine Lines

The intensity of a particular hyperfine transition between sublevels is determined from the interaction of two
nuclear angular momentum states. It is expressed as the product of two terms which are angular dependent and
angular independent. The angular-dependent term averages to unity when all the orientations are equally prob-
able, e.g. in a randomly oriented polycrystalline sample. The intensity of the angular-dependent term is given by
the square of the appropriate Clebsh–Gordan coefficient, i.e.
intensity ∝ < Ig J – m1m⏐Iem2 >2 (10.64)
where the two nuclear spin states Ig and Ie have Iz values of m1, and m2 and their coupling obeys the vector sum
J = Ie + Ig and m = m1 − m2. J is called the multipolarity of the transition and the intensity is greater if J is small.
J = 1 for dipole transition and J = 2 for quadrupole transition. If there is no change in parity during the decay
it is classified as magnetic dipole (M1) or electric quadrupole (E2) and for a change in parity, the transition is
classified as electric dipole (E1). The angular-dependent term q (J, m) is expressed as the radiation probability in
a direction at an angle q to the magnetic field axis. q is the angle between the direction of the magnetic field at
the nucleus and the direction of propagation of g -ray. The angular-dependent probability functions q (J, M) are
given in the tabular form as foot note (2) under Table 10.8.
The intensity for polycrystalline samples is obtained by integration over all orientations to obtain average
3 2 1 2π π ⎛ 3 2 ⎞
θ (JJ m), e.g. sin θ ∫ ∫ sin θ sin
i ϕ dθdϕ = 1 (10.65)
2 4π 0 0 ⎝ 2 ⎠
and the total emitted radiation given by ∑<I 1 J − m1m I 2 m 2 > 2 θ (J , m ) is independent of q i.e. the g -ray emis-
m1m 2
sion is isotropic and is normalised to unity, i.e.

1
∑ 4<I 1 J − m1m I 2 m 2 > 2θ (J , m ) = 1 (10.66)
m1m 2
The Clebsch–Gordan coefficients and angular-dependence functions needed to compute the relative intensities of
3 1 7 5
hyperfine lines for − and − transitions with M1 multipolarity have been given in Tables 10.7 and 10.8
respectively. 2 2 2 2
3 1
Table 10.7 Relative probabilities for a , transition.
2 2
Magnetic spectra (M1)
m2 –m1 m C C2 q(J, m) q2 = 0º q = 90º

(1) (2) (2) C2q (3) C2q (3)
3 1 1 3
+ + +1 1 (1 cos 2 θ ) 6 3
2 4
2 4
1 1 1 3 2
+ + 0 s in θ 0 4
2/3 2
2 2 6
1 1 1 3
− + –1 (1 cos 2 θ ) 2 1
1/ 3 4
2 2 12
3 1
− + –2 0 0 0 0 0
2 2
3 1
+ − +2 0 0 0 0 0
2 2
1 1 1 3
+ − +1 1/ 3 (1 cos 2 θ ) 2 1
12 4
2 2
1 1 1 3 2
− − 0 2/3 s in θ 0 4
2 2 2
6
3 1 1 3
− − –1 1 (1 cos 2 θ ) 6 3
2 2 4 4
Quadrupole spectra (MI ) when h = 0
Transition C2 q (J, m) q = 0º q = 90º

(2) (2) C q (3)
2
C2q (3)
1 1 1 1 3 2 1 5
± ,± + sin θ
2 2 2 2 4 4 8
3 1 1 3 3 3
± ,± (1 cos 2 θ )
2 2 2 4 4 8
Table 10.8 Relative probabilities for a 7/2, 5/2 transition.
Magnetic spectra (M1)
m2 –m1 m C C2 q(J, m) q = 0º q = 90º

(1) (2) (2) C q(3)
2
C2q(3)
7 5
+ + +1 1 21⎫ 42 21
2 2 ⎪
⎪
⎪
5 3 5 ⎪
+ + +1 15 ⎪ 30 15
2 2 7
⎪
⎪
⎪
3 1 10 ⎪
+ + +1 1+cos2q 20 10
2 2 21 ⎪
10 ⎪
⎬
1 1 2 ⎪
+ − +1 6 ⎪ 12 6
2 2 7 ⎪
⎪
⎪
−
1
−
3 1 ⎪
2 2 +1 7 3 ⎪ 6 3
⎪
⎪
3 5 1 ⎪
− − +1 1 ⎪ 2 1
2 2 21 ⎭
⎫
5 5 2 ⎪
+ + 0 0 12
2 2 7 6 ⎪
⎪
⎪
⎪
3 3 ⎪
+ + 10 ⎪
2 2 0 0 20
21 10⎪
⎪
⎪
⎪
1 1 4 ⎪
+ +
2 2 0 7 ⎪ 2 sin2q 0 24
⎬
12⎪
⎪
1 1 4 ⎪
− − 0 0 24
2 2 7 ⎪
12⎪
⎪
3 3 10 ⎪
− − 0 ⎪ 0 20
2 2 21 10⎪
⎪
⎪
5 5 2 ⎪
− −
2 2 0 7 6 ⎪⎭ 0 12
Table 10.8 Relative probabilities for a 7/2, 5/2 transition—Cont’d.
Magnetic spectra (MI )
m2 –m1 m C C2 q(J, m) q = 0º q = 90º
(1) (2) (2) C2q(3) C2q(3)
3 5 1
+ + –1 21
1 ⎫ 2 1
2 2 ⎪
⎪
⎪
1 3 1 3 ⎪
+ + –1
⎪
1 + cos2q 6 3
2 2 7
⎪
⎪
1 1 2 6 ⎪
− + –1 ⎪ 12 6
2 2 7 ⎪
⎬
3 1 10 ⎪
− − –1 10 ⎪ 20 10
2 2 21 ⎪
⎪
5 3 5 ⎪
− − –1 15 ⎪ 30 15
2 2 7 ⎪
⎪
7 5 ⎪
− − –1 1 21⎪⎭ 42 21
2 2
Quadrupole spectra (M1)
Transitions C2 q(J, m) θ=0 θ = 90’

(2) (2) C2 q(3) C2 q(3)
5 7
± ,± 21 1 + cos2q 42 21
2 2
5 5
± ,± 6 2sin2q 0 12
2 2
5 3
± ,± 1 1 + cos2q 2 1
2 2
3 5
± ,± 15 1 + cos2q 30 15
2 2
3 3
± ,± 10 2sin2q 0 20
2 2
3 1
± ,± 3 1 + cos2q 6 3
2 2
1 3
± ,± 10 1 + cos2q 20 10
2 2
1 1
± ,± 18 2 12 30
2 2 + sin2q
3
1
31 1 3 + m1 4
1. The Clebsch−Gordon coefficient: For < m2 1 2m > is < , 2 − m1m m 2 > = ( ) 2
22 2 2 3
5
31 7 5 5 7 + m1 8 75
< m 2 m1 2 m > . For < m2 1 1m > is < 1 − m1m m 2 > = ( ) 2 < m 2 m1 1m > .
22 2 2 2 2 3 22
2. The angular dependent probability functions q(J. M) are
m\j 0 ±1 ±2
1 2 1
1 sin θ (1 cos 2 θ ) −
2 4
3 2 1 1
2 sin 2θ (cos 2 θ coss 2 2 ) (sin 2 θ sin 2 2θ / 4)
8 4 4
3. C2 and q(J, m) are the angular independent and angular dependent terms normalized to a total radiation
probability of
∑ C θ (J , m ) = 1
m1 m 2
2
4. Relative intensities are observed at 0° and 90° to the principal axis: normalisation is arbitrary.
1 3
The theoretical stick spectra for → transition deduced from (2I + 1) multiplicity rule; ΔMI = 0, ±1 selection
2 2
rule and the probability of the transition given in Table 10.7 are shown in Figs 10.17(a)−(c) whereas, the observed
1 3
experimental spectra for → transition for a −Fe foil at room temperature with hyperfine magnetic field of
2 2
Hint = 330 kOe are shown in Fig. 10.18(a)−(c).
3 3
2 2
1.00
1 1 0.98
0.96
(a) 0.94
3 3 0.92 (a)
1.00
0.98
Relative Transmission
1 1
0.96
0.94
(b)
0.92
4 4 0.90
3 3 0.88 (b)
1.00
0.98
1 1
0.96
0.94
(c)
0.92 (c)
Fig. 10.17 MB stick spectra (a) for a randomly ori-
ented crystalline sample, when no external field is 0.90
−8 −6 −4 −2 0 2 4 6 8
applied, (b) q = 0, a magnetic field is applied parallel
to the direction of the internal magnetic field of the Velocity (mm/s)
sample, and (c) q = 90°, a magnetic field is applied Fig. 10.18 Experimentally observed MB spectra of a − 57Fe foil at
perpendicular to the direction of the internal mag- room temperature (a) H = 0, Random (b) H = 50 kOe, q = 0 and (c)
netic field of the sample. H = 3.5 kOe, q = 90°.
Further, at very low temperatures, T << 1 K, the magnitude of magnetic hyperfine splitting and the thermal
energy κT are of the same order. Thus, the populations of the nuclear levels are governed by Maxwell–Boltzmann
distribution law. Accordingly, the higher levels will be thinly populated and transitions from these levels will be
less frequent. Consequently, the spectra become asymmetric with respect to relative line intensity. This effect is
observed for the parent isotope 57Co (I = 7/2) which has a large magnetic moment and the polarisation is main-
tained through the decay to 57Fe.
10.8.2 Sign of Nuclear Zeeman Interaction and Internal Magnetic Field

The spectrum in Fig. 10.18(b) reveals that when large magnetic field is applied perpendicular to the foil plane (q
= 0) in excess of the demagnetisation field (≈21 kOe), the splitting is reduced by the applied field. This indicates
that hyperfine interaction has a negative sign, i.e. atomic moment and the magnetic field at the nucleus (Hint)
are in opposite directions, e.g. 57Fe in most alloys and ferrites. Thus, the sign of internal magnetic field can be
determined by applying external magnetic field (H = 30 − 50 kG). When the internal field is aligned parallel to
external field, i.e. magnetization, the internal field is increased to Hint + H, then the sign of the internal field is
positive. In the antiparallel alignment, i.e. demagnetisation, a field of Hint − H indicates a negative sign. However,
this method fails when magnetic interaction anisotropy of the matrix is so large that it prevents the rotation of the
internal field into the direction of the applied field, i.e. it prevents complete polarisation of the ordered spins by
the external field.
10.8.3 Relaxation Effects in Relation to Magnetic Hyperfine Splitting

Further it is to be noted that the magnetic hyperfine interaction depends upon the magnetic nature of the sample,
viz., paramagnetic, ferromagnetic or antiferromagnetic. For paramagnetic samples, the time lag tf elapsed between
two successive flips of the electron spin needs to be considered. The flips could be because of electronic relaxation
1
or above Curie point, due to exchange coupling between neighbouring electron spins. When the frequency of
1 τf
the flips is >> than the Larmor frequency of the nuclear spin in the internal field, there will be no magnetic
τL
hyperfine pattern, since the value of the internal field experienced by the nucleus averages out to zero over the time
of the frequency of the nuclear spin. On the other hand, if the electron-flips frequency is << than the frequency
of the nuclear spin, the spectrum will exhibit a magnetic hyperfine structure, e.g. Fe and rare earth Mössbauer
nuclei. When the relaxation time of the electron spin (spin-spin and spin-lattice) tf = tL, complicated spectra with
broad lines are observed, which enables us to determine the electron relaxation time. The line broadening in the
Mössbauer–Zeeman pattern is closely connected to the motional narrowing in NMR. In case of ferro-, ferri-, and
antiferromagnetic substances, the electron s-s interaction is >> than the nuclear Zeeman splitting, consequently
the nuclear spin interacts with the average value of the internal magnetic field and the MB spectrum exhibits a
magnetic hyperfine pattern. From the foregoing description, it is pertinent to note that magnetic dipole interac-
tions are rarely encountered in chemical applications of Mössbauer spectroscopy. Metallic iron is an exception.
However, the magnetic dipole interactions provide a technique for the study of magnetically ordered materials,
sublattice magnetisation, Curie or Neel temperature and atomic magnetic moments can be deduced from the
magnetic hyperfine field.
The magnetic hyperfine pattern in paramagnetic substances serves as a tool for the study of spin relaxation
time and the nuclear Lande g-factor. The two parameters which give information on the magnetic properties of
the material under study are (i) the position of the six fingers gives information on the splitting of the excited and
ground states, and (ii) the relative intensity of the lines gives information on the polarisation. The magnetic split-
ting in a-iron (in mm/s) is determined from the appropriate first differences of line positions.
The ground state splitting is
ΔE (g ) = g g = v4 − v2 (10.67a)
= v 5 − v3 ;
and the excited-state splitting is
ΔE ( ) = g e = v3 − v 2 (10.67b)
= v 5 − v4
1 ⎡1 1 ⎤
δ IS (mm/s) ( v1 + v 6 ) + ( v 2 + v 3 + v 4 + v 5 ) ⎥ (10.68)
2 ⎣2 4 ⎦
1
Problem 10.2: For 57Fe, me = −0.15 nm, Ie = 3/2, mg = 0.09 nm and I g = . Prove that the separation between
2
any two consecutive excited magnetic hyperfine levels is small as compared to their counter parts in the ground
state.
μI H M I
Em = − (10.69)
I
Let a and b be the separations between any two consecutive Zeeman levels in the excited and ground states
respectively.
Applying the selection rule for permissible transitions, i.e. ΔMI = 0, ±1, we get
μe H
α=+
Ie
μg H
β=+
Ig
α ⎛ μe ⎞ ⎛ I g ⎞
So = (10.70)
β ⎜⎝ μ g ⎟⎠ ⎜⎝ I e ⎟⎠
Substituting the values of the respective parameters, we get
α ⎛ −0.15( nm) ⎞ ⎛ 1 2 ⎞ −5
= × = = −0 55
β ⎜⎝ 0.09( nm) ⎟⎠ ⎜⎝ 2 3 ⎟⎠ 9
or a = −0.55 b
b>a
Problem 10.3: Prove that in the ground state of 57Fe (14.4 keV), the hyperfine magnetic field is given by
⎛ Eγ ⎞ ⎛ − g g I g ⎞
H =⎜
⎝ cβN ⎟⎠ ⎜⎝ μ g ⎟⎠

MI
Em μH (Eq. 10.63)
I
1
For ground state of 57Fe, I g = .
2
1
Therefore, MI = ±
2
Thus, the energy difference between the Zeeman levels corresponding the ground state is
− μg H
ΔE (g ) = gg = (10.71)
Ig
I g gg
H=− (10.72)
μg
E
If gg is in mm/s then gg γ is in ergs, also if mg is in units of nuclear magneton, bN.
Then c
⎛ Eγ ⎞ ⎛ − g g I g ⎞
H =⎜ (10.73)
⎝ cβN ⎟⎠ ⎜⎝ μ g ⎟⎠
Now, Eg = 14.4125 keV

bN = 5.04929 × 10−27 J T−1
1 eV = 1.6021 × 10−9 J
Substituting there values in Eq. (10.73), we get
⎛ ggI g ⎞
H = − (15.250 kOe s/mm nm) ⎜ (10.74)
⎝ μ g ⎟⎠
The hyperfine field could also be derived from the excited state splitting ge. Since this method is inherently less
accurate and non-independent, ge is not used to calculate H.
10.9 ELECTRIC QUADRUPOLE INTERACTION

Recall from Chapter 8 on nuclear quadrupole resonance spectroscopy that when a crystal has low symmetry or its
∂ 2U
high symmetry is broken locally, an electric field gradient UZ Z = 2 = eq will result at the site of the nucleus and
∂z
interact with the electric quadrupole moment eQ of the probe isotope nucleus, causing a splitting of the nuclear
state into substates, e.g. in case of 57Fe a double peak spectrum will appear when the crystal is paramagnetic, and
an asymmetric six-line pattern is observed in case of ferromagnetic ordering. The electric quadrupole coupling in
MB spectroscopy is similar to that in nuclear quadrupole resonance spectroscopy. The main difference is that the
latter deals with radio-frequency transitions within hyperfine multiplets of a nucleus in the ground state whereas
the former is a g -ray transition between hyperfine multiplets of the nucleus in the ground and excited states.
The classical expression for the quadrupole interaction is given by
e 2qQ ⎡ 1 U xx U yy ⎤
HQ = 1− (10.75)
6 ⎢⎣ 2 U zz ⎥
⎦
where terms which are independent of nuclear orientation have been neglected.
The quantum mechanical analogue of the classical quadrupole Hamiltonian is
eQU zz
HQ = ⎡3 2
− I 2 + η(I +2 + I −2 ) / 2⎤⎦ (10.76)
4I ( 2I − 1) ⎣
z
where, Iˆz is the Z-component of the nuclear spin operator Iˆ and Iˆ+ and Iˆ– are the raising and lowering operators
defined as
Iˆz Iˆx ± i Iˆy .
The eigen value of the operator Iˆ 2 is I(I + 1)
The eigen values of the Hamiltonian HQ, i.e. the energies of the sublevels are given by
1/ 2
eQU zz ⎛ η2 ⎞
EQ = ⎡⎣3M I2 − I (I + 1) ⎤⎦ ⎜1 + ⎟ (10.77)
4I ( 2I − 1) ⎝ 3⎠
where the nuclear magnetic spin quantum number MI can take on (2I + 1) values of I, I − 1, I − 2, ..., − I. Here the
substates can’t be distinguished by the sign of MI, since Eq. (10.77) contains only the second power of MI. There-
fore, the nuclear substates arising from nuclear quadrupole splitting remain doubly degenerate and are denoted by
( I , ± M I > . The two-fold degeneracy of substates can only be removed by magnetic perturbation. For spheri-
cally symmetric fields (Uxx = Uyy = Uzz) degeneracy corresponding to MI is not removed and no resonance can take
place. For axial symmetrical fields (I= 3/2, 5/2, 7/2, ...,) the energy level splits into a series of Kramer’s doublet
⎛ 1⎞
by ⎜ I + ⎟ rule and for integral spins, i.e. I = 1, 2, 3..., the number of sublevels is given by I + 1. It is to be noted
⎝ 2⎠
that only those transitions among the substates are allowed for which ΔMI = 0, ±1. A lack of axial symmetry in the
EFG introduces matrix elements which are off diagonal with ΔMI = ±2. For 57Fe and 119Sn, Ie = 3/2 and Ig = 1/2. The
excited state will split into two doubly degenerate sublevels whereas the number of sublevels corresponding to the
⎛ 1⎞
ground state will be one only. The nuclear ground state ⎜ I g = ⎟ is not split, because Q = 0. Accordingly, for 57Fe,
⎝ 2⎠
3 3 3 1
the energy difference ΔEQ between the two substates , ± > and , ± > can be worked out as follows.
From Eq. (10.77), we write 2 2 2 2
1/ 2
⎛ 3⎞ U ⎛ η2 ⎞ 1 3
EQ 3eQ zz 1 + ⎟ for I , MI = ±
⎝ 2⎠ 12 ⎝ 3⎠ 2 2
1/ 2
⎛ 1⎞ U ⎛ η2 ⎞ 3 1
EQ
⎝ 2⎠
3eQ zz
12 ⎜⎝1 + 3 ⎟⎠ for I
2
, MI = ±
2
Thus, the quadrupole transition energy is given by
⎛ 3⎞ ⎛ 1⎞
Δ Q = EQ ⎜ ± ⎟ − EQ ± ⎟
⎝ 2⎠ ⎝ 2⎠
1/ 2
⎛ eQU zz ⎞ ⎛ η ⎞
2
=⎜ 1 + (10.78)
⎝ 2 ⎟⎠ ⎜⎝ 3 ⎟⎠
1/ 2
1⎛ c ⎞ 2 ⎛ η2 ⎞
Δ Q ( )= ⎜ ⎟⎠ e qQ ⎜⎝1 + 3 ⎟⎠
2 ⎝ Eγ
For h = 0, Eq. (10.78) becomes
eQU zz eQeq e 2Qq
Δ Q=
ΔE = = (10.79)
2 2 2
1⎛ c ⎞ 2
Δ ( )= ⎟⎠ e qQ
2 ⎜⎝ Eγ
Q
Note that ΔEQ is half the quadrupole coupling constant defined in NQR spectroscopy. Further, it is always
better to refer to the sign of e2qQ or of q since in the term eq, it is not always mentioned whether e is taken the
charge of electron or the proton. The observed sign of e2qQ may be an important factor in deciding the origin
of the EFG. As we know that the quadruple moment Q for 57Fe is positive so the quadrupole coupling constant
e 2qQ
will be positive or negative according as q, the Z-component of EFG, is positive or negative. In these two
2
⎛ 3⎞
possible situations, the excited-state sublevels interchange. The energy-level diagram for 57Fe ⎜ I e = + ⎟ exhibit-
⎝ 2⎠
ing the quadrupole splitting and the resultant Mossbauer spectrum under different possible situations is shown in
Fig. 10.19.
10.9.1 Intensity of Nuclear Quadrupole Doublet

The quadrupole split spectrum of 57Fe consists of two lines. The intensity of the lines can be predicted from the
Clebsch-Gordon coefficients and angular dependance functions listed in Table 10.7. The angular dependence of
intensity ratio when field gradient is axially symmetric, i.e. h = 0 is given by
⎛ ⎞
I ±1/ 2, ± 3 / 2 I ⎜ 1 + cos 2 θ ⎟
= = (10.80)
I ±1/ 2, ±1/ 2 I σ ⎜ 2 ⎟
⎜ + sin 2 θ ⎟
⎝3 ⎠
where, q is the angle between the Uzz axis and direction of g -emission.
The intensity ratio ranges from 3(q = 0°) to 0.6 (q = 90°).
It may be pointed out that the above expression does not include the effect of vibrational anisotropy
(Goldanskii–Karyagin effect) and thus for any meaningful interpretation one should give proper consideration
MI MI
3 1
± 2 ± 2
I= 3 DEQ
2
1 3
± 2
± 2
1 1
I= 1
2
± 2 ± 2
q=0 q>0 q<0
(a) h=0
MI MI
I+>,I+⬘ I−>,I−⬘
I= 3 DEo
2
I−>,I−⬘ I+>,I+⬘
1,1 1,1 1,1 1,1

I= 1 >, > >, >
2 q=0 q>0 2 2 2 2 q<0 2 2 2 2
h≠0
(b)
d I.S.
It
DEQ
E
(c) Velocity (mm/s)
Fig. 10.19 An energy-level diagram illustrating the quadrupole splitting for 57Fe nucleus under different situations (a) axially symmetric
field gradient h = 0 (q > 0 and < 0). (b) asymmetry parameter h ≠ 0 (q > 0 and q < 0, and (c) theoretical Mossbauer spectrum.
while comparing intensities. Further it is interesting to note that for polycrystalline sample the intensity
ratio is unity.
Note that for q = 54°44′, Eq. (10.80) yields an intensity ratio of unity. Thus, the presence or absence of any
asymmetry parameter is not evident from any ordinary MB spectrum of a powdered sample.
10.9.2 Magnitude and Sign of Quadrupole Interaction

The magnitude of the quadrupole interaction is a product of two factors: (i) eQ which is a nuclear constant for the
resonant MB isotope, and (ii) eq which is a function of chemical environment. For a 3/2−1/2 transition, it is not
possible to determine the sign of e2qQ or the magnitude of h from the line positions alone. However, for higher
spin states, the sign of e2qQ can be determined from unequally spaced lines. If both Ie and Ig are greater than 1/2
then eq can be estimated because the ground state quadrupole moment can be determined from other measure-
1
ments, i.e. NQR. If Ig = 0 or , then Q for the excited state is calculated from estimated values eq in chemical
2
compounds. Although e2qQ and h can be evaluated from a MB spectrum, it is difficult to relate these parameters
to electronic structure which generates them.
10.10 COMBINED MAGNETIC AND QUADRUPOLE INTERACTIONS

As has already been discussed in Chapter 8 on nuclear quadrupole resonance spectroscopy, when both magnetic
and quadrupole interactions (both are directional effects) exist simultaneously, their respective principal axes may
or may not be in the same direction because of which the resulting behaviour may be complex. Consequently,
the general interpretation of the spectrum can be quite complex. The combined effects may cause a change in the
line positions and/or in the relative line intensities or width of the hyperfine pattern. The combined effects can
operate in different ways, either the magnetic hyperfine interaction is perturbed by an electric quadrupole interac-
tion or the quadrupole interaction is perturbed by magnetic dipole interaction, for instance by an applied external
magnetic field. In case the quadrupole interaction is very weak as compared to magnetic hyperfine interaction
(i.e. e2qQ << mH), it may cause a perturbation of the magnetic hyperfine interaction. The resultant energy level
for such cases is given by
MI +
1
Q ⎛ 3cos
e 2qQ 3 cos 2 θ − 1⎞
EQ g N βN H M I + ( − ) 2
(10.81)
4 ⎜⎝ 2 ⎟⎠
3 1
Here, q is the angle between magnetic axis and the major axis (Z-axis) of EFG. For the transition − , all
2 2
the magnetic hyperfine levels in the excited state and hence the lines in the resultant MB spectrum are shifted by
the quantity.
e 2qQ
ε ( c 2
θ − 1) (10.82)
4
3
The symbol e is often used if q is unknown. For a positive EFG (a−Fe foil), e is positive for MI = ± and
2
negative for MI = ± 1/2, which implies that the substates with MI = ± 3/2 will be raised by amount ε while those
with MI = ± 1/2 will be lowered by the same amount. On the other hand, for a negative EFG (naturally occurring
chalcopyrite) the substates with MI = ± 3/2 are lowered by an amount ε and those with MI = ± 1/2 are raised by the
same amount. The schematic diagram showing the splitting of the excited (3/2) and ground state (1/2) levels in the
combined action of internal magnetic field and quadrupole interaction and the resultant six-fingered MB spectrum
for positive EFG is shown in Fig. 10.20. The angle q and value of e2qQ cannot be determined separately from the
H ≠ 0, Uzz > 0
H ≠ 0, Uzz = 0 6 3, 3
+ >
6 ε 2 2
3 5
3 ε 3 5 3, 1
I= + >
2 2 4 2 2
2 3, 1
4 − >
1 2 2
1 3, 3
− >
2 2
3 2 1 1 2 3
1, 1
1 − >
I= 2 2
2
Isomer shift 1, 1
+ >
Magnetic dipole Magnetic dipole splitting 2 2
splitting + electric quadrupole
(a) perturbation
Relative transmission
1 2 3 4 5 6 1 2 3 4 5 6
v (mm/s)
(b)
Fig. 10.20 Magnetic dipole splitting without (H ≠ 0, Uzz = 0) and with electric quadrupole splitting (H ≠ 0, Uzz > 0) (a) and the resultant
spectra (b) of Fe57.
position of the lines. The asymmetrical nature of the spectrum about the centroid is an evidence for the presence
3 1 1
of quadrupole perturbation in the − magnetic hyperfine splitting. However, if cos θ , the second term of
2 2 3
Eq. (10.81) will vanish and the spectrum resembles with that of an unperturbed magnetic hyperfine splitting.
The ground state (gg) and excited state (ge) splittings can be determined from Eqs (10.67) and (10.68) respectively
while the isomer shift from Eq. (10.69), i.e.
g g = v 4 − v 2 = v5 − v3 (Eq. 10.67)
g e = v3 − v 2 = v5 − v 4 (Eq. 10.68)
1 ⎡1
( v1 + v6 ) + ( v 2 + v3 + v 4 + v5 )⎤⎥
1
and δ IS = (Eq. 10.69)
2 ⎣2 4 ⎦
The quadrupole coupling, eQUzz is never between 4ε(q = 0°) and −8ε(q = 90°). Twice the difference between
the centres of gravity of lines 1 and 6 and lines 2−5 gives the best value of 4ε:
⎡1
( v1 6 ) ( v 2 v3 + v 4 + v5 )⎤⎥
1
4ε (10.83)
⎣2 4 ⎦
Thus the following quantities can be estimated from the hyperfine interactions (i) magnetic moment ratio,
me/mg = 3ge/gg, (ii) first excited-state magnetic moment, me = mg (3ge/gg) provided the value of the ground state
magnetic moment mg is known from other source, and (iii) hyperfine magnetic field using the expressions (10.76)
or (10.77), i.e.
⎛ ggI g ⎞
H = − (15.250 kOe s/mm nm) ⎜ (10.84)
⎝ μ ⎟⎠ g
H (T K )
Hyperfine magnetic field ratio and temperature shift δ IS ( ) δ IS ( ) can also be estimated by
H (T K )
performing the experiment at two different temperatures T1 and T2 say T1 = 298 K and T2 = 4.3 K.
In case of naturally occurring chalcopyrite, the observed spectra at 300, 373 and 448 K (Fig. 10.21) conform
to the energy-level diagram shown in Fig. 10.22. This implies a negative EFG, i.e. ε. The magnetic interaction
(a) T = 300 K
39
38
37
36
35
Relative counting rate
(b) T = 373 K
69
67
65
63
(c) T = 448 K
68
66
64
62
−7 −6 −5 −4 −3 −2 −1 0 1 2 3 4 5 7
Velocity (mm/s)
Fig. 10.21 The observed Mössbauer spectra of naturally occurring chalcopyrite at (a) 300, (b) 373, and (c) 448 K.
M1
3
e −
2
D1 1
−
2
3
1=
2
1
+
2
3
+
2
1 2 3 4 5 6
1
+
2
1 D0
1=
2
1
−
2
No magnetic Magnetice dipole Magnetice dipole +

field interaction electric quadrupole
interactions
(a)
6
1
5
4
2 3
D1 D1
D0
D0
Spectrum
(b)
Fig. 10.22 (a) Schematic splitting of 57Fe levels of excited and ground states under the combined interaction of magnetic dipole and
electric quadrupole. (b) The six-fingered spectrum to a first approximation.
first leads to the splitting of the ground state and the first excited state of 57Fe characterised by ΔEm(g) and ΔEm(e).
3 1
The substates of the excited state with MI = ± are then displaced to lower and those with MI = ± raised by the
2 2
electric quadrupole interaction, i.e. ε. The six possible transitions with their Doppler energies required to have the
observed spectrum have been listed in Table 10.9.
Table 10.9 Six possible transitions and their Doppler energies.
Doppler energy EQM

Transition (Number) ΔMI
required for resonance
3 1 ΔE m ( g ) 3ΔE m ( e )
+ → + (1) −1 δ IS ε
2 2 2 2
1 1 ΔE m ( g ) ΔE m ( e )
+ → + ( 2) 0 δ IS ε
2 2 2 2
1 1 ΔE m ( g ) ΔE m ( e )
− → + (3) +1 δ IS ε
2 2 2 2
1 1 ΔE m ( g ) ΔE m ( e )
+ → − ( 4) −1 δ IS ε
2 2 2 2
1 1 ΔE m ( g ) ΔE m ( e )
− → − (5) 0 δ IS ε
2 2 2 2
3 1 ΔE m ( g ) 3ΔE m ( e )
− → − (6) +1 δ IS ε
2 2 2 2
These Doppler energies are the respective positions of lines in the observed spectra. Now the determination
of the parameter dIS, e, ΔEm(g), i.e. gg and ΔEm(e), i.e. ge is feasible in the following manner. dIS is obtained as the
mean values of the positions of the lines at v1, v2, v5 and v6 or v1, v3, v4 and v6 as the sum of the energies of interac-
tions EQM for the four excited substates must be zero in the absence of any isomer shift. Further defining S1 as the
difference in EQM for the lines v1 and v2 and S2 as the difference in EQM for the lines v5 and v6 we see that
S1 S2 = v 2 v1 − v 6 − v 5 = 4ε
i.e. e, the quadrupole interaction energy given by Eq. (10.83) is one quarter the difference in splitting of the outer
pairs of the six-line spectra. ΔEm(g) as shown in Fig. 10.22 is the difference in EQM of the lines v2 and v4 or v3 and
v5, i.e. v4 − v2 or v5 − v3. Similarly, Δ1 = v3 − v2 or v5 − v4. The value of the internal magnetic field can be estimated
from Eq. (10.74) provided the value of magnetic moment in the ground state (mg) is known. Alternatively the mea-
sured value ΔEm(g), i.e. gg for the specimen may be correlated with the corresponding ΔEm(g) for Fe which has been
determined from precise measurements related to Hint = 330 kOe. The values of dIS, e, and internal magnetic field
for naturally occuring chalcopyrite are given in Table 10.10.
Table 10.10 Isomer shift, quadrupole splitting and internal magnetic field
in chalcopyrite at various temperatures.
Temperature (K) δIS (mm/s) ε (mm/s) Hint (kG)
300 −0.03 ± 0.02 0.05 ± 0.02 325 ± 10

373 −0.03 ± 0.03 0.05 ± 0.03 312 ± 10
448 −0.04 ± 0.02 0.04 ± 0.02 290 ±10
10.10.1 Effect of Pressure on Quadrupole −0.001

Splitting
−0.002
e2qQ /4 (cm/s)
The behaviour of quadrupole splitting with pressure in 57Fe

in CoCl2 (rhombohedra1 structure) and57 Fe in hcp cobalt is −0.003
shown in Figs 10.23 (a) and (b) respectively. Theoretically,
quadrupole splitting varies as 1 / R I3 L where RI−L is the ion-
−0.004
ligand distance. The shape of curves in Fig. 10.23 follows
the theoretical behaviour. This is expected because as the
pressure grows, the ion-ligand distance decreases. 100 200
P (kB)
(a)
10.10.2 Effect of Temperature on
0.12
Quadrupole Splitting
The effect of temperature on quadrupole splitting depends 0.11
e2qQ /4 (cm/s)
on the nature of the compound. Quadrupole splitting in

several ferrous compounds, e.g. FeF2, FeSO4,FeSO4.7H2O, 0.10
FeSiF6.6H2O, FeCl2.4H2O, FeC2O4.2H2O grows very slowly
and then also decreases very slowly in the temperature range 0.09
0−350 K. The dependence of quadrupole splitting on tem-
perature is usually small for Fe3+ ions since EFG arises only
from lattice contributions. The temperature dependence of 100 200
P (kB)
quadrupole splitting (in the temperature range 0−400 K) for
(b)
both Fe3+ sites for three compositions of the (Mn1−x Fex)2O3
Fig. 10.23 (a) Quadrupole splitting of 57Fe in CoCl2 as a
where x = 0.028, 0.060 and 0.082, system indicates little function of pressure (b) Quadrupole splitting of 57Fe in hcp
temperature dependence in the high-temperature region cobalt as a function of pressure.
(>250 K, for x = 0.028, >150 K for x = 0.060), as expected
for Fe3+ containing materials. However, for the specimens with x = 0.028 and 0.060, the variation of ΔEQ with
temperature increases markedly below specific temperatures (i.e. below 250 K for x = 0.028 and <150 K for
x = 0.060). The inflexion at 250 K for x = 0.028 and 150 K for x = 0.060 suggests the phase transition from
Table 10.11 Effect of temperature on various configurations of iron.
Configuration Symmetry (ΔEQ)0 Temperature dependence
Oh ⎤
Fe2+ (high spin) ⎥ large yes
Td ⎥⎦
Oh ⎤ no ⎤
Fe(II) (low spin) ⎥ large ⎥
Td ⎥⎦ yes ⎥⎦
Oh ⎤
Fe3+ (high spin) ⎥ small no
Td ⎥⎦
Oh ⎤
Fe(III) (low spin) ⎥ Intermediate yes
Td ⎥⎦
cubic to orthorhombic. Samples with x = 0.082 remain cubic to below 55 K. It should be borne in mind that not
all paramagnetic configurations of iron exhibit a temperature-dependent quadrupole splitting as evident from
Table 10.11.
However, in the second- and third-row transition metals, the spin-orbit coupling is much larger because of
which an almost temperature independent or much reduced quadrupole splitting below 300 K is observed.
For tetrahedral environment, the relation between ΔEQ and temperature is given by
1 − e − Δ1 / κT
Δ Q = (Δ
ΔE
EQ ) 0 (10.85)
1 + e − Δ1 / κT
where (ΔEQ)0 is the value of quadrupole splitting at absolute zero, i.e. (ΔEQ)T→0, = (ΔEQ)0. Δ1 is the energy differ-
ence between two dg lines. At high temperature ΔEQ → 0, i.e. two dg levels have equal electron population when
kT >> Δ1.
10.11 SIGN OF QUADRUPOLE INTERACTION

The determination of sign of quadrupole constant and hence Uzz is important because it plays an important
role in studies of electronic and molecular structure of compounds containing MB nuclides, e.g. since for 57Fe,
Q = +0.18 barn, a positive value of e2qQ suggests that Uzz > 0. The MB effect makes it possible to determine the
sign of quadrupole interaction. A number of methods are reported in the literature but only the three are being
described here.
The sign of quadrupole coupling constant can be determined in a single crystal by studying the angular varia-
tion of area ratios. The increase of area ratios as a function of angular variation suggests negative sign for quadru-
pole interaction and vice versa.
The second method due to Ruby and Flinn is based on the perturbation of quadrupole doublet by the strong
external magnetic field. When a strong external magnetic field is applied, the doublet of the quadrupole spectra
(in paramagnetic or diamagnetic state), split into a triplet and a doublet provided the value of h is small. For
positive e2qQ, the doublet will be on the higher energy side while on the lower energy side for negative e2q Q.
This method fails if h deviates significantly from zero. The doublet/triplet observation is then not applicable. The
doublet/triplet observation in case of ferrocene and butadiene iron tricarbonyl is shown in Figs 10.24 and 10.25
respectively.
The third method to determine the sign of quadrupole interaction in case of a powdered sample is based on
the simultaneous presence of magnetic dipole and electric quadrupole interaction. The structure of the hyperfine
multiplet depends upon whether the EFG is parallel to the direction of the hyperfine magnetic field, i.e. direction
of propagation of g -ray beam or perpendicular to it. Thus from the splitting pattern, it is possible to determine the
magnitude as well as the sign of quadrupole interaction. When splitting is reduced on the application of external
magnetic field parallel to the direction of the internal magnetic field, the sign of quadrupole splitting will be nega-
tive. The sign of quadrupole interaction is also obtained by comparing the splitting pattern of hyperfine multiplet
of unknown absorber with the hyperfine structure of a system with known sign of quadrupole interaction.
0 0
Triplet Doublet Doublet Triplet
Per cent absorption

Per cent absorption
10
5
−1 0 1 2
−2 0 2 Velocity (mm/s) relative to iron
Velocity (mm/s) relative to iron
Fig. 10.25 Mossbauer absorption spectrum of butadiene iron
Fig. 10.24 Mossbauer absorption spectrum of ferrocene at 4.2 K tricarbonyl at 4.2 K and 26 kOe. Uzz is negative here.
and 40 kOe. Uzz is positive here.
10.12 GOLDANSKII–KARYAGIN (G-K) > (G-K) EFFECT (LATTICE

VIBRATION ANISOTROPY)
The phenomenon of induction of angular independent asymmetry in the intensity of the magnetic and/ or quadru-
pole hyperfine pattern by anisotropic lattice vibrations of the crystallite or anisotropic recoil free fraction of g -ray
is called Goldanskii–Karyagin effect.
At high temperature, the mean-square displacement x i2 of the atom in the lattice is linearly proportional to
temperature. Therefore, the straight line plot of <x2> against temperature should pass through the origin. Such a
behaviour has been observed in hydrated sodium hexacyanoferrate (II). However, in case of potassium hexacy-
anoferrate (II), the line intercepts the vertical axis. This suggests the presence of anisotropy in the lattice vibra-
tions. Large mean-square displacements have been observed even at low temperatures, e.g. low temperature
anharmonicity has been observed in FeCl2; potassium hexacyano ferrate(II) becomes ferroelectric below 251 K
because of which frequencies of some of the optical modes of the lattice vibrations change.
Now, if we define three principle components of x i2 in the coordinate system X, Y, Z such that ‘i’ is defined
by the polar angles q and j, then
x i2 ⎡ x x2 cos 2 ϕ + x y2 sin
i 2 ϕ ⎤⎦ sin 2 θ + x z2 cos 2 θ (10.86a)
⎣
In case of axial symmetry,
x x2 = < x y2 > and j = 0; Thus Eq, (10.85) becomes
x i2 = x x2 si 2 θ + x z2 cos 2 θ (10.86b)
It is the convention to write
x z2 as x 2 and
x x2 as x ⊥2 , so Eq. (10.86b) is written as
x i2 ⎡ x 2 − x ⊥2 ⎤ cos 2θ + x ⊥2 (10.87)
⎣ ⎦
We already know that the recoiless fraction of g -rays is related to < x 2 > by the expression (10.21), i.e.
⎡ − Eγ2 x 2 ⎤
f = exp ⎢ ⎥ (Eq. 10.21)
⎢⎣ ( / 2π ) ⎥⎦
2
Here, < x 2 > is assumed to be an isotropic function which of course is not always true. The isotropic concept
is valid when absorber is a random microcrystalline sample with single MB resonance line and an average value
2
of x can be defined. However, the vibrations of an atom at the lattice site of noncubic symmetry are in gen-
eral anharmonic. This anharmonicity of lattice vibrations induces anisotropy in the recoiless fraction of g -rays
and has been observed in single crystal samples, e.g. anisotropic recoilless emission has been noticed in experi-
ments using 57Co source doped in a single crystal of zinc metal. The crystal is cut at an angle of 45° to the c-axis.
At room temperature, the values of f⊥ = 0.64 and f ⎜⎜ = 0.41 reflect large anisotropy in vibrations of Co impurity
atom. In view of it, the expression (10.21) should be written as,
⎡ − x ⊥2 ⎤ ⎡ − x 2 + x ⊥2 ⎤
f ( ) = exp ⎢ ⎥ exp
p ⎢ cos 2 θ ⎥
⎢⎣ ( hc / 2π Eγ ) ⎥⎦ h / 2π Eγ )
2 2
⎢⎣ ( hc ⎥⎦
or
⎧⎪ ⎡ − x ⊥2 ⎤ ⎡ x 2 − x ⊥2 ⎤ ⎫
2 ⎪
f ( ) = exp ⎨ ⎢ ⎥ − ⎢ ⎥ c
co s θ ⎬ (10.88)
⎪⎩ ⎢⎣ ( / 2π Eγ ) ⎥⎦
2 2
⎢⎣ ( h / γ) ⎥
⎦ ⎪⎭
Let us now see how the anisotropic recoilless fraction of g -rays influences the intensity of the hyperfine
quadrupole doublet corresponding to 3/2 − 1/2 transition. The intensity I3/2 / I1/2 of the lines involving |± 3/2 > and
|± 1/2 > levels of the excited state depends upon the angle q between the direction of propagation of g -ray and
principal Z-axis of the EFG and is expressed as,
I 3 / 2 ( ) 1 + cos 2 θ
= (10.89)
I 1/ 2 ( ) 2
+ sin 2 θ
3
For a random polycrystalline samples the averaged value of intensity ratio is obtained by integrating over all
the orientations, i.e.
π
∫0 (1 cos θ )sin
)si θ dθ
2
I 3/ 2 ( )
= =1 (10.90)
I 1/ 2 ( ) π ⎛ 2 2 ⎞
∫0 ⎜⎝ 3 + sin θ ⎠ sin
i θ dθ
These formulae hold good only for axial symmetry and become more complex for non-axial field gradient
because of state mixing. The above argument is valid provided the probability for recoiless event (emission or
absorption) is equal in all directions. However if the recoiless radiation is anisotropic, the recoil free fraction ‘f’
for a thin absorber is a function of q, then Eq. (10.90) modifies to,
π
∫0 (1 cos θ ) (θ ))sin θdθ

2
I 3/ 2
= ≠1 (10.91)
I 1/ 2 π ⎛ 2 2 ⎞
∫0 ⎜⎝ 3 + sin ⎠ (θ ))sin θdθ
3 1
Thus, an anistropic recoiless fraction of g -rays generates an angular independent asymmetry in the − qua-
2 2
drupole doublet. This is termed as the Goldanskii–Karyagin effect and in case of a iron silicate glass is shown in
Fig. 10.26, e.g. the value of I3/2/I1/2 determined from the hyperfine quadrupole split spectrum of Me2SnF2 at room
temperature has been found to be 0.78. The root mean square amplitudes of vibration of tin deduced from x-ray
come out to be 13.7 pm along the Sn–F bonds and 2.10 along the Sn–C bonds. By making use of Eq. (10.91) and
the known values for x ⊥2 and x 2 , given above, the intensity ratio comes out to be 0.72 which is in agreement
with the experimental value, i.e. 0.78.
It is pertinent to note that the main features of asymmetry, because of vibrational anisotropy in polycrystalline
materials are identical to the effects of partial orientation of crystallites, molecules or spins (texture). Texture has
a sensitive effect on the relative line intensities of a quadrupole hyperfine pattern. It is difficult to differentiate
experimentally between G−K and texture effects. So, it is to be checked whether the possibility of orientation has
been eliminated by checking for the angular dependence of spectra. The safest method of preparing unoriented
absorbers is by grinding the sample with large bulk of an abrasive powder such as Al2O3 or quartz glass.
The principle of G−K effect is also applicable to magnetic hyperfine spectra and to isotopes with higher spin
states and in all cases deviation from the predicted line intensities have been noticed.
Absorption per cent

1
5
−1.5 −1.0 −0.5 0 0.5 1.0 1.5
Velocity (mm/s)
Fig. 10.26 Goldanskii–Karyagin effect in the Mössbauer spectrum of an iron silicate glass.
In case of magnetic hyperfine interactions, the intensities of the ΔMI = 0 lines (I2, I5) and the inner ΔMI = ±1
lines (I3, I5) are defined by the ratio,
I 2 ,5 sin 2 θ
=
I 3, 5 1
(1 cos 2 θ )
4
π
∫0 sin (θ ))sin θdθ

2
I 2 ,5
So = (10.92)
I 3, 5 π 1
∫0 4 (1 cos θ ) (θ ))sin θdθ
2
The G−K effect may be employed to deduce the sign of Uzz in both 57Fe and 119mSn spectra.
The dependence of intensity ratio upon (x 2
) may be employed to identify the lines (i.e. p and s com-
x2
ponents), thus the sign of Uzz can be determined if the sign of ( x x ) is known. Conversely, the sign of
2 2
vibrational anisotropy factor can be determined, if sign of Uzz is known. In Fe and Sn, the MI = ± 3/2 line is the
more intense for negative values of the anisotropy parameter x2 x2 and is less intense for positive values.
x 2
= 1.76 × 10 −18
cm and x 2 2
⊥ = 0.48 × 10 −18
cm . 2
10.13 PARTIAL QUADRUPOLE SPLITTING (PQS)

Similar to partial centre shifts (p.c.s.); quadrupole splitting is also an additive property. The resultant quadrupole
splitting could be expressed as an algebraic sum of partial quadrupole splitting. For any low-spin iron (II) com-
pound of the type FeA2B4 or FeAB5, ΔEQ can be expressed as
ΔEQ (trans) = +4 (p.q.s.)A − 4(p.q.s.)B (10.93)
ΔEQ (cis) = −2(p.q.s.)A + 2(p.q.s.)B (10.94)
ΔEQ(FeAB5) = +2(p.q.s.)A − 2(p.q.s.)B (10.95)

The partial quadrupole splitting for various ligands are listed in Table 10.12.
Table 10.12 Tentative partial quadrupole splitting values (mm/s) for

some ligands.
Br− −0.32 depb −0.02

I −
−0.31 depe +0.02
Cl −
−0.30 dmpe/2 +0.07
N −
3
−0.22 RNC +0.09
SnCl −
3
−0.17 CO +0.16
NCO −
−0.10 SiH−3 +0.40
NCS− −0.11 H− +0.44
CN −
−0.06 NO +
+0.91
Problem 10.4: Using the data on p.q.s. in Table 10.12, calculate the quadrupole splitting for (a)
cis-[FeCl2(ArNC)4] (b) trans−[FeCl2 Ar(NC)4] (c) cis-[Fe(CN)2(CNEt)4], (d) trans-[Fe(CN)2 (CNEt)4] and (e)
[FeCl(ArNC)5] (ClO4).
Solution (a) cis-[FeCl2(ArNC)4] is of the type FeA2B4. So by Eq. (10.94)
ΔEQ(cis) = −2(p.q.s.)Cl + 2(p.q.s.)ArNC

= −2(−0.30) + 2 (0.09)
= 0.60 + 0.18 = 0.78 mm/s (obs: −0.78 mm/s)
(b) For trans−[FeCl2 Ar(NC)4];
ΔEQ (trans) = 4(–0.30) − 4(0.09) = −1.20 – 0.36 = −1.56 mm/s (obs: +1.55 mm/s)
(c) For cis-[Fe(CN)2(CNEt)4];
ΔEQ = −2(−0.06) + 2(0.09) = 0.12 + 0.18 = 0.30 mm/s (obs: +0.29 mm/s)
(d) For trans-[Fe(CN)2 (CNEt)4];
ΔEQ = 4(−0.06) − 4(0.091) = −0.24 – 0.36 = –0.60 mm/s (obs: −0.60 mm/s)
(e) For [FeCl(ArNC)5] (ClO4)
ΔEQ = 2 (−0.30) – 2 (0.09) = −0.60 – 0.18 = 0.78 mm/s (obs. +0.70 mm/s)
10.14 MÖSSBAUER SPECTROMETER

Basically, there are four elements of MB-spectrometer (i) Mössbauer drive to provide controlled and well-defined
relative velocity to source or absorber for Doppler tuning, (ii) Mössbauer source, (iii) Mössbauer absorber (or
scatterer), and (iv) a suitable counting assembly for detecting the transmitted or scattered g -beam intensity. The
elements and their functions are shown in Fig. 10.27.
Source Absorber g –ray detector Amplifier

g
Multichannel analyser Discriminator

Mössbauer
drive
Chart recorder
Computer
Oscilloscope
Fig. 10.27 A block diagram of a Mössbauer spectrometer.
Several additional auxiliary systems (e.g. cryostate, furnace, arrangement for pressure variation, etc.) may be
necessary to perform a particular experiment.
(a) Mössbauer Drive Systems The desired velocity to the source is imparted by the system called the MB
drive. It is of two types.
(i) Constant Velocity Drive In the constant velocity mode either the source or target is moved with a con-
trolled constant velocity for the pre-set time during which the transmitted g -ray counts are recorded. With this
drive system, we record the whole spectrum point by point. One can select a particular velocity of interest for
specific investigations, e.g. measuring the resonance signal as a function of temperature (thermal scan). A change
in count rate might indicate (Curie, Neel and Morin temperatures phase transitions, etc.) which can be determined
with high accuracy by this method. The limitation of this drive system is that the range of velocities covered is of
about 10 or 20 mm/s. This limitation makes these drives useful only for 57Fe and 119Sn isotopes.
(ii) Constant Acceleration Drive In the constant acceleration mode, the velocity of the source is changed
continuously from −v to +v in one cycle. The main advantage of this method over constant velocity method
is that in this method the count rate for each velocity over the entire range is made in each and every cycle so
that drifts in the count rate do not affect the observed spectrum and very small MB effects can be measured.
Moreover velocities up to 1000 mm/s can be attained by this method. This method is disadvantageous to con-
stant velocity method because the velocity measurements in this method are not direct, one has to calibrate the
drives often.
(b) Mössbauer Source and Mössbauer Isotopes The 57

7 Co (270 days)
Mössbauer source is a system which emits a fraction of recoil- −
2
free g -rays. There is no unique method for the preparation of
good source. However, a good source should have long life, a
Electron capture
narrow line width, high Lamb MB factor, etc.
Generally, MB source is a solid matrix called ‘host matrix’
5
containing the excited nuclei of a suitable isotope. The selec- − 137 keV
2
tion of a suitable host matrix is very important from exper-
imental point of view. For the sources, mostly high melting
metallic matrices with high coordination symmetry are chosen
in which the parent isotopes are diffused, implanted, Coulomb 3
− 14.37 keV
2
excited, or produced by nuclear reaction. The metals have a 1
− 0 Stable
number of advantages over other materials, viz., nonmagnetic 2 57
Fe
cubic metals (bcc/fcc) with low impurity concentration produce Fig. 10.28 Decay scheme of 57Co leading to
single-line emission spectra, i.e. complexity in the spectra due 14.37 keV γ-ray photons from 57Fe.
to hyperfine splitting does not occur, the Lamb MB factor is
relatively high for most metals and the electronic relaxation
process are extremely fast as compared to the lifetime of the 5
− 137 keV
excited state of the MB nuclei. High-melting refractory oxides 2
of cubic structure are also very popular host matrices. e.g. host
for 119mSn is the oxide BaSnO3.
Although these days, MB sources are commercially avail-
able, the two alternate methods used to populate the 14.4 keV
MB transition in 57Fe are illustrated by Figs 10.28 and 10.29. 3
− 14.37 keV
In the first method, 57Co with half life of 270 days is prepared 2
by the (d, p) reaction on 57Fe. The carrier free 57Co isotope 1
− 0 keV
is plated on a base; such metals like Pt, Pd, Cu, Cr, Rh, etc., 2 Fe57
and alloys like some of the stainless steels. The plated iso-
Fig. 10.29 Decay scheme of 57Fe leading to 14.37 keV
tope is then diffused into the base metal by heating at ~700°C Mössbauer transition of 57Fe.
in an inert atmosphere or better, ‘in a reducing atmosphere
of hydrogen. Certain stainless steels, give a single line, but
these are broader than the Heisenberg width due to self-resonant absorption. 57Co: Cr and 57Co : Pd sources
have been found to be the best. These source matrices are not self-resonant and yield high recoilless fraction of
g -rays with no significant line broadening of the natural width. The Lamb MB factor for 57Co:Cr source is 0.81
at room temperature and MB effect can be studied with this source up to very high temperatures. In the second
method, EC is produced by the 1.5 mA beam from 3 MeV a-particle Vande Graaff accelerator. The target and
absorber are 1.9 mg/cm2 of 91 per cent enriched 57Fe. The recoilless emission is observed following Coulomb
excitation of the 137 keV level. The 14.4 keV is detected in coincidence with the 122 keV g -ray in order to
reduce the background of x-ray and bremsstrahlung. It may be pointed out that in the (d, p) (n, g ), etc., reac-
tions produced sources there is radiation damage, which results into a recoil-free fraction ‘f’. The variety of host
matrices, mostly of metallic form are not without limitations. Pt which is chemically inert and gives large room
temperature recoilless fraction, ‘f’, is not well suited because of large photoelectric and Compton scattering due
to its high atomic number, as the latter will make the diffusion requirements for such a source extremely strin-
gent. On the other hand, Cu matrices, while having large recoil-free fraction at room temperature and low atomic
number, are chemically unstable, since unprotected sources tend to react with atmospheric oxygen and sulphides
which may be present in the ambient environment. Pd though is a reasonable compromise between the various
characteristics, which in addition also gives a narrow line has also one disadvantage of having an x-ray of 21 keV
falling near the 14.4 keV MB g -ray level, which may increase nonresonant background (in case of scintillation
detector). A very good 23.8 keV g -ray source (119Sn) is made using Mg2Sn or KSnO3 lattices. The Mg2Sn sources
have to be sealed or kept in dry atmosphere, otherwise they will decompose. The line width of these sources is
12 times the natural line width and the Lamb MB factor is also very high. The other sources of 119Sn are Pd−Sn;
Ag−Sn.
The decay scheme of some typical MB nuclei, viz., 57Fe leading to 14.40 keV, 193Ir leading to 73.00, 138.9 keV,
l29
I leading to 27.8 keV, 127I leading to 57.6 eV and 119Sn leading to 23.9 keV MB transitions are shown in Figs.
10.30−10.34 respectively. Inspection of the decay schemes indicates that the radiations emitted from the source
are complex, and mostly consist of (i) recoil-free resonant g -rays from the same transition, (ii) nonrecoil free
nonresonant g -rays from the same transition, (iii) radiation from all other transitions, and (iv) secondary radiation
produced in the matrix mainly x-ray by conversion electrons. Except the recoil-free resonant g -rays which cause
the MB effect, all the other radiations contribute to the background and produces complications especially when
the energy of the background radiation is close to that of MB g -ray, e.g. in 1l9Sn one gets 26 keV x-rays and are
filtered by Pd filter of suitable thickness.
In order that the MB resonance is observed easily, the MB isotope should qualify the following requirements:
(i) The energy of g -ray must lie between 10−150 keV preferably less than 50 keV, since the Lamb MB factor ‘f’
and resonant cross-section ‘s0’ both decrease as Eg grows. Consequently, isotopes higher than 40K do not show
57Co(270 days)
EC
0.16 percent
EC
26Fe57 (∼0.6 MeV)
5/2
Complex 99.84 percent

366.8
3/2
Complex
5/2 136.4 keV
136.4 keV 122.0 keV

9.0 percent 91 percent
3/2 14.4 (t½ = 10−7 s)
Mössbauer transition α = 9.0
1/2 0.0
Ground state
Fig. 10.30 Decay scheme of 57Co showing 14.4 keV Mössbauer transition of 57Fe.
193Ir (73.5 keV, 138.9 keV)
139Os
30 h
b−
Complex
12 percent
14 percent 5−
2 138.95 136.96
0.080 ns
59 percent
1/2+ 73.03 73.03

6.3 ns
3/2+
0
193
Ir
Fig. 10.31 Decay scheme of 193Os leading to 73.03 and 138.95 keV Mössbauer transitions from 139Ir.
129
53I(27.8 keV)
129mTe 105.6 105.6 34 d

11/2−
3 percent
33 percent Complex
129Te
3/2+ 69 m
b− Complex
1 percent
250.6 278.4 278.4
89 percent
5/2+ 27.77 27.77

16.8 ns
+ 1.57 × 107 y
7/2
129I
−
b
129
Xe
Fig. 10.32 Decay scheme of Te showing 27.8 keV Mössbauer transition from 129I.
129m
MB effect because of high energy of emitted g -rays. (ii) The half-life of the first excited level of MB nuclei which
determines the line width Γ should be between 1 and 100 ns. If t1/2 is very long, T becomes so narrow that reso-
nance overlapping is destroyed and for short t1/2, Γ becomes so broad that hyperfine interactions can’t be resolved
and detected. (iii) The internal conversion factor at, should be <10, otherwise the probability of detection of
g -ray decreases. (iv) The ground-state isotope should be stable and have a high natural abundance so that isotopic
enrichment may not be required. (v) The precursor (i.e. the isotope which decays by b-decay, electron capture or
isomeric transitions) should have longer life, otherwise the population of the required excited level will not be
maintained.
127 I(57.6 keV)

53
88.26
11/2 − 109 d
b−
127Te Complex
3/2 + 0
9.35 h
b−
0.07 percent
2.4 percent 1/2+ 374.96 172.1 375.0
98.8 percent
202.84 145.22 202.84
3/2+
57.60 57.60
7/2+
1.90 ns
0
5/2+ 127
I
Fig. 10.33 Decay scheme of 127

Te leading to 57.60 keV Mössbauer transition of 127l.
119mIn
119
50Sn (23.9 keV)
1−
2 300
18 m
5 percent b−
9+ 119In
2 0
2.1 m
β− Complex
Complex
11−
6
.6
65
50 percent 2 89.5
245 d
40 percent
3+
7
.8
23
2 23.87
17.75 ns
1+
2 0
119Sn
Fig. 10.34 Decay scheme of 119mIn showing 23.87 keV Mössbauer transition from 119Sn.
Electron capture: A nucleus emits no particles but decreases in Z by one unit by capturing an
orbital electron, most frequently from the K shell; e.g.
EC
133
Ba56 Cs55
133
This is an alternative process to positron ( b+) emission.

22
Na11 → 22Ne10 + b+
Isomeric transition: A nuclide in a metastable nuclear state passes to a nuclear state of low
energy, with emission of a γ-ray; e.g.
197m
Au79 IT 197
Au79 + g
Conversion electron: Internal photoelectric effect which takes place in the excited nuclear
state of the atom gives rise to electrons called conversion electrons
In spite of these limitations, an MB effect has been observed in 100 transitions of 83 different isotopes in
44 elements. Some of the typical MB isotopes along with their characteristics parameters are listed in Table 10.2.
(c) Mössbauer Absorber (or Scatterer or Sample) Just like the source, the standard reference absorber
should be stable and should not undergo any chemical or physical change that may affect the MB spectra. It
should have narrow line width and large Lamb MB factor. Metallic a-57Fe foil (i.e. Fe at room temperature) and
single crystal of sodium nitropruside, Na2[Fe(CN)5NO]2H2O have been accepted as standard reference absorbers
for Mössbauer nuclei 57Fe.
The absorber contains the same MB isotope as the source, in its ground state. In most of the cases, the absorber
is the material to be studied. It may be pointed out that the absorber mounting materials must be MB-isotope free
as otherwise even in traces it may sometimes complicate the whole MB spectra.
The absorber physical thickness ta, plays a significant role in the quantitative analysis of the spectra. The effec-
tive thickness of an absorber is defined by a dimensionless parameter,
Ta f aσ 0 nat aaa (10.96)
where, fa is the Lamb MB factor in the absorber, na is the number of resonant atoms/cm3 of the particular element,
ta is the physical thickness of the absorber in cm, s0 is the maximum absorption coefficient (cm2) and aa is the
isotopic abundance of the MB isotope.
The following relations exist between apparent width G app and effective absorber thickness Ta.
Γ app
= 2.00 + 0.27 Ta , 0 ≤ a ≤5 (10.97)
Γ nat
Γ app
= 2.02 + 0.29Ta − 0.005
00 Ta2 , 4 ≤
0 005 a ≤ 10 (10.98)
Γ nat
Γnat is the natural line width.

Further, the absorption is defined as
(I ∞ I )
A= (10.99)
I∞
Here, I0 is the transmitted intensity at the resonance maximum, I∞ is the transmitted intensity at the large Dop-
pler velocity where the absorption is zero.
The absorption is related to the effective thickness by the expression
Ta
A fs [ e 2 J (i Ta / )] (10.100)
where, fs is the recoilless fraction of the source. J0 (i Ta/2) is the zero-order Bessel function, i.e.
4 6
⎛x⎞ ⎛x⎞
2 ⎜⎝ ⎟⎠ ⎜⎝ ⎟⎠
⎛ ⎞
x 2 2
J 0 ( x ) = 1 + ⎜ ⎟ + 2 2 + 2 2 3 + ...
⎝ 2⎠ 1 2 1 2 ⋅ 3
The plot between A/fs and Ta shown in Fig. 10.35 reveals that the absorption shows a saturation with thick-
ness. Therefore, an appropriate value of thickness should be used to obtain adequate absorption. If Ta << 1,
the intensities of resonance lines are small. The line can be approximated to Lorentzian curves. If Ta >>1, the
shape of the absorption curve is nearly of Lorentzian type but with

a broadened width 2G , given by
2Γ
= 2 0.27 Ta (10.101)
Γ nat A/fs
provided Ta < 5.
(d) g -ray Detectors Sodium iodide, NaI (Tl), scintillation coun-

ters, proportional counters and lithium-drifted germanium and sili-
con detectors may be used for registration of g -rays. Photographic Effective thickness
emulsion can also be used in special cases. The pulses from the Fig. 10.35 The behaviour of absorption with
detector are amplified and passed through a discriminator which effective thickness (Ta) of the absorber.
filters out most of the nonresonant background radiation. The reso-
nant radiation is then fed to a multichannel analyser which is synchronised with an MB drive. The stored data in
the multichannel analyser is traced in the recorder. A g -ray emitted along the axis of motion of the source has a
Doppler shift relative to the absorber of Eγ v , but any g -ray which travels to the detector along the path at an angle
c
q to the axis has an effective Doppler shift of only Eg v cos q/c. This cosine effect of solid angle causes a spread
in the apparent Doppler energy of the g -rays and hence line broadening. The cosine effect can be reduced by
maintaining an adequate separation between source and absorber or by collimation of g -ray beams. Pulse height
spectra from 57Co source is shown in Fig. 10.36(a-c). Inspection of Fig. 10.36 indicates that resolution depends
on the type of g -ray detector. However, the poor resolution of a detector is counter balanced by the high count
rate which can be achieved. It may be pointed out that the relatively low g -photon flux density necessitates much
longer counting times to achieve significant counting statistics than for example in optical spectroscopy. The sta-
tistical behaviour of g -emission results in a standard deviation of n for a number of n registered counts. Hence
the standard deviation in 10,000 counts is 100(1 per cent), in 1000,000 counts is 1000(0.1 per cent). The longer
the counting time, the better the definition of resonance line, but the improvement to be gained must be balanced
against the experimental time required and the long-term stability or reproducibility of the apparatus. Efficiency,
resolution, count rate and price of the g -ray detector must be considered before its selection.
10.14.1 MÖSSBAUER SPECTRUM

A Mössbauer experiment can be performed either in a transmission geometry or scattering geometry. The scat-
tering geometry experiments have some outstanding and special advantages over the transmission geometry but
require high source strength (as the scattering cross-section is less than absorption cross-section by a factor of
1 σ ( )
[= 0 ] and careful attention to geometry and shielding makes this technique less attractive. To find
(1 ) σ0 ( )
maximum energy at which MB effect can still be observed, we consider the case where source and scatterer are at
low temperatures. The Debye–Waller factor is then given by
⎡ 3Eγ2 ⎤
f( ) = exp ⎢ − ⎥ (10.102)
⎣ 4 M 0c κθ D ⎦
2
Counting rate (arbitrary units)
10 20 40 60 80 100 120 Energy (keV)

(a)
10 20 30 40 Energy (keV)
(b)
10 20 30 40 50 100 Energy (keV)
(c)
Fig. 10.36 Pulse height spectra (of 57Fe from 57Co source) of g-ray detectors: (a) Proportional counter (b) NaI (Tl) scintillation counter
(c) Solid state Si (Li) Detector.
Mössbauer scattering have been observed with recoil-less fractions as small as 5 × 10−3 and it seems feasible
to lower the limit even further. For a given fcrit, a critical g -ray energy, Ecrit can be calculated from Eq. (10.102):
1/ 2
⎡4 ⎛ 1 ⎞⎤
⎢ M 0c κθ D ln ⎜
2
E crit ⎥
⎣3 ⎝ f ccrit ⎟⎠ ⎦
1/ 2
⎡ ⎛ 1 ⎞⎤
or E crit = 0.33 ⎢ Mass number × θ D × ln ⎜ ⎥ (10.103)
⎣ ⎝ f ccrit ⎟⎠ ⎦
for fcrit = 5 × 10−3, we get for Ecrit,
Ecrit ( f = 5 × 10−3) = 0.76 [Mass number × qD]1/2 (E in keV, qD in K ) (10.104)

In the scattering technique, the re-emission (scattering) of resonant g -radiation in the target is detected. Further,
in many Mössbauer isotopes, the g -radiation is internally connected into an x-ray and conversion electron. In such
cases, the different competing radiation (i.e. nonresonant g -rays, x-rays, conversion electrons) accompanying the
resonant g -nuclear transitions with their characteristic penetration depth can be used for detecting the MB effect.
Because of the low power of re-emitted secondary radiations, only atoms near the surface of the target (0.1−1 mm)
will contribute to the observed resonance effect. For 57Fe, the x-ray energy is 6.3 keV and 7.3 keV for conversion
electron. The geometry for back scattering is shown in Fig. 10.37. The x-ray and conversion electron Mössbauer
back scattering spectra for 57Fe are shown in Figs. 10.38 and 10.39 respectively. For Fe, the internal conver-
sion coefficient is 9, and hence the detection of the 14.4 keV radiation is insufficient. x-ray fluorescence by the
122 keV precursor increases the noise for x-ray detection. In proportional counter, with 2.p back scattering geometry
an argon −10 per cent methane gas is most efficient for x-ray detection [Fig. 10.40(a)], while the conversion electrons
can be detected by a helium −10 per cent methane mixture without x-ray interference [10.40(b)]. MB nuclei suitable
for scattering experiments are 57Fe (136 keV), 190Os (187 keV), 192Os (206 keV), 139La (163 keV), 101Ru (127 keV),
123
Te (159 keV) and 175Lu (114 keV). Scattering is particularly
useful for (a) surface analysis, (b) very thick samples, and (c) small
Mössbauer effects due to the sample size or low ‘f’ factor. By this Detector
technique, material of nearly any shape can be investigated in the Shield
original condition, ‘in situ’, e.g. painting, locomotive boilers, etc.

Corroded surfaces less than 1000 Å can be studied.
In order to obtain the MB spectrum either in the transmission
x-ray
mode or scattering mode, the source is moved towards the target
by the MB drive system. And, the number of resonant g -rays pass-
ing through the target are counted by a g -ray detector as a func- γ-ray
Drive
tion of velocity of source, i.e. the source is moved towards the
target until the velocity at which the maximum absorption/mini-
mum transmission, i.e. dip of g -rays occurs, is determined. It is Source
to be noted that the transmitted g -rays from the absorber are scat-
tered in the 4p solid angle because of which a very small amount Sample
of these rays reaches the detector. Consequently, the number of surface
secondary events recorded at the detector in a collimated trans-
mission experiment is few and are usually neglected. The sign
convention is that the positive velocity is defined as the motion of Fig. 10.37 Schematic representation of geometry for
the source towards the target and vice versa, whereas the negative back scattering.
velocity is defined, as the movement of the source away from the
target and vice versa.
The MB spectrum is then obtained by plotting the counts on the vertical axis as a function of Doppler velocity
on the horizontal axis. A hypothetical spectrum in both the geometries is shown in Fig. 10.41, whereas shapes of
some typical experimentally observed spectra are shown in Figs 10.42(a– d). It is to be noted that (i) the energy of
a photon emitted by a moving source is displaced by an amount (v/c)Eg (Eq. 10.15). If the source is moved with a
constant velocity v, the emitted photons will have an energy = Eg ⎛⎜1+ v ⎞⎟ . What does this energy shift, i.e. Eγ do?
v
⎝ c⎠ c
Emission and absorption lines both show a Lorentzian shape with width Γ. If their centres coincide, the absorption
will be maximized. If one of the lines is shifted, the overlap will be smaller, the absorption will be smaller, and the
Stainless steel
peak
108.0
(a)
100.0
Relative intensity
106.0
(b)
100.0
Source velocity (mm/s)

Fig. 10.38 x-ray Mössbauer back-scattering spectra for (a) 0.2 mil, and (b) 0.5 mil thick iron foils in stainless steel foil. (1 mil = 0.0254 mm).
Stainless steel
peak
(a)
102.0
Relative intensity
100.0
(b)
103.2
100.0
−6.4 −3.2 0.0 3.2 6.4

Fig. 10.39 Conversion electron Mössbauer back-scattering spectra for (a) 600 Å, and (b) 3000 Å thick vacuum-deposited iron on stainless steel.
104.0 (a)
Relative intensity
100.0
(b)
102.0
100.0
−4.0 0 4.0
Fig. 10.40 x-ray and conversion electron Mössbauer back scattering spectra for iron foil using (a) argon—10 per cent methane, and
(b) helium-10 percent methane respectively on a proportional counter with 2p back-scatter geometry.
counting rate in the detector will increase. The resulting transmission line is again a Lorentzian, but with width 2Γ.
These lines are shown in Fig. 10.43. (ii) The doppler velocity needed to obtain a spectrum is in the order of the line
width or of the hyperfine splitting. It is related to the Heisenberg width Γ of the emitted g -photon by.
v 2Γ (10.105)
=
c Eγ
Transmission Scattering
Resonant
Source Source scatterer
moving at moving at
velocity v velocity v
Resonant Detector
absorber Shield
Detector
Transmission count-rate
Scattering count-rate
‘Effect’ ‘Background’
(Recoilless part (Recoilless plus ‘Effect’
resonantly nonrecoilless part) Background
(Resonantly
absorbed) (Due to nonresonant
scattered intensity)
process)
Source velocity v Source velocity v

Fig. 10.41 Comparison of transmission and scattering geometries and hypothetical spectra in both the geometries.
98
Retative counts of rate
Relative counts of rate
97
Γ
96
95 d IS = 0.0 d IS
94
−4 −2 0 2 4 −0.4 −0.2 0 0.2 0.4

Velocity (cm/s) Velocity (cm/s)
(a) (b)
47.0
129XeF
4
29.0
129
XeCl4
δ IS
0 −1.0 −0.8 −0.6 −0.4 −0.2 0 0.2 0.4 0.6 0.8 1.0
Velocity (mm/s) Velocity (cm/s)
(c) (d)
Fig. 10.42 Various shapes of Mössbauer spectrum: (a) 99Ru at 85K (90 keV transition, metallic source and absorber) (b) Ferricinium bromide
(c) 129Xe F4 and 129Xe Cl4(40 keV-transition (d) 57Fe (14.4 keV transition).
and (iii) the Doppler velocity given to the source relative to the target is not to compensate the recoil-energy loss
but to compensate the energy difference between the MB transition states of the source and the target due to the
difference in the electronic environment.
For the transitions with large recoil-less fractions, the numbers N(R) and N(NR) of resonantly and nonreso-
nantly scattered g -rays can be determined by analysing the resonantly scattered radiation by a resonant absorber
rather than by a source. A typical arrangement for such a measurement and the data obtained with it are shown in
Counter rate
Γ Γ 2Γ
vsource
Emission line Absorption line Transmission spectrum
Fig. 10.43 Emission, absorption and transmission lines in a simple Mössbauer arrangement.
Fig. 10.44. With source and scatterer off resonance, the back- Source Resonant
scatterer
ground level ‘B’ is measured. If the source and scatterer are on
resonance, the transmission through a black absorber on and off
resonance is given by
T(0) = B + N(NR) (10.106) Absorber
velocity, v
T(∞) = B + N(NR) + N(R)
Detector
For a thin scatterer, the MB Lamb factor for the absorber is
given by
f abs T ( ) −T ( )
= (10.107)
( f abs ) T( )−B Transmission
T(∞) through moving
Counting rate
It should be noted that an MB spectrum resembles UV, IR absorber with

and NMR spectra that the intensity is plotted against energy of T(0) source and
scatterer on
radiation. resonance
B = Count-rate
with source
10.14.2 Data Computation and scatterer
B
off resonance
Position, width, intensity and splitting are the parameters char-
acterising the MB line. In order to locate the centre of gravity Absorber velocity v
of the spectrum with respect to zero velocity, it is necessary to Fig. 10.44 Geometry for measurement of resonantly
analyse the spectrum and do the line fitting very carefully. It is and nonresonantly g-photons and the resultant hypo-
thetical spectrum.
presumed that the spectrometer is calibrated for velocity. This
is normally done by taking standard spectra for 57Fe such as
K4Fe(CN)6.3H2O (single line), Fe2O3 (magnetically split), FeSO4.7H2O (quadrupole split) and checking the results
with the reported values for isomer shift, magnetic hyperfine splitting and quadrupole splitting. In the absence of
computing facilities, the analysis of the experimental data is done by fitting the experimental points with theoreti-
cally computed Lorentzians. The curves may be computed using the expression
I0
I= 2 4
(10.108)
3⎛ v⎞ 1 ⎛ v⎞
1+ ⎜ ⎟ + ⎜ ⎟
4 ⎝ Γ⎠ 4 ⎝ Γ⎠
Here, I0 and Γ are the intensity and full width of the experimental line respectively.
These days a number of softwares which synthesize spectra from assumed interaction parameters, viz., the isomer
shift d, the hyperfine magnetic field Hint, the principle component of the electric field gradient, Uzz, the polar and azi-
muthal angles relating the direction of Hint and the direction of propagation of g -ray to the principle axes (X, Y, Z) of
EFG and the asymmetry parameter h are available and are being used for the analysis of the spectra, i.e. experimental
data in terms of line positions and relative line intensities.
10.15 APPLICATIONS OF MÖSSBAUER SPECTROSCOPY

The applications of Mössbauer spectroscopy are based on the MB parameters, namely line width, isomer shift,
magnetic and quadrupole hyperfine interactions, and factors influencing them. A wealth of information about
the physical and chemical properties of the MB nuclei in the materials can be obtained from these parameters.
For instance, nuclear physical properties in the excited state, i.e. Lande splitting g factor, magnetic dipole moment,
lifetime, internal conversion coefficient, etc., and physical phenomena such as diffusion and gravitational red shift
are studied by this technique. Further, this technique is routinely applied to study the electronic configuration and
nature of chemical bond not only of coordinated and organometallic compounds but also of molecular complexes
and oxidation and corrosion phenomena on surfaces as well. This technique has also made deep inroads to the
field of biological sciences. It not only enables us to understand the electronic structure of biological molecules
in vivo and in vitro but also about their coordinating behaviour.
Furthermore, MB spectroscopy has made a unique contribution to the study of conformational changes of
organometallic compounds resulting from the change of environment. Such studies are important because of the
ambiguity in the structural information derived from x-ray diffraction data on single crystal and from NMR and
IR measurements on solutions. Since MB spectrum can be recorded both on neat solids and frozen solutions,
this tool bridges the gap between x-ray diffraction and spectroscopic, i.e. IR, NMR measurements to resolve the
ambiguity in structural information so derived.
10.15.1 Lifetime of the Excited State

The relation between apparent width, Γapp (sum of the emission and absorption half-widths, i.e. 2Γ and effective
absorber thickness Ta is expressed by the equation (10.97), i.e.
Γ app
= 2.00 + 0.27 Ta , 0 ≤ a ≤5 (10.109)
Γ nat
The expression indicates that
⎛ Γ app ⎞
⎜⎝ Γ ⎟⎠ ≈2 (10.110)
nat Ta → 0
Γ app
Thus, from the plot of Γapp, as a function of Ta, it is possible to obtain the true value of Γ nat = (Ta → 0).
The lifetime of the excited state is then obtained from Eq. (10.8), i.e. 2
6.58 × 10 −16
τ (s) = (Eq. 10.8)
Γ app ( )
The method only provides a lower limit to lifetime as the possibilities of intrinsic broadening from unresolved
hyperfine effects cannot be eliminated.
10.15.2 Determination of Quadrupole Moment Ratio Qe/Qg

Since the quadrupole hyperfine structure in the MB spectrum arises because of the transitions between the
sublevels of the excited state and the ground state of the MB nucleus, so it is possible to determine only the ratio
Qe/Qg but not the values Qe and Qg separately.
We know that for axial symmetric fields (i.e. h = 0),
e 2Qq
EQ = ⎡⎣3M
M I2 − I (I + 1) ⎤⎦ (Eq. 10.77)
4II ( I −1)
Since the MB nucleus experiences this interaction in both its ground and excited states, Eq. (10.77) for the
ground and excited states is expressed as
e 2qQ g
EQ ( g ) = ⎡3M I2g I g (I
( g ) ⎤⎦ (10.111a)
4I g ( I g ) ⎣
e 2qQe
EQ (e ) = ⎡⎣3M I2e I e (I
( e ) ⎤⎦ (10.111b)
4I e ( I e )
Thus, the energy of g -ray for a transition from the ground state sublevel M I g to the excited state-sublevel M I e
will be given by
δ e 2q
Eγ ( g e) E0 + E0 + [Q C (I e , M I ) − Q g C (I g , M Ig )] (10.112)
c 4
⎡⎣3M I2 − I (I
( ) ⎤⎦ δ
where C (I , M I ) = , and E is the isomer shift-energy term.
I( I ) c 0
The positions of the quadrupole hyperfine lines in the MB spectrum can now be represented as
Δ Q ( g → e ) = Eγ ( g → e ) − E 0
δ e 2q
= E0 + [Q C (I e , M I ) − Q g C (I g , M Ig )]
c 4
ce 2qQ g ⎡Q ⎤
or δ( )
4 Eγ
⎢ ( e, Ie ) ( g , Ig )⎥ δ IS (10.113)
⎣ Qg ⎦
Here, d(g → e) in (mm/s) represents the shift of the transition from the ground state level |Ig, M I g > to the
excited-state level |Ie, M I e >, dIS is the isomer shift of the MB nucleus with respect to the source.
Equation (10.113) shows that the line positions change as a function of Qe/Qg. The ratio Qe/Qg and isomer shifts
dIS can be obtained from the observed values of d(e → g) for the lines by least square fit to Eq. (10.113), which rep-
resents a straight line with slope Qe/Qg and an intercept dIS. Further, from the ratio Qe/Qg, the quadrupole moment
Qe can be determined provided the ground-state quadrupole moment Qg is known from other experiments.
Note that for MB nuclei with h ≠ 0, Eq. (10.113) becomes complex, i.e. nonlinear due to the term h and the fitting
of the observed line positions to the differences in energy for the sublevels of the ground and excited states becomes
cumbersome, e.g. MB spectra of 129I in solid I2 (h = −0.16) gives Qe/Qg = 1.232 ± 0.004 and 1.237 ± 0.002.
Problem 10.5: (a) The MB spectrum of 129I in KIO3 (Fig. 10.45) (h = 0, Ie = 5/2, Ig = 7/2, ZnTe source) consists
of eight quadrupole hyperfine lines (left to right) with relative intensity ratios:
⎛ 3 5⎞ ⎛ 5 5⎞ ⎛ 7 5⎞ ⎛ 1 3⎞ ⎛ 3 3⎞
1⎜ ± − ± ⎟ , 6 ⎜ ± − ± ⎟ , 21⎜ ± − ± ⎟ , 3 ⎜ ± − ± ⎟ , 10 ⎜ ± − ± ⎟ ,
⎝ 2 2⎠ ⎝ 2 2⎠ ⎝ 2 2⎠ ⎝ 2 2⎠ ⎝ 2 2⎠
⎛ 5 3⎞ ⎛ 1 1⎞ ⎛ 3 1⎞
15 ⎜ ± − ± ⎟ , 18 ⎜ ± − ± ⎟ , 10 ⎜ ± − ± ⎟ .
⎝ 2 2⎠ ⎝ 2 2⎠ ⎝ 2 2⎠
The observed position of lines yield Qe/Qg = ±1.23 and dIS = 0.156 cm/s. If the measured (at 80 K) K127IO3
Q129
quadrupole coupling constant is 996.7 MHz and the observed quadrupole moment ratio = + 0.70121, what
will be the value of eqQ (129)? Q127
e
(b) A comparison of the MB spectra of 129I in KIO3 (h = 0) and I2 (h = −0.16) indicates that the two spectra are
approximately inverted from right to left. If in the former, eqQ is positive and the quadrupole moments of both the
ground state and the 26.8 keV excited state are negative, what will be the signs of q in case of KIO3 and I2?
Solution (a) From,
eq Q127 = 996.7 MHz
Q129
and = 0.70121
Q127
we obtain the K129IO3 ground-state coupling constant

eq Q129 = 0.70121 × 996.7 (MHz) = 698.9 MHz
Further, since the quadrupole moments ratio
Qe
= 1.23
Qg
Therefore, the K129IO3 excited state coupling constant
eqQe = eqQg ×1.23 = 698.9 (MHz) × 1.23 = 859.6 MHz
1 2 3
± 5
2
4 5 6
± 3
2
5+
2
7 8
± 1
2
1 3
18
M*I (e)
10 10
6
21 15
Eso
(o) ± 7
Eabs 2
± 5
2
7+ ± 3
2
2
± 1
2
MI (g)
Fig. 10.45 A nuclear energy-level diagram illustrating the quadrupole splitting for I in KIO3. 129
(b) In case of KIO3, eqQ is positive since both Qe and Qg are negative and therefore q should always be negative
for eqQ to be positive. The inversion suggests that the sign of eqQ should be negative in case of I2 with the result q
is positive in I2. Further since the field gradient at the iodine nucleus is due to a p-electron vacancy which behaves
like a positive charge, we expect a positive value of q, i.e.
e ∂ 2U 2e
U =+ and =+ 3.
r ∂r 2
r
10.15.3 Determination of g-Factors and Nuclear Magnetic Dipole Moments

Just like quadrupole hyperfine lines in the MB spectrum, the magnetic hyperfine pattern is also caused by the
transitions between the sublevels of the excited state and the ground state of the MB nucleus because of which it
is possible to determine only the ratio me/mg but not the values of me and mg separately.
Now, in order to determine me/mg we are to work out a relationship between me/mg and the position of the mag-
netic hyperfine lines. This is done as follows.
The magnetic hyperfine Hamiltonian for a nuclear magnetic dipole mI in a magnetic field H is given by
Hm μI H = − g β Iˆ Hˆ (Eq. 10.61)
and the energy of the magnetic substates is expressed as
E g N βN H M I (Eq. 10.63)
where the terms and symbols have their usual meanings. It can be readily proved that the 2I + 1 magnetic substates
are equally spaced with a separation of gN bN H between the substates.
Further since the MB nucleus experiences this interaction in both its excited and ground states, so Eq. (10.63)
for the ground state and the excited state is expressed as
E (g ) g g βN H M I g (10.114)
E ( ) g e βN H M I e (10.115)
Thus, the energy of g -ray for a transition from the excited state sublevel M I e to the ground-state sublevel M I g will be
δ
Eγ (e g) E0 + E g g βN H M I g g β H M Ie (10.116)
c
δ
where, E 0 is the isomer shift-energy term.
c
Note that the NMR spectroscopy measures the energy differences between the magnetic substates of the ground
state.
ΔE
Δ g N βN H , [M I ( = I 1) M I ( = I )] (Eq. 7.16)
The position of the magnetic hyperfine lines in the MB spectrum can be expressed as
δ ⎛ g ⎞
Δ m ( → g ) = Eγ (e → g ) − E = E − βN H g g ⎜ M I g − e M I e ⎟
c ⎝ gg ⎠
cβ N H g g ⎛ g e ⎞
or δ( )= ⎜⎝ g M I e M I g ⎟ + δ IS (10.117)
Eγ g ⎠
Here, d(e → g) in mm/s represents the shift of the transition from the excited state level |Ie, M I e> to the ground-
state level |Ig, M I g >, dIS is the isomer shift of the MB nucleus with respect to the source.
Equation (10.117) shows that the line positions change as a function of ge/gg and isomer shiff can be obtained
from the measured values of d(e → g) for the lines by least square fit to Eq. (10.117) which represents a straight
line with slope ge/gg and an intercept dIS.
μ
Once ge/gg is known, one can readily obtain the magnetic dipole moment ratio e from the relation
μg
μe g e I g
= (Eq. 10.70)
μg g g I e
Unlike the quadrupole case, the ground state g-factors and the magnetic dipole moments mg are well known
since they can be determined very accurately from NMR measurements, i.e.
ΔE
gN = and μ g = Ig g βN
βN H
By using the known ground-state value of m, i.e. mg , one can determine the value of me from Eq. (10.70), as well
−g g I g
as the field strength H; H = .
μg
Further, it is to be noted that for the 2+ → 0+ even–even nuclei, the magnetic interaction in the ground state will
not exist, and the hyperfine splitting will only be from the excited state. The magnetic measurements will then
only give a value for ge bN H. Thus in order to obtain the value of ge or H, one must know that value of H or ge from
another measurement.
10.15.4 Determination of Internal Conversion Coefficient

The intensity or area of an absorption line is a function of the various parameters involved in the effective thicknees Ta
⎡ λ2 ( 2I e + 1) ⎤
⎢ = nσ 0 f at aaa (10.96) h σ0 = ⋅ (10.28) ⎥ of a Mössbauer absorber.
⎣ 2π (1 + α t ) ( 2I g + 1) ⎦
Thus the internal conversion coefficient at can be determined from the area of an absorption line which is given by
+∞
A fs ∫ ddEE ⎡⎣ ( E f a t a naa )⎤⎦ (10.118)

−∞
σ 0Γ 2
where, σ( ) = (10.119)
4 (E E0 ) + Γ 2
2
By evaluating the transmission integral and by measuring the ratios of the areas in order to eliminate the recoil-
free fraction of source, fs, one can determine s0 and hence from Eq. (10.28), the internal conversion coefficient
at, If absorbers of varying thickness are used, the amount of photoelectron absorption from atomic electrons will
vary from absorber to absorber. This results is an inaccurate values of area, and hence the value of at. Such an
effect can be eliminated by using a series of absorbers having the same thickness of MB atoms but whose density
of atoms varies.
10.15.5 Determination of d R/R or Nuclear Radius Changes

from Mössbauer Isomer Shift
δR
We know that isomer shift is a product of nuclear term, and chemical term, |ys (0)abs|2 −|ys(0)so|2. So, if either
R
of the two factors is determined independently by some method, the other can be determined from the isomer
shift. The chemical term suggests that isomer shift is proportional to the difference in the total s-electron density
at the source and the absorber nuclei. Thus the change in the radius, dR/R can’t be determined independently of
the chemical environment, which means that electron density has to be estimated by molecular orbital methods
δR
in at least two compounds before a value of can be estimated. The values of electron densities for 1s, 2s; 3s,
R
4s electrons for different 3d configurations of the MB celebrated iron atom computed by Hartree Fock method
n
δR
have already been listed in Table 10.3. for a number of nuclei, namely 57Fe, 119Sn, 121Sb, 127I, 129I, etc., have
R
been estimated from the isomer-shift data. It is found that the values of dR/R determined from chemical isomer
shift data are sensitive to the various methods of computation of |ys(0)|2. It appears, therefore, that isomer shift is
not an ideal parameter to determine dR/R. It is better to determine dR/R values independently and then use these
values to interpret |ys(0)|2, the s-electron density at the nucleus and hence the nature of chemical bond. dR/R can
be determined by measuring internal conversion coefficients for valence electrons as a function of the chemical
state of the element. dR/R values in deformed nuclei have been estimated from the energy shifts of muonic x-rays.
For 57Fe (14.4 keV), the value of dR/R determined from MB isomer shift is −1.28 × 10−3. The values of dR/R
determined by molecular orbital and density calibration methods lie in the range −0.4 to −5.4 × 10−4. The nega-
tive value of dR/R indicates that radius of the 14.4 keV excited state (I = 3/2) of 57Fe is smaller than the radius of
the ground state (I = 1/2). On the other hand, in case of divalent 119Sn, the value of dR/R determined by different
methods (lie in the range 0.3 − 3.5 × 10−4) and from isomer shifts 9.21 × 10−5. The positive sign of dR/R in case of
divalent 119Sn shows that the radius of the excited state (I = 3/2) is larger than its counterpart in the ground state
(I = 1/2). The values of dR/R in 57Fe and 119Sn are likely to be changed if their relativistic correction factors S(Z)
are taken into account. The values of S(Z) for some MB isotopes are Fe(26) = 1.29; Ni(28) = 1.33, Zn(30) = 1.39,
Sn(50) = 2.30, I(53) = 2.53.
The discrepancies in the values of (dR/R) obtained from the isomer shift data are attributed to the inadequate
knowledge of the charge density of an ion when it is situated in a lattice. This difficulty is solved by making ratios
of the dR values from the isomer shift data obtained from two different Mössbauer levels.
(δ R / R )2 Eγ δ 2 Z 1R12 S ( Z 1 ) Δψ ( Z 1 )
2
= 2
⋅ ⋅ ⋅ ⋅ (10.120)
(δ R / R )1 Eγ 1
δ1 Z 2 R 22 S ( Z 2 ) Δψ 2 ( Z 2 )
For the case of two excited states in the same nucleus, e.g. the 97.4 and 103.2 keV levels in 153Eu, Eq. (10.120)
reduces to
(δ R / R )2 Eγ δ2
= 2
⋅ (10.121)
(δ R / R )1 Eγ 1
δ1
For the case of two different isotopes of the same element, e.g. the 57.5 keV level in 127I and the 27.7 keV level
in 129I, Eq. (10.120) becomes
(δ R / R )2 Eγ δ 2 R12
= 2
⋅ ⋅ (10.122)
(δ R / R )1 Eγ 1
δ1 R 22
This method can also be extended to isoelectronic compounds, e.g. the ratios of nuclear radii for nuclei from
tin, Z = 50, to Xe, Z = 54; i.e. for isoelectronic pairs such as (129I O4)− vs (129XeO4), 129I− vs 129Xe0 and K2 125Te O4
vs K 129IO4, etc. The calculations with Hartree–Fock self-consistent field wave functions show that the ratio of
electron densities for any pair of iso-electronic compounds is approximately a constant Δy 2(Z + 1)/Δy 2(Z) = 5/4
for the various 5sm 5pn configurations.
Problem 10.6: The isomer shift for Fe3+ ions has been observed at about 0.5 mm/s and around 1.2 mm/s for
Fe2+ ions. (a) Account for the different values of isomer shifts of Fe2+ and Fe3+ ions. Using the required data from
Table 10.3, determine the (b) contribution of chemical term to isomer shift, (c) contribution of nuclear term to
isomer shift, and (d) effect of relativistic term on the nuclear term.
Solution (a) The electronic configuration of iron in the various oxidation states is Fe (metallic): [Ar] 3d74s1,
++
Fe , Fe : [Ar] 3d6, Fe3+ : [Ar] 3d 5. Normally s-electron density at the nucleus of Fe3+ ion in a salt is expected to
2+
be the same as that at the nucleus of Fe2+ ion. The difference in the isomer shifts of Fe2+ and Fe3+ ions suggests that
the shielding effect of 3d6 electrons of Fe2+ ion and 3d5 of Fe3+ ion on the respective 3s electrons is different.
(b) The contribution of chemical term to isomer shift is determined from the expression
2 2 2
ψ ( 0) ψ 3 ( 0 ) 2 + − ψ 3 ( 0 ) 3+
From Table 10.3, we write
2
ψ 3 (0) 2+ = 68.274 au
2
ψ 3 (0)3+ = 69.433 au
Thus,
2
ψ 3s (0) = (68.274) − (69.433)
= −1.159 au.
δR
(c) can be determined from the relation
R
4π 2 2
⎛ δR ⎞ 2
v IS = ⎜⎝ ⎟⎠ ψ 3s (0) (Eq. 10.54)
5E γ R
2
4 × 3.14 × 3 × 1010 × 26 × ( 4.8 × 10 −10 ) × ⎡⎣1.4 × 10 −13 × (57)1/ 3 ⎤⎦
10 2
⎛ δR ⎞ 2
= ×⎜ ψ 3s ( 0 ) ( )
5 × 14.4 10
10 3
1.602 × 10 1122
× (0.53 10 ) 8 3
⎝ R ⎟⎠
⎛ δR ⎞ 2
or v IS = 381.9 ⎜ ⎟ ψ 3 (0) (mm/s)
⎝ R ⎠
Now,
v IS v IS ( Fe 2+ ) − v IS (Fe
( Fe 3+ ) = 1.20 − 0.50 = 0.70 mm/s
2
and from Part (b), ψ 3s (0) = −1.159 au.
Substituting these values in the above expression, we obtain
δR 0 70
= = −1 58 × 10 −3
R 381.9 × ( −1.159)
The negative value of dR/R indicates that the radius of 14.4 keV excited state (3/2) of 57Fe is smaller than the
radius of the ground state.
(d) For 57Fe (Z = 26), the relativistic correction factor S(Z) = 1.29
Therefore,
2 2
ψ ′ (0) = ψ 3s (0) × S ( Z ) = −1.159 × 1.29 = −1.495
and
δR 0 70
= = −1.23 × 10 −3
R −1.495 × 381.9
δR
The increase in the value of indicates that the increase in radius of 14.4 keV excited state of 57Fe is rela-
R
tively more than that of the ground state when the relativistic correction factor is taken in consideration.
10.15.6 Determination of Gravitational Red Shift

The change in energy of g -ray photon travelling between source and absorber of different gravitational field is
termed gravitational red shift and is expressed as
dg = gg .d mass of g -photon
⎛ Eγ ⎞
= gγ d ⋅ ⎜ 2 ⎟ (10.123)
⎝c ⎠
Here, gg is the acceleration of the photon in the gravitational field and d is the distance between the source and
the absorber.
This effect was first predicted by Einstein in 1907. The magnitude of the shift is very small (i.e. 1.1 parts in
10−16 m−1).
The experiments were performed using the isotope 57Fe (14.4 keV) over a distance of d = 22.5 m. The source
is mounted at a distance d above the absorber so that the photon travels through the gravitational field which
would introduce an acceleration of gg in the atom, if it is free to fall. The g -photon interacts with the gravitational
field and a change in the g -ray energy occurs and thus a change in position of the MB resonance line. The shift
has been found to be 0.9970 ± 0.0076 at a height separation of 22.5 m. For a shift of full line width, d ≈ 3000
m, the experiments become unfeasible. Other isotopes like 181Ta can be used because of their very small natural
line widths.
10.15.7 Diffusion in Solids and Liquids

The MB resonance of hydrated iron (III) sulphate in glycerol (in the temperature range −100 to 0°C) shows the
presence of Fe2+ and Fe3+ ions at −100°C at which temperature the matrix is frozen. The resonance becomes
broadened with the growth in temperature and at 0°C when the sample becomes liquid, the very broad lines (Fe2+
doublet + Fe3+ singlet) have been observed. The relationship between viscosity of the liquid and the broadening
of the MB line can be thought to be related to the relative importance of jump and continuous/Brownian types of
diffusion processes.
MB spectroscopy can be used to study diffusion and related processes in solid systems since the appearance of
spectrum depends upon the lifespan of the motion executed by the resonant atom. If the nucleus is vibrating har-
monically about a fixed position at a frequency which is high compared with the inverse of excited state lifetime
then the intensity of resonance line decreases without any line broadening. The decrease in the MB Lamb factor
is given by the Debye−Waller formula
2 2
λγ2
f = e −4π x
(10.124)
where, x2 is the mean square amplitude of oscillation in the direction of observation of g -ray, λg is the wavelength
of the g -photon. This means that the value of x of the order of 10 pm can be measured with MB isotopes with
excited state lifespans similar to those of 57Fe or 119Sn.
In continuous diffusion, the broadening of MB line without reduction in overall recoil-less factor and
change in the shape of the basic Lorentzian line occurs. The increase in experimental line width at half height
is given by
2
⎛ 2π ⎞
Δ exp = 2h ⎜ ⎟ Dc (10.125)
⎝λ ⎠ γ
Here, Dc is the self-diffusion coefficient. Dc value of the order of 10−15 m2s−l gives broadening of the order of
natural line width for 57Fe or 119Sn sources.
At elevated temperatures, the relaxation time for diffuse jumps of the resonance atoms is in the same order
of magnitude as the mean lifetime of the excited state. The line broadening without any change in intensity and
Lorentzian line shape for random jumps of a diffusing MB atom is given by
h
Δ exp = ( − g ) (10.126)
πτ 0
where, t0 is the mean stay time of an atom at a lattice site (1/t0 is the jump frequency), values of t0 equal to 280 ns
and 53 ns give broadening equal to natural line width for 57Fe and 119Sn respectively. ag is a geometric factor. The
diffusion coefficient is given by
1 2
Dj d j fc (10.127)
6τ 0
where, dj is the jump distance and fc is a factor which describes the correlation in direction between successive
jumps of one and the same atom. Lumping together Eqs (10.126) and (10.127), we obtain
6h
Δ exp = Dj( − g ) (10.128)
d 2j fc
The diffusion coefficient (and jump frequency) and jump direction can be determined from this expression.
This unique feature of the Mössbauer effect is very useful in the study of diffusion mechanism especially for
the anisotropic diffusion including surface diffusion. Diffusion phenomena of Fe in b-Ti have been studied by
MB technique. Tracer measurements for the diffusion in b-Ti indicate an anomaly in the log D versus 1/T. The
observed diffusional broadening could be analysed in terms of an appropriate diffusion mechanism for this alloy
system. A 57Co/Cu source or a 57Fe/Cu absorber shows line broadening between 1000−1060°C due to solid state
diffusion, although data show deviation from simple diffusion jump model.
10.15.8 Study of Surfaces

MB spectroscopy represents an important addition to other surface spectroscopy tools like LEED (Low-
Energy Electron Diffraction), ESCA (electron spectroscopy by chemical analysis) and Auger electron spec-
troscopy and can provide information which classical methods can’t. Because of its sensitivity to strength and
angular distribution of binding, to magnetic and electric fields, and to the s-electrons density at the nucleus,
this tool can be used to study surfaces. The fraction of recoil-free g -rays emitted depends on the mean square
displacement of the emitting nucleus from its equilibrium position. The displacement will be different on
surfaces than in the bulk of a crystal and will depend on the direction of the g -ray emission. The structure
of the MB line will depend on the magnetic fields and on the electric-field gradient at the surface. The vari-
ous components of a line split by a quadrupole interaction will have different directional distributions with
respect to the field axis. An anisotropy in the binding at the surface may result in isomer shift. The difference
between the mean square velocity of surface and bulk atoms will lead to a second-order Doppler effect. For
experimental studies, ultra-high vacuum should be maintained, the surface must be checked and the radioac-
tive material must be deposited in less than a monolayer. The surface temperature should be kept below the
room temperature, otherwise the tracer atoms may rapidly diffuse to a crack or a surface imperfection and
lead to erroneous results.
Before discussing some of the experimental facts, let us first describe briefly the basic aspects of the theory
related to the MB effect on surfaces.
The fraction ‘f’ of recoil-less emitted g -rays is expressed as
⎡ − Eγ2 x 2 ⎤
f = exp ⎢ ⎥ (Eq. 10.21)
⎢⎣ ( / 2π ) ⎥⎦
2
The three typical positions of a nucleus undergoing, an MB γ

transition have been shown in Fig. 10.46. n
According to Eq. (10.21), the recoil-less fraction ‘f’ is smallest A
in the direction of the largest excursion for the emitting nucleus.
Thus, in a cubic crystal, f will be isotropic for a nucleus C resid- B Surface
ing in the bulk. The fraction ‘f’ will be smaller parallel to the
normal n than perpendicular to it, when the nucleus B resides
in the surface whereas reverse will be true for the nucleus A on C
the surface. Furthermore, fC will be greater than either fA or fB.
Further, if we assume that for a periodic lattice in the high- Bulk
temperature region, all frequencies contribute equally to the cal-

Fig. 10.46 Three typical positions of an MB nucle-
culation of ‘f’ then on the basis of Einstein model of a solid, it is us adsorbed on a surface.
found that a single typical frequency related to the fraction f of
g -rays emitted without recoil is given by
1
v typical v max (10.129)
3
or equivalently,
1
θE θD (10.130)
3
Accordingly, a nucleus in a crystal can be considered as a single particle of mass m located in an anisotropic
harmonic potential with force constants k1, k2 and k3.
In the high-temperature limit, the recoilless fraction f is expressed as
⎡ ⎛ h⎞2 T ⎧ 1 1 1 ⎫⎤
f = exp ⎢ − ⎜ ⎟ ⎨ 2 siin θ cos ϕ + 2 si θ ϕ+ cos 2 θ ⎬ ⎥
2 2 2 2
(10.131)
⎢⎣ ⎝ λ ⎠ m ⎩ θ E ( ) θ E ( ) θ 2
E ( 3) ⎭ ⎥⎦
where λ is the wavelength of the g -ray. For surface studies q is assumed to be the angle between the surface
normal and the direction of the emitted g -ray. qE(i) is an Einstein temperature defined by
1/ 2
h ⎛ ki ⎞
θ E (i ) = ⎜ ⎟ (10.132a)
2πκ ⎝ m ⎠
Moreover,
i 1
x γ2 ∝ ∝ (10.132b)
θ 2
E (γ ) kγ
where kg is the effective force constant in the direction of the g -ray emission.
The isomer shift is given by
δ IS
2
5
π Ze 2 ( − ) ⋅ ⎡⎣ ψ (0) 2
ψ ( 0 ) so ⎤
2
⎦
(Eq. 10.40)
where the terms and symbols have their conventional meanings.

A given MB transition investigated on surfaces and in the bulk of a crystal will give information about the
electronic functions in the various situations. This effect will be temperature independent provided a crystal phase
transition does not take place.
The second-order Doppler shift is given by
Eγ
δ SOD = 2
v2 (Eq. 10.44)
2c
Here,
3 ( h / 2π ) 2 ( k 1 + k 2 + k 3 )
v2 = + (10.133)
m 12 2κTm 2
Let k1 + k2 + k3 = keff where keff for atoms in the direction of the emitted g -rays in the bulk, in the surface and on
the surface of a single cubic lattice have been approximated to be in the ratio 3 : 2 : 1.
Lumping together Eqs (10.44) and (10.134) we obtain
Eγ ⎡ 3κT ( h π ) 2 k eff ⎤
δ SOD = ⎢ + ⎥ (10.134)
2c 2 ⎣ m 12κTm 2 ⎦
The central line shift dc is
dc = dIS + dSOD (10.135)
3κTEγ 1
If we assume dIS to be independent of temperature and plot δ c − 2
against , the slope of Eq. (10.135)
2mc T
( / π ) 2 k eff
will be equal to . Thus from the slope one can determine the value of keff, i.e. the effective force
122κ m 2
constant which gives us information about the details of surface lattice binding.
The effect of adsorption on quadrupole coupling constant can be estimated from the electric field splitting of
the nuclear energy levels as follows.
We know that for axial symmetric fields,
e 2qQ
EQ = ⎡3M
4II ( I −1) ⎣
M I2 − I (I 1) ⎤⎦ , [ U zz = eq ] (Eq. 10.77)
⎛ 3⎞ ⎛ 1⎞
For 57Fe, ΔEQ
Δ EQ ± ⎟ − EQ ± ⎟
⎝ 2⎠ ⎝ 2⎠
2
e qQQ 1 ∂U
2
⎡ ∂ 2U ⎤
= = eQ 2 ⎢ ∵U zz = = eq ⎥ (Eq. 10.79)
2 2 ∂z ⎣ ∂z 2
⎦
The actual splitting is larger because of polarisation of the d-electrons by external field gradient. The polarisa-
tion is taken into account by multiplying Eq. (10.79) by Sternheimer antishielding factor, g ∞.
The order of magnitude of a field gradient on a surface is estimated by assuming a charge −e placed at a typical
lattice spacing ‘a’ above the surface, simulating a missing positively charged ion. We then obtain
∂ 2U −2e
= 3 (10.136)
∂z 2 a
Assuming, a = 3 × 10−8 cm, Sternheimer antishielding factor for Fe = 15, quadrupole moment eQ = 0.2 b, we
get from Eq. (10.79),
ΔEQ ≈ 10−8 eV
Since the natural width of the MB line in 57Fe is about 10−8 eV, surface field gradients may be observable.
The field gradient at a surface is expected to have a well-defined axis. The two components that result from
the splitting of the 57Fe 14 keV line by quadrupole interaction will exhibit different directional behaviour. The
directional dependence of intensities of the two lines is
I ±1/ 2, ± 3 / 2 I π 1 + cos 2 θ
= = (Eq. 10.80)
I ±1/ 2, ±1/ 2 I σ ⎛ 2 2 ⎞
⎜⎝ + sin θ ⎟⎠
3
Thus, the direction of the field gradient can be obtained from the intensities of a quadrupole split MB line in
different directions.
Let us now examine the MB spectra of 57Fe on the surface of polycrystalline tungsten (in the temperature range
100−500 K) and at angles of 0° and 60° relative to the surface normal. The spectra as deduced from least square
fit computer analysis consists of three unresolved lines. The middle line conforms to the measured spectra of 57Fe
in the bulk of tungsten and the outer two lines are attributed to the surface effects. MB spectrum of a single crystal
surface showed no central peak substantiating there by the assumption that the central line is due to diffusion of
57
Fe into the bulk of tungsten.
The asymmetry of the surface spectra, the angular dependence of shape and the magnitude of splitting of the
two lines are attributed to the surface quadrupole splitting. EFG at the nucleus is −3.6 × 1017 V/cm2 perpendicular
to the surface for quadrupole moment = 0.2 × 10−28 m2 of the 3/2 state in 57Fe. The average ratios of the intensi-
ties of the two lines are 2.1 at 0° and 1.4 at 60°. The expected ratios with corrections for experimental facts are
between 2.5 and 1.9 at 0° and between 0.9 and 1.0 at 60°. The discrepancies at 60° may be attributed to the dif-
ficulties in the analysis of the three unresolved lines.
The intensity of the central component of the unresolved line depends on temperature and qD has been found to
be 403 ± 70 K whereas MB spectrum of 57Co diffused into polycrystalline tungsten gives a value of qD = 406 ± 12 K.
The value of qD as determined from the combined intensity of the outer components of the line is 340 ± 30 K
perpendicular to the surface and, 273 ± 30 K at 60° from the normal. It is to be noted that solid angle has not
been taken into account while calculating these values. When solid angle is considered, the effective value of
qD = 354 ± 30 K perpendicular to the surface and 255 ± 30 K parallel to the surface. It can easily be shown by
Eq. (10.132b) that these results correspond to the following mean square displacement ratios (recall that the mean
square displacement in the high temperature limit is inversely proportional to the square of the effective Debye
temperature).
x2
= 1.9 ± 0.4, (1.2)
x ⊥2
x ⊥2
2
= 1.3 ± 0.2 ( 2) (10.137)
x bulk
x2
and 2
= 2.5 ± 0.5 ( 2.5)
x bulk
The quantities with in braces are because of an extension of the ‘bond counting’ analysis to an atom on the sur-
face of a body-centred cubic lattice such as tungsten. More sophisticated assumptions about the effective bonding
in various directions need to be made to have good agreement with the experimental data.
The isomer shift relative to iron at 300 K is 0.0151 ± 0.0002 cm/s in bulk and 0.018 ± 0.002 cm/s on the sur-
face. The position of the central component matches with that of the bulk measurement.
This tool is also suitable to study oxidation and corrosion phenomena on surfaces, the initial states of internally
oxidised atoms, and internally oxidised phases. The isomer shift of internally oxidised Sn atoms of an Ag−Sn
alloy showed that the ionization state is close to +4 and that most of the oxidised Sn has been precipitated as SnO2
particles.
The internal oxidation of Fe to FeO and subsequently to Fe3O4 and CuFeO2 which occurs by heat treatments
of Cu–Fe samples in a relatively low partial pressure atmosphere of O2, can be followed by MB spectroscopy.
Although FeO has the cubic NaCl structure, it exhibits, at room temperature, a quadrupole split spectrum with an
isomer shift characteristic of divalent Fe. The removal of the local cubic symmetry in the iron deficient compound
(Fe0.87 − 0.95 O) could be the reason for the presence of the electric field gradient at the site of the nucleus.
Fe3O4 exhibits, at room temperature, two superimposed six-line Zeeman spectra. The two patterns are because of
Fe3+ at tetrahedral sites (Hint = 492 ± 5 kOe) and Fe2+ and Fe3+ with rapid charge exchange at octahedral sites (Hint = 464
± 5 kOe). The terminal oxidation product has been identified as CuFeO2, which occurs naturally as mineral delafos-
site and in copper smelter furnance slags. The isomer shift observed is characteristic of trivalent iron which provides
evidence of ionic charge state of delafossite: Cu+Fe3+O2 (not Cu2+Fe2+O2). A magnetic transition has been found with
a Neel temperature of 190 K.
The following reactions have been summarised in the Cu−Fe system.
Fesolution ⎯annealing → γ − Fe(pption ) (i)
γ F deformation
o )
→ α − Fe ( phase transformation (ii)
1
Fe + O 2 annealing
FeO (internal oxidation ) (iii)
2
1
3FeO + O 2 annealing
Fe 3O 4 (internal oxidation ) (iv)
2
Fe3O 4 3Cu + O 2 annealing

3CuFeO 2 (internal oxidation ) (v)
10.15.9 Structure of Coordination Compounds

The tool of MB spectroscopy is routinely employed these days to
study the structure of coordination compounds. Besides iron, MB OXYGEN
investigations have been made with coordination compounds of
+ IRON
Ni, Ru, Os, Ir, Pt, Au, Eu and Np among the transition elements,
+
and of Sn, Sb, Te, I, Xe and Kr among the nontransition elements.
The importance of this technique regarding structural elucidation of
coordination compounds is demonstrated with some representative
examples cited here. Type-1 Site
(a) Crystalline Structure The MB spectrum of polycrystal-

line hydrated ferrous formate (Fe(HCOO)22H2O) at 6, 77 and 298
K consists of four equally sharp lines. The inner and outside lines OXYGEN
are attributed to two quadrupole doublets arising from different + + IRON
electric field gradients corresponding to two inequivalent sites. The
results are listed in Table 10.13. The smaller quadrupole splitting is WATER
expected to be associated with the more symmetrical site. The near-
est-neighbour symmetry at the two sites is Oh and D4h respectively Type-2 Site
and is shown in Fig. 10.47. These MB spectral observations conform Fig. 10.47 Models of the approximate nearest
to x-ray diffraction data. neighbour symmetry at the two ferrous ion sites in
hydrated ferrous formate.
Table 10.13 Isomer shifts (mm/s) and quadrupole splittings (mm/s) of ferrous formate.
Temperature (K) ΔEQ δIS (Cr) δIS (Fe)
298 0.59 1.34 1.19
2.97 1.39 1.24

77 1.37 1.56 1.41
3.42 1.56 1.41
6 1.48 1.48 1.33
3.36 1.53 1.38
(b) Complex Isomerism Compounds having the same stoichiomet- H2O

ric composition but different arrangement of ligands are called isomers. H2O H2O
The MB spectrum may be of help to decide about the correct structure

of the isomer, e.g. hydrated ferric chloride (FeCl3.6H2O) may have an Fe
iron ion surrounded by an octahedral environment of water molecules as

H 2O H2O
shown in Fig. 10.48. However, the MB spectrum of this molecule exhibits
much quadrupole splitting (ΔEQ = 0.91 mm/s, dIS (Pd)dIS (Fe) = 0.42/0.53 H2O
mm/s at 77 K, and ΔEQ = 0.97 mm/s, dIS(Pd)/dIS(Fe) = 0.27/0.45 mm/s Fig. 10.48 Model of the [Fe(H2O)6]3+ ion
at 300 K), suggesting that the symmetry around the ferric ion should be with octahedral structure.
lower than octahedral. The x-ray diffraction studies have then shown that
this molecule has a trans octahedral arrangement 2Cl− and 4H2O around
H2O
the iron atom as shown in Fig. 10.49.
Cl
This result is of interest in understanding the properties of aqueous H2O
solutions of the ferric ion in the presence of chloride ions, since the dif-
ferent complexes which are formed are derived from’ this basic distorted Fe
structure.
(c) Cis-trans Isomerism The differences in ligand should induce dif- H2O
Cl
ferent values of the electric field gradient at the central ion. In the trans
H2O
case there is an axial symmetry (D4h), which is absent in the cis ca se
(C2v). Based on the point charge model, the field gradient is given by
Fig. 10.50. Fig. 10.49 Model of the [FeCI2(H2O)4 ]+ ion.
(I)
Z Y
q1
r
X
(II)
+
q1 Y
Z
q2 X
q2 r
Fig. 10.50 The EFG in a point charge model for cis and trans isomers.
(I) Transcase (D4h): q = 0
[q ]trans ( γ ) 4q / r 3 (10.138)
(II) Cis case (C2v): q = p/2
[q ]cis = ( − γ ∞ )2q / r 3 (10.139)
In general,
⎡ ⎤
q ( γ ) ⎢ ∑ qi ( cos 2 θi − 1) / ri3 ⎥ (10.140)
⎣ i ⎦
where terms and symbols have their usual meanings. qi is the magnitude of the ith charge whose coordinates are
ri and qi.
π
Since q = 0 for trans case and q = for the cis case, so the ratio of quadrupole splittings is
2
EQ (tran
r s)
=2:1 (10.141)
EQ (cis )
This procedure can be used to identify which is the cis and which is the trans isomer, e.g. ΔEQ/dIS:
(cis-Fe(CNMe)4(CN)2 = 0.24/0.18 and trans-Fe(CNMe)4(CN)2 = 0.44/0.18 mm/s. ΔEQ/dIS: cis-Fe(CN Et)4(CN)2 =
0.29/0.23 mm/s. d values are with respected to SNP.
(d) Ligand Linkage Isomerism The x-ray structure of metal ferrocyanides (the Prussian blue and similar
compounds, e.g. K2NiFe(CN)6 shows that one metal is bound to the nitrogen and the other to the carbon atom, i.e.
Fe—C ≡ N—Ni. The carbon end of the cyanide linkage creates a strong field and tends to form low-spin complexes,
whereas the nitrogen end is a weak field ligand and usually gives high-spin complexes. The phenomenon of cya-
nide linkage isomerism has been studied in iron (II) hexacyanochromate. This complex with composition Fe3II
(CrIII(CN)6) exhibits linkage isomerism. At room temperature it changes spontaneously from form (I) to form (II)
as shown in Fig. 10.51. Because of the difference in strength of ligand field induced by carbon and nitrogen ends
of the cyanide ligand, the high-spin S = 2, Fe2+ is expected in form (I) and a low-spin S = 0 Fe(II) complex in
form (II). The MB spectra of form (I) show indeed the typical isomer shift and quadrupole splitting of the high-
spin Fe2+ whereas form (II) exhibits low values of isomer
shift and quadrupole splitting corresponding to low-spin Fe2+ N C Cr3+ Cr3+ N C Fe2+
ferrous ion configurations. The MB observations find
(I) (II)
support from x-ray powder diffraction, infrared and mag-
Fig. 10.51 Ligand linkage isomerism in iron hexacynochromate.
netic susceptibility studies on this complex.
(e) Structure of Complex Coordinated Molecules Structure of complicated complex molecules can be
accomplished from MB parameters, namely intensity, isomer shift, quadrupole splitting, internal magnetic field,
etc., since they are sensitive to the electron configuration of the molecules. Let us first examine whether qua-
drupole splitting alone is competent enough to rule out the possible structures of a particular compound or not.
This is demonstrated here by taking the example of Fe3(CO)l2. The MB spectrum of Fe3(CO)12: Cr (57CO) at 78 K
consists of three lines. Three structures shown in Fig.

10.52 are possible corresponding to formula Fe3(CO)12.
In one of the structures (a) the three iron atoms may
be equivalent. In such a situation, the MB spectrum
should exhibit an unexpected two-line spectrum. Thus
such a structure is ruled out. In the other two structures
(b) and (c), two of the three atoms may be equivalent.
In one of these structures (b), if the two atoms are ir-
regularly octahedrally arranged and the third one is not,
(a)
the spectrum would consist of four lines due to quadru-
pole splittings which is contrary to the expected spec-
trum. On the other hand, if the two equivalent atoms
are irregularly octahedrally arranged and the third one
is nearly octahedral as shown in Fig. 10.52(c), the spec-
trum would consist of three lines as expected, the cen-
tral line is because of nearly octahedral Fe atom and the
other two lines are due to quadrupole splitting at two
equivalent Fe atoms.
The other structural problem which demonstrates
the potential of MB tool is that of Prussian and
Turnbull’s blues. Prussian blue is prepared by mixing
the solutions of an iron (III) salt (in excess for insoluble (b)
(c)
Prussian blue) (Fe2(SO4)3) and potassium ferrocyanide
(K4Fe(CN)6) and Turnbull’s blue by mixing the solu- Fig. 10.52 Three possible structures of Fe3
(CO)12
,● −Fe, O − CO.
tions of ferrous salt (FeCl2) and potassium ferricyanide
(K3Fe(CN)6). Now the problem arises as to (i) what are
the electronic configurations of iron in these three compounds: ferric ferrocyanide FeIII[FeII(CN)6], ferrous fer-
ricyanide FeII[FeIII(CN)6] or a more complex configuration, and (ii) whether soluble Prussian blue, insoluble Prus-
sian blue and Turnbull’s blue have the same formula or not.
However, before discussing these aspects, it will be more appropriate to report the MB spectra of potassium
ferrocyanide and potassium ferricyanide. Ferrocyanides exhibit a single-line spectra because of the octahedral
ligand symmetry and the lack of a nonbonding 3d electron contribution to EFG, e.g. K4Fe(CN)6 at 143 K: dIS(Fe) =
+ 0.04; K4Fe(CN)6. 3H2O at 298 K: dIS.(Fe) = −0.047.
Further, a decrease of chemical shift and broadening of line occurs under the influence of applied pressure.
This is attributed to an increase in s-electron density at the nucleus and a possible distortion of octahedron. High
pressures at above 100°C on M2Fe(CN)6 where M = Cu, Ni and Zn, cause a partial change into high spin (S = 2)
Fe2+ species. The process is reversible but with hysteresis.
A small quadrupole splitting of 0.28 mm/s has been observed in H4Fe(CN)6. This is attributed to H-bond struc-
ture which has a symmetry lower than octahedral.
On the other hand, the MB spectra of K3Fe(CN)6 in the temperature range 4.2−300 K, consist of a doublet:
ΔEQ = 0.28 mm/s, and dIS (Fe) = −0.124 mm/s. However, below Neel’s temperature at 0.025−0.04 K, a six-line
pattern has been observed with Hint =193 kG. Moreover, the increase of quadrupole splitting, together with the
appearance of a third peak in the spectrum which has isomer shift value similar to that of ferrocyanide has been
observed with the increase of pressure. This may be either due to an increase in the degree of back donation from
3d orbitals to the empty ligand p*- antibonding orbitals or a change in the 4s-admixture in the binding or both.
Near 50 kilobar (kb), there is a phase transition which causes discontinuity in the isomer shift-pressure curves for
both the ferrocyanide and ferricyanide components and at the same time produces a partial reversion to the oxida-
tion species. All these changes are fully reversible on the release of pressure. M2Fe(CN)6 also shows a reduction
to ferrocyanides with increasing pressure. At elevated temperatures, a second change to high spin (S = 2) Fe(II)
is also observed.
Let us now describe the two methods which are employed to identify and confirm the oxidation states of iron
in the three blue complexes. The natural abundance of 57Fe is only 2.17 per cent. The first method is based on the
MB studies on samples prepared from reactants containing unenriched iron whereas the second is based on the
samples prepared from reactants containing enriched iron.
(i). All the three complexes become ferromagnetic below 5.5 K and the MB spectra at 1.6 K consist of a well-
resolved six-line pattern typical of ferric iron and a superimposed singlet owing to diamagnetic ferrocyanides
Velocity of source (mm/s)
−8 −6 −4 −2 0 2 4 6 8 10
10
(a)
−8 −6 −4 −2 0 2 4 6 8 10
Absorption percent
10
(b)
−8 −6 −4 −2 0 2 4 6 8 10
0
10
(c)
S1 S2
(d)
Fig. 10.53 The Mössbauer spectra at 1.6 K for (a) soluble Prussian blue, (b) insoluble Prussian blue, and (c) Tumbull’s blue. The line posi-
tions for the two kinds of iron are indicated in (d) with solid lines and dashed line. The ferric six lines spectrum shows a quadrupole splitting
1
S1 − S2 = c2qQ (3 cos2q − 1).
2
Table 10.14 Mössbauer Parameters for Iron Ferro/Ferricyanides.
Parameter Prussian blue Turnbull’s blue
Soluble Insoluble
Fe3+ Fe(II) Fe3+ Fe(II) Fe3+ Fe(II)

dIS (SNP)/ 0.84/0.57 0.33/0.06 0.84/0.57 0.31/0.04 0.83/0.56 0.27/0.0
dIS (Fe) mm/s
S1 − S2 (mm/s) 0.37 0 0.48 0 0.52 0
Hint(kOe) 536 ± 20 0 ± 10 541 ± 20 0 ± 10 543 ± 20 0 ± 10
S1 −S2 =
1 e2qQ(3cos2q − 1)
2
(because of octahedral ligand symmetry and the lack of non bonding 3d electron contribution to EFG) (Fig. 10.53).
The results obtained are listed in Tables 10.14 and 10.15 and are compared with characteristic values of MB
parameters for the various states of iron recorded in Table 10.16.
Table 10.15 Intensity ratio between two kinds of Iron, Fe3+ and Fe(II), for Prussian Blue and
Turnbull’s Blue.
Fe3+/Fe(II) (observed)a Fe3+/Fe(II) (normalised)b
Soluble Prussian blue 1.39 1.00
Insoluble Prussian blue 1.78 1.28
Turnbull’s blue 1.84 1.32
(a) Fe3+/Fe (II) observed with the absorbers containing iron of about 20 mg/cm2 (b) Normalised to the
thin absorber.
Table 10.16 Characteristic Mössbauer parameters for various charge states of iron.
Fe3+ (ionic) Fe2+ (ionic) Fe(III) (CN) Fe(II) (CN)
dIS(SNP) (mm/s) ~ 0.7 ~ 1.6 ~0 ~0
ΔEQ ~ 0.5 1.0 − 3.4 ≤1.0 ~0
Hint (kOe) 500 ~ 600 0 ~ 300 170 ~ 270* 0
* The value of internal magnetic field of low-spin Fe(III) combined with (CN)6 are for K3Fe(CN)6 and
M3[Fe(CN)6]2 (M : Mn, Co, Ni, Cu).
The internal magnetic fields of 540 kOe and zero kOe are typical of high-spin S = 5/2, ferric iron and low-spin
S = 0 ferrous iron respectively. The isomer shift and quadrupole splitting for each iron moities also conform to
typical value for Fe3+ and Fe(II) respectively. Thus these compounds exist in the Fe3+ ionic and covalent Fe(II) elec-
tronic states. It is to be noted that blue colour is still retained at 1.6 K, which confirms the charge transfer model
and discards the valence oscillation or resonance between structures (i.e. Fe(II) (CN)6Fe3+ Fe(III) (CN)6 Fe2+)
at 1.6 K because the charged states are stable for at least for 10 s or at least occurs much more rapidly that 10−8 s;
−7
time of Larmor’s precesion of the 57Fe nucleus. The preparation of Turnbull’s blue, therefore, involves a mutual
redox reaction.
Almost the same value of quadrupole splittings in
the three compounds suggests that the electronic struc-
100
ture of these compounds is the same from the MB
spectral point of view. Therefore, in Turnbull’s blue,
the charge transfer from Fe2+ (high spin) to Fe(III) (low
Transmission
spin) or flipping of the CN ligand by 180° should occur

at the instant of combination.
The intensity ratios of the MB spectra are consis- 95
tent with the stoichiometric formula KFe(Fe(CN)6) for

soluble Prussian blue (the electronic spectra of solu-
ble Prussian blue could be interpreted in terms of a −3 −2 −1 0 1 2 3
Velocity (mm/s)
KFe3+[Fe(II)(CN)6] formulation with a charge transfer
(a)
excitation to the unstable KFe2+[Fe(III)(CN)6] alterna-
tive) and Fe4 (Fe(CN)6) for insoluble Prussian blue and 100
Turnbull’s blue. Thus, from the comparison of the MB
spectra, the basic information on the structure of these
Transmission
90
coordinated complexes has been obtained.
(ii) The MB spectrum of Prussian blue at 77 K pre- 80
pared from enriched 57Fe, Fe2(SO4)3 and unenriched
K4Fe(CN)6 exhibits a dominant feature of iron (III) 70
quadrupole doublet [Fig. 10.54(a)]. There is little or
−3 −2 −1 0 1 2 3
no transfer of enriched iron to the low-spin moites. Velocity (mm/s)
The equivalent Turnbull’s blue preparation 57Fe Cl2 (b)
and unenriched ferricyanide also gives the enhanced Fig. 10.54 Mössbauer spectrum at 77 K for Prussian blue pre-
iron (III) spectrum [Fig. 10.54(b)], thereby confirming pared by mixing: (a) 57Fe2(SO4)3 and unenriched K4 [Fe (CN)6 ], and
the redox reaction. In both spectra the weaker singlet (b) 57FeCI2 and unenriched K3 [Fe (CN)6 ].
component arising from the unenriched ferrocyanide is obscured within the unresolved central portion of the
spectra.
Application of pressure to insoluble Prussian blue causes considerable reversible reduction of high-spin ferric
iron to high spin ferrous iron.
10.15.10 Spin-State Equilibria

The MB spectra of iron (II)-bis-(1, 10-phenanthroline) at 293 and 77 K exhibit quadrupole doublet respectively.
The MB parameters: ΔEQ = 2.67 mm/s, dIS (SNP)/(Fe) = 0.34/0.98 mm/s at 293 K correspond to high-spin
S = 2, Fe2+(5T2) configuration whereas ΔEQ = 0.34 mm/s, dIS (SNP)/Fe = 0.62/0.37 mm/s at 77 K correspond
to low-spin, S = 0 Fe2+ (1A1) configuration. However, the more extensive measurements over this temperature
range show lines due to both the spin states in the vicinity of transition region (133−173 K). This shows that
the interconversion takes place on a time scale, greater than 10−7 s, a factor also common to other 5T2 − 1A1
transitions. Thus the variation of hyperfine parameters as a function of temperature from the range of values
of high-spin to low-spin configuration of the ferrous ion reveals the coexistence at a given temperature of
different configurations of ferrous ion in the complex as expected on the basis of magnetic susceptibility mea-
surements on transition metal ions. Thus, this data provides evidence for the existence of spin-state equilibria
between equienergetic 5T2 and 1A1 states and is the first reported quintet-singlet equilibrium in a transition
metal complex.
Fe(phen)2 (NCSe) also exhibit spin-state equilibria. The MB parameters for this complex are: ΔEQ = 2.52 mm/s,
dIS (Fe) = 1.03 mm/s at 293 K for 5T2 configuration and ΔEQ = 0.18 mm/s, dIS (Fe) = 0.35 mm/s at 77K for 1A1
configuration.
10.15.11 Structure of Organometallic Compounds

The MB spectroscopy is used to study the structure, bonding, conformational changes with the change in envi-
ronment; and symmetry in organometallic compounds. The solvents play a key role in such studies. They should
have good solvating power and should be inert and glass formers at low temperatures (so that the solvent accom-
modates the structure of the solute rather than imposing a crystalline environment on the latter), they should have
a high lattice temperature (so that the resonance effects are large at the temperature of the measurement, usually
liquid nitrogen temperature). For the study of cyclooctatetraene iron tricarbonyl [COTFe(CO)3] problem, these
criteria are met by methyltetrahydrofuran (MTF) and ethanol + 1 propanol + diethyl ether (EPA 16 : 42 : 42).
The correlation diagram (dIS vs ΔEQ) for the several iron tricarbonyl compounds [RFe(CO)3] as neat solids and
as frozen solution samples is shown in Fig. 10.55. Figure 10.55 reveals that the quadrupole splitting parameter
0.36 C8H8Fe(CO)3 C8H18

C8H8Fe(CO)3 C6H5NO2
C8H8Fe(CO)3 MTF
C8H8Fe(CO)3 EPA
C8H6(CO2CH2) Fe(CO)3 MTF C8H6(OCOCH3)2Fe(CO)3
C8H6(CO2CH2)Fe(CO)3
0.32
Isomer shift (mm/s)
C8H8Fe(CO)3
C7H8Fe(CO)3
0.28 (CH2)3C Fe(CO)3 EPA C4H4Fe(CO)3
(CH2)3C Fe(CO)3
C4H4Fe(CO)3 EPA
0.24
C8H8Fe2(CO)6
C8H12Fe(CO)3
0.60 1.00 1.40 1.80 2.20

ΔEQ (mm/s)
Fig. 10.55 Correlation diagram between isomer shift and quadurpole splitting for several iron tricarbonyl compounds as neat solids and as
frozen solution samples. Metal in these complexes is bonded to a C4H4 fragment. Data is relative to SNP at 294 K.
(which is most sensitive to variations in the structures that are being considered) of COT Fe(CO)3 is the same
in the solid and in EPA, nitrobenzene, MTF (methyl tetra hydrofuran), n-octane and identical to the quadrupole
splitting in the binuclear complex COT[Fe(CO)3]2. Moreover, this parameter i.e. ΔEQ in the COT complexes (i) is
markedly different from the value observed for C8H12Fe(CO)3 a compound which from other evidence is found to
be a 1, 5-diene complex of iron tricarbonyl, and (ii) is much smaller compared to ΔEQ value observed for norbor-
nadienyl iron tricarbonyl [C7H8Fe(CO)3] in which pseudooctahedral structure is not realised.
The MB data on COT Fe(CO)3 show that the configuration of this compound in solid and solution phases is the
same as that observed by x-ray techniques for the solid. NMR and IR measurements also confirm this observation.
PMR spectrum of this compound in solution phase at room temperature consists of a single peak. Accordingly,
all the eight protons in the complex should be equivalent. This is possible provided the complex is an ‘open-faced
sandwich’ with cyclooctatetraene ring approximately planar and symmetrically bonded to the Fe(CO)3 moiety.
On the other hand, the x-ray diffraction data show that in the solid state, the COT Fe(CO)3 molecule has
a pseudo-octahedral structure, i.e. the Fe(CO)3 moiety is asymmetrically situated above the C4H4 fragment of
the ring and the remaining hydrocarbon framework is separated by distances which preclude a direct bonding
interaction. The ambiguity in the room temperature NMR and x-ray diffraction results has been resolved from low-
temperature NMR spectrum which exhibits a splitting of proton resonance into two structured peaks symmetrically
displaced from the resonance peak at room temperature. Note that at low temperature, the frequency of rotation of
eight protons around the bond becomes large compared to NMR time-scan, i.e. 10−6 s. This results in eight equiva-
lent protons. The low-temperature NMR spectra may be interpreted in terms of bonding of the Fe(CO)3 fragment to
(a) a biplanar COT ring acting as a 1, 3-diene, (b) a tub shaped COT ring acting as a 1,3-diene and (c) a tub-shaped
COT ring acting as a 1, 5-diene. The unwanted structure of the low-temperature solution conformation has been
ruled out from MB measurements on neat solids and on frozen solution samples of the COT Fe(CO)3 complex and
related molecules. Quadrupole splitting parameter of COT Fe(CO)3 is same in both the phases (1.16–1.24 mm/s)
and is also identical to the quadrupole splitting of binuclear complex, COT [Fe(CO)3]2. This parameter of COT
complex is different from the value observed for C8H12 Fe(CO)3 (1.8 mm/s) which is known from other evidence
to be a 1, 5,-diene complex of iron tricarbonyl and is also much smaller compared to that of norbornadienyl iron
tricarbonyl [C7H8Fe(CO)3] (ΔEQ = 2.15 mm/s in which pseudo octahedral structure is not realised.
Further, the possible, conformational changes in other, organametallic compounds when these are subjected to
changes in environment can also be followed by MB spectroscopic tool, e.g. the 119Sn (relative to SnO2, dIS = 0)
MB resonance (Sn) spectrum of neat [(p − C5H5) Fe(CO)2]2 SnCl2 consists of a sharp doublet (dIS = 1.95 mm/s and
ΔEQ = 2.38 mm/s). However, the spectrum (Sn) in frozen solution in poly (methylmethacrylate) matrix becomes
broad and can be resolved into two quadrupole doublets of esentially identical quadrupole splitting but differ-
ent isomer shifts, (i.e. ΔEQ = 2.25/2.29 mm/s and dIS = 1.96/1.31 mm/s), Fe(SNP): Neat/2-methylfuran: ΔEQ =
1.68/1.68 mm/s and dIS = 0.36/0.36 mm/s. The two doublets suggest the presence of two different tin (Sn) atoms
which are associated with two rotational conformers of this molecule in solution. These results provide confirma-
tion of the speculation based on infrared spectral data.
{Me2 Sn[Fe(CO)4]}2 Sn also exhibits rotational isomerism as is evident from the MB data indicated below in
the tabular form.
57
Fe ll9
Sn(SnO2 )
ΔEQ dIS ΔEQ dIS

0.30 −0.10 1.24 1.45
− 2.20
80 K and values for dIS (Fe) converted from SNP. Parameters are in mm/s.
10.15.12 Electronic Structure and Bond Nature

of Charge Transfer Complexes
The change in the MB parameters namely, quadrupole coupling constant and chemical shift of MB nucleus pres-
ent either in the donor or acceptor in the donor–acceptor systems may be fixed a criterion for the charge-transfer
complexation between the donor and the acceptor. From the number of quadrupole splitting patterns in the MB
spectrum, one can infer about the chemical species of MB nuclei present in the molecular complex. Further, it
is also possible to estimate the charge densities on the chemical species of MB nuclei and the bond nature in the
charge-transfer complex provided a relationship exists between the MB parameters and the bond parameters.
For instance, the MB spectrum of 129I (at 16 K in CS2, relative to ZnTe source) of pyridine–iodine complex
exhibits two quadrupole splitting patterns say A and B; the pattern A is more intense than pattern B and the MB
parameters corresponding to these patterns in the pyridine−I2 system are (A/B): e2qQ (MHz) = −2631/1439,
d (mm/s) = 1.82/0.39. For I2 (in CS2): e2Qq (MHz) = −2242; d(mm/s) = 0.93, h (per cent) = 11.(The quadrupole
coupling constant (e2Qq) is converted to I127). The MB data indicate that there are two chemical species (say IA
and IB) containing two electronic charge states of the MB nucleus in the complex (in the inert solvent). This fact
may suggest that the iodine molecule lies along the symmetry axis of the donor molecule resulting in a linear N−
IA − IB arrangement. The large absorption intensity of A as compared to that of B lends support to the assignment
of IA to the bridged iodine atom and IB to the terminal iodine atom. In order to determine the charge densities on
the atoms involved in charge-transfer complexation and the nature of chemical bond formed, use is made of the
following relations between MB parameters and bond parameters.
The number of the unbalanced p-electrons is given by
−e 2Qqobs
Up = = 1− s2 ± i − π (10.142)
e 2Qqatom
where s2 is the amount of s hybridisation of the s bond, i is the ionic charge, and p is the double-bond character
and/or the intermolecular bond. The atomic iodine quadrupole coupling constant for 127I e2Qqatom = 2293 MHz.
Thus, from the observed values of e2 q Q and Eq. (10.142), we get
−2631 (MHz)
U p (I A ) = = 1.147
−2293 ( )
−1439 (MHz)
U p (I B ) = = 0.627
−2293 ( )
Assuming that s2 and p are negligible with respect to the quantity i, we obtain
1 ± iA = 1.147
and 1 ± iB = 0.627
Thus,
iA = + 0.147,
iB = − 0.373,
and iN = + 0.226
N IA IB
+0.226 +0.147 −0.373
These values are reasonable compared to the values obtained from the far infrared spectroscopy and the mea-
surement of the dipole moment.
The correctness of the assumption that the chemical bond of the bridging iodine atom is formed by only pure 5
ps orbital can be checked from the correlation between dIS value and the charge density on the iodine atoms.
dIS (mm/s) = −9.2 hs+ 1.5 hp − 0.54 (ZnTe source) (10.143)
where, hs and hp are the number of s- and p-electrons removed from the 5s25p6 (I−) configuration respectively.
Assuming that there is no s hybridisation in the I−I bond, we obtain from observed value of dIS and Eq. (10.143),
the values of charge densities on the terminal and the bridged iodine atoms as follows:
0.39 = 1.5 hp − 0.54
0.39 + 0.54 0 93
or hp = = = 0.620
15 15
Thus, iB = 0.620 −1 = −0.38

Similarly, we get, hp = 1.573 and iA = 0.573.
The agreement between the values of iB obtained from quadrupole coupling constant and dIS values respec-
tively shows that the I−I bond in the pyridine-iodine complex consists of mainly a pure s bond of the p-electrons.
On the other hand, the unexpected large value of hp(IA) compared to Up(IA) suggests that the intermolecular charge-
transfer bond cannot be described by only a pure s bond of the 5p electrons. The s hybridisation of the s bond
does not contribute to the CT bond since it greatly increases the hp value and slightly decreases Up. The discrep-
ancy of hp(IA) and Up(IA) may be due to the presence of p bond which arises due to the donation of 5pp electrons
by iodine to the donor molecule. Based on this assumption the p-bond character can readily be obtained from the
following expressions:
hp = 6 − (Nx + Ny + Nz) (10.144)
1
Up = (N + Ny) − Nz (10.145)
2 x
where, Nx, Ny and Nz are the number of 5p electrons along the X, Y, and Z-axes, respectively.
Substituting the values of hp(IA) and Up(IA) into the respective expressions we obtain
1
1.15 = (Nx + Ny) − Nz (10.146)
2
1.57 = 6 −(Nx + Ny + Nz) (10.147)
From the solution of the equations represented by (10.146) and (10.147), we get
Nz = 0.71, Nx + Ny = 3.72.
The p-bond character comes out to be about 14 per cent which of course is unreasonably high. In other words
we can say that the bridging iodine atom (IA) has abnormally high positive charge, i.e. + 0.573.
The unreasonably high value of charge density on the bridged iodine atom may be due to the non consideration
of 4d electrons contribution partially to the formation of the CT bond between the nitrogen atom in the donor
molecule and the bridging iodine atom. The calculations on the basis of ds ps hybridisation in which the 4ds
electrons promote to the 5ps orbital yield IA = +0.11, IB = −0.38 and 4d electron defect on the iodine atom = 0.33.
Further, the low value of h (10 per cent) for the bridged iodine atom rules out the contribution due to back dona-
tion of 4dp electrons in this complex although this donation causes an increase in the isomer shift greatly and the
quadrupole coupling constant slightly.
It is interesting to note that unlike pyridine −I2 complex the observed spectra of benzene, mesitylene and hex-
amethylbenzene, −I2 complexes exhibit only one type of quadrupole splitting pattern suggesting thereby that only
one chemical species of iodine is present in these complexes. The correlations between MB parameters and bond
parameters also exist in these type of complexes. They, being complicated, are not treated here.
10.15.13 Studies of Lunar Soil

An interesting problem is the oxidation state of iron in the constituents of the soil, namely earthly soil, lunar
soil, Martian soil etc. The primary aim has been to study the amount of iron in the soil, distribution of iron over
the soil constituents and the particular valence states. Such type of studies on earth is important because plants
take Fe2+ form of iron but not Fe3+ from the soil. These days due to soil pollution, soils may become F2+ deficient,
consequently the growth of plant kingdom may be retarded. As an example, sandy soil being Fe2+ deficient is not
recommended for the cultivation of paddy. Thus for survival of mankind in the future, it is important to study the
composition and characteristics of of soil from other planets also.
The constituents of the lunar soil are of complex nature. Since the first landing on the moon in July 1969,
a large number of physical techniques, viz., x-ray diffraction, electron microscopy, and x-ray emission micro-
analysis and chemical methods have been employed to study lunar samples returned in the course of the Apollo
and Luna missions. The main constituents include glassy agglomerates with a wide range of colour and trans-
parency, and with angular, ellipsoidal, or spherical shapes, or formed like dumb-bells. Some glass particles
show devitrification or inclusions of partially molten crystals. Additional components of the soil are isolated
mineral particles of plagioclase (Na, Ca) [(Al, Si)4O8], pyroxene [(Ca, Mg, Fe)SiO3], the minerals of this group
are augite [~Ca0.15−0.4(Mg, Fe)0.85−0.6 SiO3], pigeonite [~ Ca0.5−0.15(Mg, Fe)0.95−0.85SiO3], and orthopyroxene [(Mg,
Fe)SiO3]; ilmenite (FeTiO3), and olivine [(Mg, Fe)2 SiO4]. The soil also contains small amounts of alloys with
compositions close to metallic iron. The principal elements which form more than 99 per cent of the lunar
surface are, in order of decreasing abundance oxygen, silicon, aluminium, iron, calcium, titanium, sodium and
potassium. The aluminium and calcium show higher concentrations in the terra regions or highlands whereas
iron and titanium in the mare regions. In view of the chemical composition of the lunar surface, applications
of the MB effect have naturally been restricted to iron which is one of the four most abundant elements. The
MB effect has made a modest, more or less descriptive contribution to the study of composite samples and
separated mineral phases from the moon. MB spectroscopy of 57Fe has been used mainly as a complement to
various other techniques to characterise the soil fraction with a grain size smaller than about one mm. Even
analytical data regarding the amount and properties of iron in grain sizes less than 1 mm, not easily obtainable
by other techniques such as x-ray emission-microprobe analysis can be obtained by MB technique. Here, we
shall be concentrating on the study of lunar soil by this technique since earthly soils have already been well
characterised and documented.
The MB absorption spectra of the bulk fines represent remarkable ‘fingerprints’ of the landing sites, e.g. the
resonant absorption spectrum of 57Fe (295K) in lunar soil from Apollo landing site at Mare Tranquillitatis is
shown in Fig. 10.56. The spectrum consists primarily of the superimposed resonant absorption patterns of 57Fe
in the silicate minerals, silicate glasses, ilmenite (FeTiO3), metallic iron and troilite (stoichiometric FeS). The
absorption in the range between 0 and 0.3 mm/s arises primarily from iron in silicate-glass, pyroxene, and olivine
whereas the peaks at lower and higher velocities are attributed to metallic iron. The MB spectra of 57Fe (at 295
and 77 K) in glass separated from Apollo 11 soil with 1 mm grain size shown in Figs 10.57 (a) and (b) did not
exhibit any signal corresponding to Fe3+ ions with a detectability limit of ±1 per cent. The doublet due to ilmenite
(FeTiO3) results from inclusions whereas the doublets 1 and 2 are because of Fe2+ ions in two distinct coordination
polyhedra sites similar to that in pyroxene suggesting a high degree of devitrification. The temperature depen-
dence of the doublets 1 and 2 is also similar to that of pyroxene.
MB hyperfine spectrum of 57Fe in plagioclase separated from grained basalt of the Apollo 11 landing site at
Mare Tranquillitatis consists of an apparent three-line pattern as shown in Fig. 10.58. The three-line pattern has
been ascribed to two overlapping quadrupole split doublets of Fe2+ ions at tetrahedrally coordinated Al3+ sites and
at the Ca2+ site. Though plagioclase did not show any signal corresponding to Fe3+ ions, the presence of these ions
in lunar plagioclase crystals has been confirmed by ESR. However, the MB spectrum of 57Fe (295K) in augite
from a terrestrial volcanic rock, exhibits in addition to an intense doublet because of Fe2+ ions, a weak peak which
is assigned to Fe3+ ions. (Fig. 10.59). The total iron content in the bulk fines of Apollo-11 soil as estimated from
the total resonant absorption is about 10–12 mass per cent. MB parameters in glasses from lunar soils have been
compared with synthetic glasses in Table 10.17.
0.00
0.02
0.04
Resonant absorption
0.06
0.08
0.10
0.12
−8.0 −4.0 0.0 4.0 8.0

Velocity (mm/s)
Fig. 10.56 Mössbauer spectrum of Fe57 at 295 K in lunar soil from the Apollo 11 landing site at Mare Tranquillitatis.
0.00
0.01
Fe
0.02
Resonant absorbtion
0.03
0.04
0.05
0.06 FeTiO3
1
2
0.07 Fe
−3 −2 −1 −0 1 2 3
Velocity (mm/s)
(a)
0.00
0.01
0.02 Fe
Resonant absorbtion
0.03
0.04
0.05
0.06
FeTiO3
0.07
1
2
0.08
−3 −2 −1 −0 1 2 3
Velocity (mm/s)
(b)
Fig. 10.57 Mössbauer spectra of Fe57 (a) at 295 K in glass particles separated from Apollo 11 soil and (b) at 77 K of the glass particles.
0.000
0.001
0.002
Resonant absorption
0.003
0.004
0.005
0.006
0.007
0.008
1
0.009 2
0.010
−4 −3 −2 −1 0 1 2 3 4
Velocity (mm/s)
Fig. 10.58 Mössbauer spectrum of 57Fe in plagioclase isolated from a coarse grained basalt of the Apollo 11 landing site at Mare Tranquil-
litatis. 2 : doublet due to Fe2+ at the Ca2+ sites; 1: doublet due to Fe2+ ions at tetrahedrally coordinated Al3+ sites.
0.00
0.01
Fe3+
Resonant absorption
0.02
0.03
0.04
0.05
−3 −2 −1 0 1 2 3
Fig. 10.59 Mössbauer spectrum of Fe (295 K) in augite from a terrestrial volcanic rock of the Apollo 11 landing site at Mare Tranquillitatis.
57
The weak Fe3+ peak is a part of doublet which includes approximately ten per cent of the total resonant absorption.
Table 10.17 Isomer Shifts, Quadrupole Splittings and Line Widths in Glasses from Lunar Soils and Synthetic Glasses at Room Temperature.
Sample State of iron Average dI.S.(a) AverageΔEQ Line-width Lower Linewidth Upper
(mm/s) (mm/s) position (mm/s) position (mm/s)
Apollo 11 doublet 1(b) 2+ 1.16 2.72 0.27 0.37
doublet 2(b) 2+ 1.17 1.87 0.49 0.64
Apollo 17
Orange glass(c) 2+ 1.05 1.99 0.64 0.78
Synthetic glass 2+ 1.04 1.97 0.57 0.72

prepared at 1335 °C
Synthetic glass 3+ 0.32 1.15 0.66 0.58

prepared at 1500 °C
1
(a) With respect to metallic iron absorber, (b) area ratio = 0.15. Additional peaks due to ilmenite inclusions are not tabulated. (c) Apollo 17 orange
(1 + 2)
glass is separated from orange soil collected at the Apollo 17 landing site in the Taurus Mountains−Littrow region, on the rim of Shorty Crater. Apollo 17
landed on 11 December, 1972.

The most celebrated MB nucleus 57Fe, is present in many biological molecules, so most of the interesting and stim-
ulating applications of MB spectroscopy are in the field of biosciences. Earlier the state of transition metal ions in
biological molecules has been accomplished by magnetic susceptibility and ESR techniques, both of which deal
with the bulk properties of the system. One of the major advantages of MB tool over these techniques is that it is
site specific. Moreover, proteins containing diamagnetic iron which could not be studied by ESR technique can be
measured by MB method. The measurements are made on frozen aqeuous solutions of protein which have been
precipitated and separated in solid form using ultra centrifuge. Care is also taken that in the separation process,
the denaturation of protein does not take place. However, the possibility of a conformational change in the solid
cannot be ruled out, and consequently the active solution form is not observed. On the basis of observations, it is
found that frozen solutions of protein concentrates give satisfactory results. Further, since natural iron contains
only 2 per cent of MB isotope, comparatively a weak MB resonance is observed in biological molecules. So in
order to obtain a strong MB resonance line in biological molecules, the natural samples need be enriched in 57Fe.
This is achieved either by growing organisms on a diet enriched in 57Fe or by incorporating 57Fe in the samples
by chemical exchange, i.e. in some proteins it is possible to remove iron and then to reconstitute the protein with
enriched 57Fe without affecting the biological activity. The growing method requires more 57Fe and is more reliable
than exchange method since in the latter, the proton may be modified in the exchange process.
MB resonance studies in biological molecules are carried out to have information on (i) the electronic structure
of iron in these molecules, (ii) location and site symmetry of iron in them, and (iii) the role of iron in biochemical
transformations in which these molecules participate.
57
Fe is present in biological systems such as (i) haemoproteins, i.e. haemoglobin, myoglobin, cytochromes, per-
oxidases, catalases. The latter three are also known as haeme enzymes, (ii) iron–sulphur proteins, i.e. rubredoxins
(1-iron/molecule), ferredoxins (a) plant type (2-iron/molecule) (b) bacterial (8-iron/molecule), high potential
iron proteins (HPIP) (4-iron/molecule), conjugated iron–sulphur proteins (8 or more iron/molecule), i.e. xanthine
oxidase; iron–molybdenum proteins, iron–zinc proteins, superoxide dismutase contain Cu−Zn and Fe−Zn pro-
teins which control the aging process in living systems by reducing the accumulation of superoxide-free radicals
(Ö2−) formed in many redox reactions, and (iii) iron storage and transport proteins such as ferritin, transferrins
and hemosiderin. The latter two types are nonhaeme proteins. The haeme proteins have many different kinds
of functions: oxygen transport (haemoglobin), electron transport (cytochrome), catalysis (peroxidase), etc. The
iron–sulphur proteins which are found in many diverse organ-
isms, in plants, animals and bacteria are involed in oxidative X
electron transfer in photosynthesis, nitrogen fixation, diges-
tion, respiration, vision, etc. It is to be noted that the function
of haeme proteins is governed by the haeme group present
in such proteins. Haeme is a small planar stable molecule in N N
which iron is coordinated to four nitrogen atoms which belong Fe
N
to the porphyrin ring and the remaining two sites of iron may N
be occupied by two ligands above and below the plane of the
four nitrogen ligands. The structure of haeme is shown in
Fig.10.60. Haemoglobin contains four haeme groups in each HIS
of which the fifth coordination site of iron is attached to globin
through a nitrogen atom in an imadizole of histidine residue PRO
below the plane of four nitrogen ligands. The sixth coordina-
tion site, i.e. X of iron is occupied by an exchangeable ligand GLU
(L) such as CO, NO, O2, etc., and the state of iron varies with THR
the nature of the labile ligand. Myoglobin differs from hae- LEU
moglobin in that it contains only one haeme group. Haemo-
globin is the biological molecule, the structure of which is Fig. 10.60 Structure of haeme group.
established by x-ray diffraction tool.
Iodine, though controls all the metabolic processes in the body, but iodine MB effect has no significant appli-
cations in biological sciences because (i) although the natural linewidth of 129I is 0.59 mm/s, its natural abun-
dance is zero and one has to use radioactive 129I obtained as a fission by-product and replace natural 129I-partially
or wholly. (ii) 127I has 100 per cent abundance, but the MB spectrum for 127I is hard to observe because of high
energy of g -ray it emits (57 keV) and the accompanied x-rays. The effect is observed at liquid nitrogen tempera-
ture. Moreover, due to large natural linewidth (2.54 mm/s), the small quadrupole interactions cannot be resolved
and detected.
Let us now examine how MB spectra of haemoglobin and its complexes are valuable in understanding the role
of iron in biological processes.
10.16.1 Studies of Haemoglobin and its Derivations

The 57Fe Mössbauer spectral studies of haemoglobin and its derivations are important in order to differentiate
between healthy (S = 0, Fe(II) in HbO2 and S = 2, Fe(II) in Hb) and unhealthy (S = 1/2, Fe(III) in HbCN, HbN3,
HbOH and S = 5/2, Fe(III) in HbH2O, i.e. methaemoglobin), HbF bloods. MB data on Hb and its derivatives have
been summarised in Table 10.18.
Table 10.18 Isomer Shifts (dIS in mm/s) and Quadrupole Splittings (ΔEQ in mm/s) for haemoglobin and its Derivatives.
Compound Temperature(K) dIS ΔEQ Configuration

Hb 195
4
0.90
0.91
2.40
2.40
} High spin S = 2,
Fe(II)
HbO2 195 0.20 1.89 ⎫
⎪
(0.00) (1.89)⎪
77 0.26 2.19 ⎪
⎪
(0.06) (2.19)⎪
⎪
⎬
1.2 0.24 2.24 ⎪
(0.04) (2.24)⎪ Low spin S = 0,
⎪
HbCO 195 0.18 0.36 ⎪ Fe(II)
⎪
4 0.26 0.36 ⎪
⎭
HbCN 195 0.17 1.39 ⎫
⎪
⎪
⎪
HbN3 195 0.15 2.30 ⎪
⎪
Low spin S = 1/2,
⎪
⎬
⎪
HbOH 195 0.18 1.57 ⎪ Fe(III)
⎪
⎪
77 0.20 1.90 ⎪
⎪
⎭
HbNO 195−1.2 Magnetically broadened

HbH2O
HbF
195
195−1.2
0.20
0.43
2.00
(0.67) } High spin S = 5/2,
Fe(III)
Magnetically broadened
The values within parenthesis are for rat HbO2.
(a) Deoxy/Reduced Haemoglobin (Hb) It is present in arterial and venous blood. Magnetic susceptibility
data show that Hb is high spin S = 2, iron (II) compound in the solution phase whereas its magnetic susceptibility
in powder form is less than 2. The data also revealed that iron atom in this compound lies out of the plane of the
haeme group.
The MB spectrum of Hb consists of a quadrupole doublet with MB parameters ΔEQ = 2.4 mm/s and dIS = 0.90
mm/s at 195 K (Table 10.18). Data reveal that Hb is a high spin S = 2, ferrous Fe(II) compound even though the
value of isomer shift is less than the expected value for high-spin ionic ferrous compounds (Table 10.4). Magnetic
hyperfine splitting is not observed in the zero field. The broadening of the quadrupole doublet at low temperatures
(4 K) in the presence of strong external magnetic field (7.5 and 30 kG) perpendicular to the direction of observa-
tion may be because of anisotropy of ferrous ions. The applied field produces a large internal field which is dif-
ferent for each of the random orientations of the haeme plane and which broadens the spectrum. The absence of
internal magnetic field hyperfine interaction may be due to the fact that an ion with even number of electrons will
have (2S + 1) spin degeneracy removed completely by a small crystal field provided the crystal-field symmetry is
low, i.e. the electron spin-lattice relaxation time for ferrous is rapid compared to the life of the MB excited state
even at low temperatures so that the magnetic field at the nuclei averages to zero.
(b) Oxygenated Haemoglobin (HbO2 ) Just like reduced haemoglobin, HbO2 also occurs in arterial as well
as in venous blood and the iron atom in HbO2 lies in the plane of the haeme group. Magnetic susceptibility data
show that iron exists as low spin S = 0 ferrous (II) state in HbO2. This also rules out the possibility of oxygen lying
perpendicular to the plane of haeme group.
MB spectrum of HbO2 at 195 K consists of a quadrupole doublet with MB parameters, ΔEQ = 1.89 mm/s,
d = 0.2 mm/s (Table 10.18). It is difficult to assign low-spin ferrous or ferric state from the isomer shift and ΔEQ
data (Table 10.18). However, the spectrum of HbO2 under the influence of external magnetic field (30 kOe at 4.2K)
shows that the total field at the nucleus equals the applied field suggesting no magnetic hyperfine field, i.e. Hint, as
expected for low-spin S = 0, ferrous (II) state. The applied field also suggests that the sign of e2q Q is negative.
One interesting feature of HbO2 spectrum is that the quadrupole splitting is unusually temperature dependent
for a diamagnetic configuration (Table 10.18), e.g. the quadrupole splitting at 4.2 K is 2.24 mm/s and decreases to
1.89 mm/s on warming to 195 K. This may be attributed to the existence of low-lying excited state, i.e. paramag-
netism which of course is not observed. However, it could be because of a possible libration motion of the oxygen
molecule at higher temperatures.
The sharp decrease in dIS in going from Hb (0.90 mm/s at 195 K) to HbO2 (0.20 mm/s at 195 K) suggests some
d–orbital and p-bond interaction between iron and oxygen in the HbO2 molecule. There may be an electron trans-
fer from iron to oxygen or one or more d-electrons are back donated. However the interaction between the metal
dyz–orbital and the p*-antibonding orbitals of oxygen responsible for the loss of oxygen molecule’s paramagnetism
and the covalent bonding effectively removes an electron from the dyz orbital. This conforms to the negative sign
of e2qQ measured by the applied magnetic field which suggests an electron hole in the dyz orbital. The scheme for
geometrical arrangement and bonding of molecular orbitals in HbO2 is shown in Fig. 10.61. The molecular orbital
(d yz + z ) has only one-half character of dyz orbital.
2
(c) Haemoglobin–Carbon Monoxide (HbCO) The MB spectrum of HbCO consists of a small quadrupole
doublet (ΔEQ = 0.36 mm/s) at 195 K (Table 10.18). The quadrupole splitting has little temperature dependence,
and there is a complete absence of magnetic interactions. The value of ΔEQ = 0.36 mm/s and dIS = 0.18 mm/s at
195 K in HbCO conforms to low spin S = 0, Fe(II) configuration as expected (Table 10.4).
Let us now examine the 57Fe MB spectrum of red blood cells in the light of the following exchange reaction
occurring in such cells.
2 − haemoglobin + CO 2 CO 2 − haemoglobin + O 2
MB spectra of human and rat blood cells enriched to 35 per cent 57Fe exhibit two doublets, a weak one and a strong
one. These are attributed to O2-haemoglobin and CO2-haemoglobin. An additional weak single resonance line is
OZ OZ
+ OX − OX
Z −
+
− X Y
+
− +
− +
Fe(dyz)
+ −
(dyz−πz∗)/√2
∗ ∗ π∗X
π Xπ Z dyz
O O
dxz
(dyz−πz∗)/√2
dxy
O2 Complex Fe
Fig. 10.61 Schematic diagram of the bonding in oxyhaemoglobin showing how the interaction between oxygen π z* molecule orbital and the
iron dyz orbital removes the degeneracy and causes spin pairing on the oxygen molecule.
also observed in MB spectrum of human blood. However CO2 in CO2-haemoglobin is not attached to iron but to
another part of haemoglobin since the spectra of blood which is exposed separately to an atmosphere of nitrogen
and carbon dioxide are the same. It is known that terminal amine groups produced in muscles bind CO2 for trans-
port to the lungs. Thus, site 6 of iron may be either vacant or loosely bound to a molecule of water.
(d) Haemoglobin Cyanide (HbCN), Haemoglobin Azide (HbN3) and Haemoglobin Hydroxide
(HbOH) In all the three compounds, a symmetrical (with respect to width and intensity) hyperfine quadrupole
doublet is observed at high temperature, i.e. 195 K. The values of quadrupole splitting and isomer shifts. (Table
10.18) lie within the expected range for the low spin S = 1/2, Fe(III) compounds (Table 10.4). This observation
complements the results obtained from ESR and magnetic susceptibility measurements on these compounds. The
marked low value of isomer shift in these compounds compared to reduced haemoglobin (Table 10.18) suggests that
—CN, —N3 and —OH groups are attached to the site 6 of iron atom.
The doublet at 195 K becomes broadens on cooling the sample and is illustrated in case of HbCN in Fig. 10.62.
Inspection of Fig. 10.62 reveals that at 77 K, the wing of the doublet on the low-velocity side is more intense
compared to its counterpart on the high-frequency side of the spectrum. The broadening at 77 K relative to 195 K
is due to slower spin relaxation rate which is responsible for the lack of magnetic splitting at higher temperatures.
The asymmetry in the spectrum is attributed to the anisotropy of the magnetic hyperfine interaction. At 4.2 K,
the relaxation is so slow that the hyperfine pattern is well resolved, but complex and asymmetrical because of the
reason reported above.
(e) Haemoglobin Nitric Oxide (HbNO) MB spectrum of HbNO exhibits paramagnetic broadening at all
temperatures between 195 and 1.2 K (Fig. 10.63). The magnetic hyperfine structure suggests a strong covalent
bonding and a large spin transfer from the paramagnetic NO molecule to iron. The broadening is so strong that it
is not possible to define the state of iron (ferrous or ferric) with certainty. This inference from Mössbauer studies
is compatible with the reports in the literature that some haemoglobin compounds previously assumed to be fer-
rous contain ferric ions also.
(f) Haemoglobin Fluoride (HbF) and Methaemoglobin (HbH2O) In both the compounds, ferric being in
a S state has zero orbital magnetic moment. This results in very slow s-l relaxation rates and isotropic magnetic
6
properties. The zero field MB spectrum of HbF exhibits magnetic broadening′ even at high temperature (195
K) because of which it becomes difficult to obtain accurate values of quadrupole splitting and isomer shift. The
zero-field spectrum remains unresolved even at low temperatures because of coupling of electron spin (trans-
ferred hyperfine interaction) with the magnetic moment of the neighbouring fluorine nucleus and to a lesser
extent with a nitrogen nuclei in the haeme group. However, under the influence of a very small external mag-
netic field (≈ 100 Oe) perpendicular to the ligand field axis (i.e. g -ray beam), an almost symmetrical six-finger
splitting is observed. This marked change in spectrum is due to the decoupling by the external magnetic field
106 × 0.37
(a)
(a)
0.98 × 106
(b)
0.35 (b)
Counts
3.28
(c) 0.92
(c)
3.19
−15 −10 −5 0 5 10 15
Fig. 10.62 Mössbauer spectra of haemoglobin cyanide at (a) 195 K, (b) 77 K, and (c) 4.2 K.
1.01
0.585
1.00
(a)
0.580
Counts × 10−6
counts × 10−6
0.575
0.99
0.570
(b)
0.565
−6.0 −2.0 0 2.0 6.0

Velocity (mm/s)
Fig. 10.63 Mössbauer spectra of haemoglobin nitric oxide at (a) 195 K, and (b) 77 K.
of electron spin from the nuclear spins of 57Fe, 19F and 14N nuclei. The small quadrupole splitting and six sharp
fingers are characteristic of the high spin S = 5/2, ferric (III) configuration. This inference is compatible with the
ESR studies of HbF.
The MB spectra of this compound at 4 K and under external magnetic fields of 30, 15 and 7.5 kOe look like a
superposition of two six-line spectra caused by different effective magnetic fields of the upper and lower members
1
of the magnetically split Ms > ± doublet. Under the influence of high external magnetic field, the higher ligand
2
field states get mixed with the ground state |MS =± 1/2> in such a way that the effective field for the |Ms = −1/2> state
increases while that of the |MS = +1/2> state decreases. The observed of value of effective magnetic field which in
5
governed by ling and field is about 300 kOe is consistent with that of 500 kOe usually found when Ms = .
2
In HbH2O, H2O is in sixth coordination position. ESR spectra and magnetic susceptibility are consistent with
high spin S = 5/2, Fe(III) configuration.
MB behaviour of this compound is similar to that of HbF, though (i) the temperature dependence of magnetic
hyperfine spectra shows that the spin-lattice relaxation is much faster than what could be expected on the basis
of pure s-state ion, and (ii) the quadrupole splitting observed in this compound is large (2.00 mm/s) compared to
HbF (0.67 mm/sec). Both the points suggest that there is considerable mixing in of orbital moment into the ferric
states by the ligand field.
10.16.2 Studies of Peroxidase and Its Derivatives

The MB spectrum of the peroxidase enzyme is of special interest because iron in these compounds can exist in a
higher oxidation state than Fe(III). The MB data on Japanese radish peroxidase-a (JRP-a) enriched in Fe and its
derivatives are compared with their respective derivatives of hemoglobin in Table 10.19.
10.16.3 Studies of Iron–Sulphur Proteins

Fe in these proteins is not bound in a stable basic structural unit as in the case of haemoglobin. It is held loosely in
the chain of amino acids via sulphur atoms. The MB effect has been able to observe directly the antiferromagnetic
coupling between the pair of iron atoms in such proteins.
(a) Ferredoxins (Scenedesmus is Involved in Plant Photosynthesis) (2-iron Proteins) Ferredoxin from
plants contains two atoms of iron and two atoms of inorganic sulphur. The MB spectra of oxidised and reduced
Table 10.19 MB Data on JRP-a and its Derivatives.
Compound T (K) ΔEQ dIS(Fe) Configuration Remarks

(mm/s) (mm/s)
JRP-a 243−77 1.67−2.20 0.3 1−0.37 5 Large quadrupole splitting which is temperature
S= , Fe(III) dependent, splitting is asymmetrical below 120 K.
4.2 Magnetically broadened 2 This behaviour is similar to methaemoglobin.
243−120 1.84−2.36 0.27−0.22 All the three show similar behaviour but more
⎡JRP - a OH ⎤
magnetic broadening in hydroxide and azide as
⎢ ⎥ 120 2.22 0.31 1
⎢JRP - a N 3 ⎥ S= , Fe(III) compared to cyanide because of longer relaxation
2 times. Behaviour is similar to respective haemo-
⎢JRP - a CN ⎥ 195−77 1.61−1.81 0.11−0.15
⎣ ⎦ globin derivatives.
Reduced JRP-a 243−77 1.79−2.32 0.79−0.81 S = 2, Fe(II) Spectrum is analogous to reduced haemoglobin
JRP-a CO 77 0.21 0.32 S = 0, Fe(II) Spectrum is analogous to HbCO.
JRP-a F 195−4.2 Magnetically 5 Substantial magnetic broadening between 105−4.2

S= , Fe(III)
broadened 2 K. Behaviour is similar to HbF
JRP-a H2O2 (I) 77 1.33 0.10 S = 1, Fe(IV) ⎫

with two un- ⎪
paired electrons
⎪ All the three give sharp quadrupole splitting which
⎪ is small and temperature dependent. Compounds I
JRP-a H2O2 (II) 77 1.46 0.11 S = 1, Fe(IV)
⎪ and II can't be distinguished from isomer shifts since
⎪
with two un- ⎬ dIS (I) ≈ dIS (II). Compound III is similar to oxyhae-
paired electrons ⎪ moglobin.
⎪
JRP-a H2O2 (III) 77 2.37 0.29 S = 0, Fe(II) ⎪
⎪
Fe3+
Oxidised
(a)
Fe2+
Reduced
Fe3+
(b)
0.5 percent
Absorption
−3 −2 −1 0 1 2 3
velocity (mm/s)
Fig. 10.64 Mössbauer spectra of (a) oxidised, and (b) reduced form of ferredoxin at 195 K.
forms of ferredoxin at 195 K (Pd 57Co source) are shown in Fig. 10.64. Broadly speaking, the spectrum of the
oxidised form [Fig. 10.64 (a)] appears to consist of two resonances of nearly equal intensity, while on reduc-
tion, clearly two doublets are observed, one due to Fe2+ and one due to Fe3+ [Fig. 10.64 (b)]. In both forms of the
protein, the iron atoms are believed to be coupled together antiferromagnetically. When oxidised, there are two
Fe3+ atoms and their total spin is zero which agrees with their observed temperature independent susceptibility.
However, the widths of the resonances are slightly more than that for a standard absorber. The analysis showed
that the spectrum is made up of two quadrupole doublets having almost equal quadrupole interaction but slightly
different isomer shifts, suggesting that the compound contains iron atoms in two inequivalent sites.
At low temperatures (4 K and below) no magnetic hyperfine interaction is noticed in the oxidized protein
even under the influence of external magnetic field, suggesting the nonmagnetic nature of the molecule in the
oxidised state. This leads to conclude that in the oxidised form of protein, the two iron atoms are coupled together
antiferromagnetically to give zero spin which is consistent with the observed temperature independent magnetic
susceptibility as stated earlier.
Based on the MB data, a model of active centre of ferredoxin has been proposed by Moshkovskii. According to
this model, seven iron atoms are distributed linearly and are bound to each other by seven atoms of sulphur that are
part of cysteine molecule and by six atoms of inorganic sulphur. Such a model has two related but non-equivalent
environments, viz., two iron atoms at the ends of the chain and the remaining iron atoms within the chain.
The spectrum of the reduced form consists of two clear doublets [Fig. 10.64(b)], i.e. ΔEQ = 0.56 and 2.66 mm/s.
The isomer shifts and quadrupole splittings corresponding to these doublets are different, suggesting that the
compound in the reduced form has iron atoms in two non-equivalent sites.
The MB data, viz., dIS and ΔEQ corresponds to high-spin S = 2, Fe2+(II) and high spin S = 5/2, Fe3+ (III)
configurations. The coupling between Fe3+ atom (S1 = 5/2) and the Fe2+ atom (S2 = 2) gives rise to a ground state
with total spin S = S1 + S2 = 1/2. This model accounts for the unusual g-values centred around 1.94 found from
ESR spectra of these molecules.
The spectrum of the reduced form of the protein shows magnetic hyperfine splitting at low temperatures, i.e.
4.2 K since at these temperatures the electron spin relaxation time becomes long compared to ⎛⎜ h ⎞⎟ where A
⎝ 2π A ⎠
is the hyperfine coupling. The spins S1 and S2 precess about their resultant S, and the observed spectrum is asym-
metrical and has broad and partly resolved lines. When a small magnetic field Hext > AS/gb is applied, a more
symmetrical spectrum with sharper lines results. The effect of the field is to cause S1 and S2 to precess about it
and results in two separate effective magnetic fields at the nuclei. The resulting spectrum is the superposition
of two contributions, one from Fe3+ and one from Fe2+ and the lines from each are strongly overlapping. Under
high magnetic field (30 kOe, at 4.2 K), the spectrum splits into two components, each having different effective
magnetic fields with opposite signs also, i.e. −180 kOe (Fe 3+) and +120 kOe (Fe2+). This observation suggests that
the two iron atoms are coupled antiferromagnetically in reduced ferridoxin also. Thus MB spectra provide direct
evidence for antiferromagnetic coupling between the Fe2+ and Fe3+ iron atoms in plant type ferredoxins. Atoms
with spins pointing up (i.e. parallel to resultant spin, S) have effective magnetic field at the nucleus reduced by
the external field (since the hyperfine field is negative) so that their spectrum is compressed by the field. On the
other hand, atoms with spins down, have their effective field increased, i.e. spectrum appears to expand in the
external field.
(b) 4- and 8-iron Proteins These proteins include the bacterial ferredoxins which may contain four iron atoms
per molecule, e.g. (Bacillus ferredoxin) or light iron atoms, e.g. from chromatium or clostridium as well as the
high-potential iron proteins (HIPIP’s) four iron atoms per molecule and the conjugated iron–sulphur proteins, e.g.
flavoproteins, iron-molybdenum proteins containing eight or more iron atoms per molecule. The MB spectrum
data for 57Fe in oxidised and reduced iron–sulphur proteins at 195 K is given below.
State/Protein Oxidised Reduced
dIS ΔEQ dIS ΔEQ
Ferredoxin (plant) 0.22 0.59 0.56(a) 2.75(a)
HIPIP 0.28(b) 0.77(b) 0.38(b) 1.01(b)
Ferredoxin (bacterial) 0.39(b) 0.75(b) 0.52(b) 1.07(b)

(a) Value for ferrous atom (b) Average value
Many features of the bacterial ferredoxins resemble with the simpler plant-type ferredoxins, i.e. their oxi-
dised form is nonmagnetic and have negative redox potentials, while their reduced form is magnetic with an
average g-value less than two (g < 2). Their MB spectra are more complex than 2-iron proteins. In eight iron
proteins, the high-temperature MB spectra do not show distinct quadrupole doublets corresponding to Fe3+
and Fe2+ states. At high temperatures (77 K and above), the spectra of both the oxidised and reduced form of
proteins consist of the superposition of two or more closely similar doublets. The average isomer shift for the
oxidised protein suggests that each of the two four-iron active centres consists of two Fe3+ and two Fe3+ atoms.
Molecular structure of Peptococuss aerogenes shows that the iron in this 8-iron protein is in units containing
four atoms close together (Jensen, Sieker, Watenpaugh, Adman and Herriott; Biochem. Soc. Trans. 1 (1973)
27). Both the average isomer shift and the quadrupole splitting increase in going from oxidised to reduced
form. This conforms to one Fe3+ and three Fe2+ atoms per sub-unit in the reduced molecule. This change in
isomer shift and quadrupole splitting on reduction shows that all the iron atoms are affected when one electron
is added to each four-iron centre. The non-observance of spectra for Fe3+ and Fe2+ (as they are in reduced two
iron plant ferredoxin) suggests that d-electrons are not localised on particular atoms but are shared equally by
all four atoms in the four-iron centres.
The low-temperature (4 K and below) spectra of oxidised form do not show hyperfine splitting even when
external magnetic field is applied confirming, thereby, the nonmagnetic nature of the molecule in the oxidised
state. This suggests that the four-iron atoms in each sub-unit are antiferromagnetically coupled together to
give zero spin in the oxidised state. On the other hand, magnetic hyperfine splitting has been observed in case
of reduced form at low temperatures. This shows that all the iron atoms are magnetic. This demonstrates that
one electron goes to each centre on reduction. Further, both positive and negative hyperfine fields have been
observed when large external magnetic field is applied to the reduced form at low temperatures. Thus MB
spectra provide direct evidence for antiferromagnetic coupling between iron atoms in bacterial as well as plant
type ferredoxins.
The HIPIPs have four iron atoms per molecule all close together in a cluster. They have positive redox
potentials and are nonmagnetic in the reduced form but they lose one electron per molecule when oxides and
become magnetic with an average g value greater than two (g >2). MB data on the reduced state confirm that it
is nonmagnetic. MB spectra of oxidised form in the presence of applied magnetic fields show both positive (+90
kOe) and negative (−120 kOe) hyperfine fields. This shows antiferromagnetic coupling between iron atoms in the
oxidised form of HIPIPIs. The isomer shifts and quadrupole splitting suggests that there are two Fe3+ and two Fe2+
atoms in the nonmagnetic reduced state and three Fe3+ atoms and one Fe2+ atom in the oxidised state. Since no
separate spectrum is observed for Fe3+ and Fe2+ shows that the d-electrons are mobile and not localised on particu-
lar iron atoms. The distribution of Fe2+ and Fe3+ atoms in the four iron centres of the oxidised and reduced forms
of 4- and 8-iron proteins is summarised in the tabular form as below.
State/Protein Reduced Oxidised
HIPIP 2Fe2+ + 2 Fe3+ 1 Fe2+ + 3 Fe3+
Clostridium 3Fe2+ + 1 Fe3+ 2 Fe2+ + 2 Fe3+
10.16.4 Studies of Cytochrome−C

The enzyme cytochrome−C also contains a haeme unit attached to a protein and is associated with the oxidation
of nutrients in the cell, i.e. the biological activity of cytochrome−C is associated with the reduction and subse-
quent oxidation of the haeme which occurs during the electron transfer down the enzyme chain. The MB spectra
as well as the valence electron population on the basis of molecular orbital calculations of both the oxidised and
reduced cytochrome−C are shown in Figs 10.65 and 10.66 respectively. The MB spectra of both the oxidised and
reduced cytochrome−C (frozen solutions) consist of a symmetrical quadrupole doublet at high temperature (225 K).
The quadrupole splitting of 2.3 mm/s in case of oxisided cytochrome−C is attributed to electron hole in the dxz
orbital. Dry cytochrome shows splitting of ~1.9 mm/s. The difference between the solid and solution state spectra
may be attributed to the hydration of various protein chains in the vicinity of the haeme. However, the observed qua-
drupole splitting of 1.2 mm/s in case of reduced cytochrome−C could not be because of electron population in the
t 2g orbital as in the case of oxidised cytochrome−C since all the electrons in the t 2g orbital exist in pairs. The split-
ting could be due to lattice EFG or because of selective delocalisation of 3d electrons. The observed isomer shift
of 0.38 mm/s and quadrupole splitting of 1.2 mm/s show that iron exists as low spin S = 0 ferrous (II) in reduced
cytochrome−C.
The MB spectrum of oxidised cytochrome-C becomes broaden and asymmetrical at lower temperatures between
150 and 100 K (broad magnetic spectrum at 4.2 K). The broadening followed by decrease in intensity of the wing
of the doublet on the high velocity side is more as compared to its counterpart on the low-velocity side of the
225 K
0.000
0.002 dX2−y2
eg
dz2
150 K
Absorption 0.005
0.010 dxz
100 K
0.00 t2g dyz
0.06
dxy
0.12
−2 −1 0 +1 +2 +3
velocity (cm/s)
Fig. 10.65 Mössbauer spectrum of oxidised cytochrome−C in frozen solution.
dX2−y2
0.0 eg
dz2
0.5
Absorption
1.0 dxz
t2g dyz
1.5
dxy
−2 −1 0 +1 +2 +3
velocity (cm/sec)
Fig. 10.66 Mössbauer spectrum of reduced cytochrome−C in frozen solution.
spectrum. The broadening is attributed to the magnetic hyperfine interaction which results because of slow-spin
relaxation rates at low temperatures. The chemical isomer shift (relative to Fe) of 0.18 mm/s at 100 K, and magnetic
broadening are compatible with low spin S = 1/2 iron (III) configuration with an electron hole in the dxz orbital.
10.16.5 Ferritin
Ferritin is widely distributed throughout the various organs of mammals, with particularly high concentrations
being found in liver, spleen and bone marrow. Physiologically, it represents a depot in which surplus iron can be
stored within the cell in a nontoxic form and from which it can be mobilised as and when required either within
that particular cell itself or in other cells of the organism. In cells of erythroid series (i.e. cells which are precursors
of erythrocytes and which synthesise haemoglobin), transferrin molecules carrying Fe3+ bind to specific receptor
sites on the cell membrane and transfer iron to the cells. This iron, probably, after reduction to Fe2+ is then incor-
porated into ferritin; the process involves oxidation to the ferric state. Ferritin iron then serves as a source of iron
for haeme synthesis and its mobilisation requires reduction to Fe2+.
Ferritin consists of a shell of protein sub-units surrounding a core of ferric hydroxy phosphate. Apoferritin is
the term used for the protein part of ferritin. Native apoferritin is the term used to describe the nearly iron-free
protein which can be isolated from ferritin by centrifugal techniques. Thus simple equation can be written to
describe ferritin formation.
Fe2+ (Ferrous ammonium sulphate or ferrous sulphate)
+ O 2 + Apoprotein ⎯buffer,
In bicarbonate
pH 6.8 − 8.0
→ Ferritin
t
The MB spectrum at liquid nitrogen temperature and above (Fig. 10.67) has an isomer shift and quadrupole
splitting of 0.47−0.50 and 0.6−0.74 mm/s respectively. At low temperatures (T = 11 K) a six-finger hyperfine
spectrum is observed, characteristic of magnetic ordering. As the temperature is raised, the hyperfine spectrum
is replaced by the paramagnetic quadrupole-split doublet spectrum. This super magnetism is because of antifer-
romagnetism of fine particles (super-antiferromagnetism) analogous to that seen in certain metal oxides reduced
to fine powders.
The ESR and MB spectra of iron in ferritin over the range 4.2−295 K confirms that the temperature dependence
of the spectra can be explained by the supermagnetic properties of the micelles. A weak ESR signal is also observed
near g = 4 at 77 K and above and this is attributed to a small number of Fe3+ which are not bound to the micelle.
The MB parameters of ferritin and a synthetic polymer derived from hydrolysed ferric nitrate has a similar room
temperature MB spectrum with isomer shift and quadrupole splitting of 0.48 and 0.67 mm/s respectively. At very
low temperatures the spectrum of the polymer is also closely similar to ferritin.
The magnetic susceptibility of ferritin is markedly affected by magnetic field. At very high field meff = 5.08
B.M. The Curie point situated at −200 K proves the existence of a strong negative coupling between the atoms.
MB studies show the progressive establishment of a magnetic order below 30 K. meff (275−301 K) = 3.81 B.M. meff
in the temperature range 4.2 − 300 K shows that ferritin is antiferromagnetic with a Neel temperature of 20 ± 3 K.
T = 11 K
−8 −4 4 8
T = 86 K
T = 293 K
−2 −1 0 +1 +2
velocity (mm/s)
Fig. 10.67 Mössbauer absorption spectrum obtained using a 57Co in Pd source at room temperature and ferritin as absorber at 11.86
and 293 K.
PROBLEMS
1. Using the data in Table 10.5 on p.i.s., predict the isomer (e) Fe(CN) (EtNC)5ClO4
shifts for [Ans. Calc. (obs) in mm/s: (a)-0.78 (0.73) (b) 0.52 (0.50)
(a) cis-Fe (SnCl3)2 (ArNC)4 (c) – 0.52 (0.32) (d) 0.30 (0.29) (e) – 0.30 (0.17).]
(b) [FeCl(ArNC)5] ClO4 8. The MB technique is only applicable for the range of life-
(c) trans-FeHCl (depe)2 time 10–7–10–10 second. From the following data on line
(d) trans-Fe(CN)2(MeNC)4 width of the 197Au transmission at different thickness of
(e) trans-Fe(CN)2 (EtNC)4 metallic gold, calculate the half life of the l97Au isotope.
[Ans. Calc(obs) in mm/s: (a) + 0.24 (+ 0.27)
(b) + 0.26 (+0.23) (c) + 0.42 (+ 0.39) Ta(mg/cm2) 20 50 75 134 180
(d) + 0.18 (+ 0.16) (e) + 0.18 (+ 0.21).]
G exp(mm/s) 2.03 2.19 2.33 2.64 2.89
2. Calculate the critical g -ray energy for
(a) 57Fe (qD = 400K, f (T = 0) = 5.4 × 10–4 [Ans. t1/2 = 9.8 ns]
(b) 192Os (qD = 315K, f(T = 0) = 1.6 × 10–3 and 9. Assign the data given below to the cis and trans-forms of
(c) 139La (qD = 270K, f(T = 0) = 1.5 × 10–3 FeCl2(ArNC)4 where ArNC is p-methoxyphenyl isocyanate.
[Ans. (a) 136.5 keV (b) 206 keV (c) 163 keV]
ΔEQ(mm/s) dIS (SS)/ dIS(Fe) mm/s
3. Calculate the recoil energy for the
(a) 14.4 keV transition of 57Fe − 0.83 0.12/0.03
(b) 23.87 keV transition of 119Sn
(c) 57.60 keV transition of 127I + 1.59 0.20/0.11
[Ans. (a) 1.928 × 10–3 eV (b) 2.537 × 10–3 eV 10. From the data on tetrahedral high spin, (Et4N)2FeCl4,
(c) 13.845 × 10–3 eV.] calculate the value of quadrupole splitting at zero K and
4. Draw the energy-level diagram and the stick spectrum the energy difference between two dg levels.
7 5
for the MB → transition of 127I in Temperature K dIS (mm/s) ΔEQ (mm/s) G (mm/s)
2 2
(a) polycrystalline state 80 1.00 2.49 0.79
(b) q = 0°
(c) q = 90° indicating intensity and polarisation of the 190 0.93 1.63 0.75
lines in the spectrum [Ans. 2.70 mm/s, 180 cm−1]
3 11. For tin-119, calculate the line width and Doppler velocity
5. (a) For 119Sn, me = + 0.67 nm; mg = –1.047 nm, Ie =
1 2 to achieve resonance absorption. The transition energy
and Ig = . Prove that the separation between any of the 3/2 level is 23.8 keV, the half life is 19 ns.
2
two consecutive magnetic hyperfine levels in the 12. The MB spectrum of 57Fe ferrocene, Fe(C5H5)2 at 20 K
excited state is small as compared to their counter- consists of two lines at –0.50 and +1.88 mm/s relative to a
parts in the ground state. stainless steel source. Calculate the isomer shift and eqQ
(b) Show that for 57Fe (14.4 keV), Q for the excited nucleus. Assume cylindrical symmetry.
f = 7.9 MHz
[Ans. Shift MH ; eq Q = 55 MHz ]
ge μ
=+ e 13. Predict the positions of quadrupole hyperfine lines in the
gg 3μ g MB spectrum of 129I in KIO3 from the data reported in
Example 10.5(a).
(c) For 127
I, Ig = 5/2, mg = + 2.809 nm, Ie = 7/2 and me = 14. The MB technique is only applicable for the range of
+ 0.202. Prove that the separation between any two lifetime of 10–7–10–10 second. Calculate the range of line
consecutive Zeeman levels in the excited state is width of the Mössbauer line.
approximately one half as compared to their coun- 15. The MB parameters as deduced from the MB spectra of
terparts in the ground state, i.e. b > a. 129
I in I2 and in triethylamine (TEA) – I2 complex (at 16K in
6. From the data given below, calculate the bandwidth (in CS2 and with respect to ZnTe source) are
mm/s) of the respective nuclei. Also, determine the half
life of the excited state. System e2qQ (MHz) δ(mm/s) h(per cent)
Nucleus v(cm/s) Eγ (keV) l2 −2242 0.93 11
57
Fe 9.6 × 10–3 14.4125 TEA−I2(A) −2604 1.56 …
67
Zn 1.5 × 10–5 93.31 (B) −1261 0.40 9
197
Au 9.4 × 10–2 77.35 The MB lines corresponding to (A) are more intense than
(B) and the quadrupole coupling constant (e2qQ) has
[Ans. G (in keV): 57Fe: 4.612 × 10–12; 67Zn; 4.665 × 10–14 ; been converted to 127I. (a) Calculate the charge densities
197
Au; 2.423 × 10–10 ] on the terminal and bridged iodine atoms from the qua-
7. Using the data on p.q.s. from Table 10.12, calculate the drupole coupling constant as well as from the isomer shift
quadrupole splitting for data. (b) Determine the p bond character in the complex
(a) FeCl(ArNC)5ClO4 and (c). Comment on the nature of chemical bond in the
(b) cis-Fe(SnCl3)2(ArNC)4 charge transfer TEA − I2 complex. e2Qqatom = 2293 MHz
(c) Fe(SnCl3) (ArNC)5ClO4 (for iodine atom)
(d) cis-Fe(CN)2 (EtNC)4 [Ans. (b) 8 per cent]
CHAPTER 11 MASS SPECTROMETRY
To keep a lamp burning we have to keep putting oil in it.

—Mother Teresa
Never give up! If the end of the world were imminent, I still would plant a tree today.
—Unknown
11.1 INTRODUCTION
Broadly speaking, Mass Spectrometry (MS) deals with the sorting out the mass to charge (m/e) ratios of the posi-
tively charged fragments produced by bombarding the analyte (a few milligrams to submicrogram or even nano-
gram quantities in case of liquids and solids provided they are in the gaseous state and 0.1 mL to 10–8 mL in case
of gases) with medium energy electrons (25–70 eV), under conditions of high vacuum (10−7−10−9 mm of Hg) as a
function of their kinetic energies/intensities by applying an electric field and a magnetic field. MS has a dynamic
history dotted with Nobel laureates and a continually advancing technology that has made significant inroads into
drug discovery, protein characterisation, and even disease diagnosis. The history of science reveals that MS had
its roots in physics, branched into chemistry and in the past two decades or so, has budded into biology.
The theoretical basis of MS lies in the Lorentz rule, according to which the force exerted by a magnetic field of
strength H on an electron travelling at the velocity v at right angles to the lines of force is
q
F= v• H (11.1)
c
which reduces to F = evH for electrons assuming c = 1 (11.2)

where, e is the charge of the particle (the law is valid not only for electrons, but for any charged particles).
When however a charged particle travels along the line of force in a magnetic field, the field has no effect on the
particle. The above equation (11.l) is called the Lorentz equation after H A Lorentz, the Dutch physicist who
proposed it in 1895. For his theoretical explanation of the Zeeman effect, he shared the Nobel prize in physics
with P Zeeman in 1902. W Wien who got the Nobel prize in 1911 for laws governing radiation of heat was the
first to show in 1898 that positively charged electrical entities produced in electric discharge anode tube were also
deflected in a magnetic field. This paved the way for mass spectrometry. But the first mass spectrometer (then
called the parabola spectrograph) was constructed by J J Thomson in 1912 in which positively charged fragments
of polyatomics (e.g. those corresponding to COCl2+, CO+, C+, O+, etc., from COCl2), produced in an electric dis-
charge anode tube, were separated by their different parabolic trajectories in electromagnetic fields and detection
occurred by the ions striking a fluorescent screen or photographic plate. Note that he was the first to obtain a plot
of the ion current as a function of mass to charge (m/e) ratios. However, it was not until 1920 that F W Aston first
introduced the term ‘mass spectrum’ for such plots. The clear demonstration of mass spectrometry came at the
end of World War I by F W Aston (Thomson’s protege) who, in 1919, designed a mass spectrometer in which he
used both electrostatic and magnetic fields to focus ions produced in electric discharge tubes on a photographic
plate. In case of neon, he obtained two mass lines on the photographic plate at 20 and 22 with the intensity ratio of
10 : 1 consistent with the average mass of 20.20, the known atomic mass of neon. Note that Thomson was not able
to assign these lines to the isotopes of Ne. Aston was presented with the Nobel prize in chemistry in 1922 for iso-
topic studies; he demonstrated the existence of isotopes in some thirty other gaseous elements. The first electron
impact ion source and magnetic deflection mass spectrometer was introduced by A J Dempster in 1918. A beam
of electrons from a hot-wire filament was used for the ionisation of volatilised molecules. Dempster’s magnetic
deflection MS could not be used for precise mass measurements, but it was better suited than Aston’s instrument
Mass Spectrometry 791
for measuring relative abundances of the ionic species and was suitable for studying electron impact processes in
gases. Electron impact ion sources are still widely used in modern mass spectrometers.
Thus, by 1920, the early instruments were capable of handing three types of measurements, viz., (i) precise
mass determination, (ii) measurement of relative abundances of ions, and (iii) electron impact studies. However,
the full potentialities of the methods of mass spectrometry were realised when in February,1942, the first com-
mercial mass spectrometer was built and this was considered the birth of the present-day mass spectrometry. The
chronology regarding the historical developments in present-day mass spectrometry is summarised below.
Year Contribution
1946 Time of flight mass analysis
1949 Ion cyclotron resonance
1953 Double focussing instruments
1953 Quadrupole analysers
1956 High resolution mass spectrometry
1966 Peptide sequencing
1966 Chemical ionisation
1968 Electron spray ionisation
1969 Field desorption mass spectrometry of organic molecules
1974 Plasma desorption mass spectrometry
1974 Fourier-transform spectrometry
1978 Triple quadrupole mass spectrometry
1981 Fast atom bombardment
1983 Matrix assisted laser desorption/ionisation
1984 Electron Spray Ionisation (ESI) on biomolecules
1990 Protein conformational changes with ESI mass spectrometry
1991 Micro ESI
1991 Non covalent complexes with ESI mass spectrometry
1993 Oligonucleotide ladder sequencing
1993 Protein mass mapping
1996−2001 Intact viral analysis
J B Fenn and K Tanaka shared the 2002 Nobel prize in chemistry of their works on developing
2002
soft ionisation techniques for large biomolecule analysis
11.2 COMPARISON OF MASS SPECTROMETRY WITH OTHER

SPECTROSCOPIC TECHNIQUES
The techniques of mass spectrometry and photoelectron spectroscopy resemble in that they are based on the principle
of ionisation of gases under high vacuum, but in the former, positively charged fragments formed when the sample is
bombarded with medium energy electrons (25–70 eV) are sorted out as a function of their kinetic energies while in
the latter, the sample is irradiated with UV (10–45 eV)/x-ray (200–2000 eV) radiations. Milligrams of a sample are
required for PES while for mass spectrometry subnanogram quantities of a sample are sufficient.
Mass spectrometry differs from electromagnetic spectroscopy in that the ions are separated according to their
masses in the former while in the latter, electromagnetic waves are separated according to wavelengths. Mass
spectrometry permits us to determine the molecular mass and molecular formula of a compound as well as certain
structural features. Infrared spectroscopy gives us information about of the types of functional groups present in
the molecule. Ultraviolet/visible spectroscopy imparts information about the conjugation contained in a molecule.
NMR spectroscopy imparts information about the arrangement of functional groups present in the molecule while
electron-spin resonance spectroscopy tells us about the presence of free radicals in the molecule.
Furthermore, mass spectrometry is not a pure spectroscopic technique since chemical reactivity is involved in
the fragmentation, i.e. decomposition process. But it is still considered an analogue of the spectroscopic methods
since it compliments the information provided by spectroscopic techniques, namely, infrared, ultraviolet/visible,
nuclear magnetic resonance, etc.
11.3 VARIOUS TYPES OF IONS ENCOUNTERED IN MASS SPECTROMETRY

When a molecule of mass M, which is in the gaseous state, under conditions of high vacuum (10−7−10−9 mmHg) is
bombarded with medium energy electrons (25–70 eV), the initial processes which can occur are the following:
(a) Parent or Original Molecular Ion As a rule, electron bombardment ejects one of the valence electrons,
i.e. one of the free (unshaired) electron pair of a hetero atom or one of the p-electrons of an aromatic system or a
multiple bond to produce a highly excited positively charged ion (in time of the order of 10−15 second) known as
the parent molecular ion or simply molecular ion. The molecular ion is formed when the energy of the bonding
electrons is equal to the ionisation energy of the molecule and the lifespan of the molecular ions is of the order
of 10−10−10−3 second.
M+ + 2e– Original moleculer ion
M2+ + 3e– Dipositive ion

M + e–
M3+ + 4e– Tripositive ion
M– Negative ion
For molecules containing multi-isotope elements, a number of different isotopologue molecular ions will be
formed. The one containing the most abundant naturally occurring isotopes of all atoms that make up the molecule
is termed the mono-isotopic molecular ion, whereas the others are termed isotopic molecular ions. The most abun-
dant of these isotopologue molecular ions is sometimes termed the top molecular ion. The mass of the molecular
ion is by definition, calculated for the mono-isotopic molecular ion using therefore the masses of the most abundant
isotopes with a correction for the mass of the electron(s) that has been removed. For instance, electron ionisation
of BF3 gives two isotopolouge ions, 10B19 F3+ of m/e : 67 and 11 B 19 F3+ of m/e : 68, the heavier one being the (top)
mono-isotopic molecular ion with a nominal mass of 68 Da since 11B is more abundant than 10B. The lighter ion of
m/e : 67 10B19 F3+ is an isotopic molecular ion. The mass of molecular ion of ethylbromide C2H579Br will be equal
to [29.039 + 79.918 − mass of the electron (me)]. This is equal to 107.957 − me, Da, based on the standard that the
mass of the isotope 12C = 12 Da exactly.
Formation of some typical molecular ions are shown here.
Removal of an electron from hetero atom.
•• •+
CH3•OH
•
+ e– CH3•OH
•
+ 2 e–
Parent molecule Original ion
Removal of p-electron from a multiple bond:

•• •+
C C + e C C + 2e
Parent molecule Moleculer ion
Removal of p-electron from an aromatic system:

•+
+ e + 2e
Parent molecule Original ion

(b) Fragment Ions A molecular ion is labile due to high degree of excitation, undergoes a number of successive fis-
sions with the formation of positively charged ions called fragment ions and neutral molecules or radicals.
The decomposition or fragmentation of the molecular ion usually involves the fission of the bonds near the site
of localisation of the positive charge. Bond fissions that occur without migration of individual atoms or groups
are known as ordinary fissions and the resulting ions are termed as ordinary fragment ions. Such ions are a part
of the original molecule.
In particular cases, fissions can also occur with migration of individual atoms or groups. Such fissions are
termed rearrangement fissions and the fragment ions which result because of such fissions are known as rear-
rangement ions. Hydrogen-transfer rearrangements are most common, though migration of an alkyl group can
also occur. For example, in the mass spectrum of n-butyraldehyde (C4H8O), the parent molecular ion is C4 H8O•+
(m/e : 72). The ordinary fragment ions are C4H7O•+ (m/e : 71), HCO•+ (m/e : 29), C2 H 3O•+ (m/e: 43) and C3H 5O•+
(m/e : 57). The ion C2 H 4 O•+ (m/e : 44) is formed via rearrangement process.
The internal energy Eint of the precursor ion may affect the rate constant, k, for a particular ion decomposi-
tion. Molecular ions containing internal energy < Ea, the activation energy, cannot decompose, irrespective of
the amount of time allowed for decomposition. If Eint >Ea, the probability of decomposition of M •+ in the source
depends on the rate constant, k, ln [M+]0/[M+] = kt. For conventional mass spectrometers, ions whose half-life cor-
responds to the ion source residence time, decompose with rate constants of ≈ 106/second.
(c) Multicharged Ions Dipositive or tripositive ions are also encountered in a mass spectrum. The peak cor-
responding to the doubly charged ion appears at m/e which is one half of that fragment while that of the triply
charged ion appear at m/e which is one third of that fragment, e.g. in the mass spectrum of CH4 there are observed
lines corresponding to the following values of m/e : 16, 15, 14, 13 and 8. The values at m/e : 16 and 8 correspond
to the singly charged ion CH4+ and doubly charged ion CH4++ (m/e :
1
2
1
( )
CH 4 + + = × 16 = 8 ) respectively. The
2
peaks at m/e :15, 14 and 13 are due to the fragment ions CH3+ , CH2+ and CH+ respectively. Mass spectrum of
chrysene (molecular mass = 228) shows molecular ion peak M+ at m/e : 228 (45 mm) and a doubly charged peak
at M2+ at m/e : 114 (8 mm).
(d) Metastable Ions Neutral fragments produced in process of fragmentation, e.g. M+ (g) → m1+ + m2 or m1 +
m2+ cannot be detected in the mass spectrum since they can never qualify the condition kinetic energy = eV (e = 0
for neutral particles). However if an ion of mass m1, called the parent or original ion having a lifetime of the order
10−5−10−6 second decomposes while traversing the field-free region before the magnetic field to another charged
fragment of mass m2 called the product or daughter ion and to a neutral fragment of mass m3= m1− m2, i.e.
+
m 1 → m 2 + + m 3 ( m1 − m 2 );
Parent
a ion Daughter Neutral
ion fragment
the resultant peak neither appears at m1, nor at m2, since the daughter ion m2+ does not qualify the condition of
kinetic energy = eV obeyed by the parent or daughter ion formed in the ion source; and the parent ion m1+ imparts
some of its kinetic energy to the daughter ion m 2+ and to the neutral fragment m3 in the ratio of their masses.
m
Accordingly, the kinetic energy of m 2+ ion formed in the field-free region equals 2 eV which is less than that
m m1
of the normal m 2+ ion by a factor of 2 . Hence the peak due to m 2+ ion formed in the field-free region neither
m1
appears at m1, nor at m2 (i.e. the normal positions for m1+ and m2+ ions on the mass scale in the spectrum) but
appears as a metastable peak (m*), the position of which is given by the relationship
m2 ( m )2
m∗ = × m2 = 2 (11.3)
m1 m1
The peaks due to metastable ions are weaker and broader compared to the normal peaks due to parent and
daughter ions. The relative abundance of metastable peaks is of the order of 10−2 or less compared to normal peaks
because the metastable peaks are given by reactions with rate constants in the region of 104−106 s−1 (contributions
from ion lifespan of about 10−6 s). The rate constant grows from 104−106 s−1 in a very narrow range of energies
given by the expression
S n −6 3 n −5
⎛ ΔE
E⎞
k =n⎜
⎝ Ea ⎟⎠ (11.4)
Here, ΔE = Eint − Ea; Eint is the internal energy of the reacting ion and Ea is the activation energy for decomposi-
tion, and n is the frequency factor and so very few ions have these energies, i.e. ΔE. The broadening of metastable
peaks occurs since some of the exciting energy leading to bond fission may be converted into additional kinetic
energy. Further in a spectrum which is linear with respect to m/e values, metastable peaks lie on the lower side of
m2 normal peak on the mass scale and the magnitude of distance of m* from the daughter ion m2 is similar to that
of the daughter ion below the parent ion. Mass spectrometers in which the mass scan has an exponential function,
m2 will be equidistant between m1 and m*. The metastable peaks usually occur at non-integral values of m/e. Note
that the absence of a metastable peak from the spectrum does not prevent a particular decomposition.
The study of metastable ions is helpful in detecting the mode of fragmentation of a particular compound and
the presence of such peaks suggests that the reaction m1+ → m2+ + m3 takes place in one step, e.g. mass spectrum of
toluene (C7H8) shows strong peaks at m/e : 91, 65 and a weak metastable peak at 46.4. The metastable peak at 46.4
reveals that the reaction m/e : 91 → 65 is occurring. Mass spectrum of tolylphenylether (C13H12O) exhibits a parent
peak at m/e : 184 (10 percent), the base peak at m/e : 91 and small peaks at m/e: 77, 65. Metastable peaks are located
at m/e : 46.5 and 45.0. The presence of two metastable peaks at 46.5 and 45.0 indicates that the reactions m/e : 91 →
65 and 184 → 91 are occurring. Isopropylpheylthioether (C9H12S) exhibits a parent peak at m/e : 152 (45 per cent)
and fragment ion peaks at m/e : 137(7 per cent), 110(100 per cent), 77(7 per cent), 66(11 per cent), 65(8 per cent)
and 43(12 per cent). Metastable peaks are observed at m/e: 123, 79.6 and 54.1. The presence of three metastable
peaks at 123, 79.6 and 54.1 indicates the reactions respectively at m/e : 152 → 137, 152 → 110 and 110 → 77.
(e) Negative Ions Occasionally negatively charged ions or anions are also encountered in a mass spectrum.
They originate from the impact of electron on the molecule by the following processes:
(i) Resonance capture or electron attachment
M + e → M− ; or AB + e → AB−
(ii) Dissociative resonance capture (dissociative electron attachment)
•
AB + e → A + B
(iii) Ion-pair production
AB + e → A + B− + e
The threshold energy of electrons for this process is above 10−15 eV. This process can also be initiated by ionis-
ing radiations such as a-particles or photons. Negative ions are not commonly formed in this process.
All the three processes can occur under normal conditions (70 eV) but each process being pressure dependent,
the relative abundance of ions will vary with the sample pressure. The formation of negative ions can be enhanced
using low-energy electrons (2 to 4 eV) from an argon source. Since high source pressure (about 10−3 torr) is neces-
sary for the production of anions by such processes, ion–molecule reactions are also common under these condi-
tions. The most common ions produced in such processes correspond to M− (resonance capture), M − 1 (loss of
hydrogen), M + 16 and (M − 1) + 16; the latter two ions result due to the presence of oxygen in the ion source.
Negative ions also result in various types of reactions observed during negative-ion chemical ionisation.
(f) Franck–Condon Principle and Stability of Negative Molecular Ions The potential energy diagrams
for the formation of negative ion by the impact of electrons are shown in Figs 11.1 (a)–(d).
The phenomenon of resonance capture may take place in two ways depending on whether the electron affin-
ity of AB is (i) positive [Fig. 11.1(a)] or (ii) negative [Fig. 11.1(b)]. When the electron affinity is negative, a
Franck–Condon transition results in an unstable anion [AB−]* which may have energy either below or above the
dissociation limit. When the energy of the excited anion is below the dissociation limit, it may disappear by the
autodetachment process, i.e. [AB−]* → AB +e− and if the energy of the excited anion is above the dissociation
limit it may dissociate to give a radical and an ion, i.e. [AB−]* → A• + B−
The molecular anions formed by resonance capture being less stable relative to autodetachment, are stabilised
by radiation emission or by collisional deactivation. For significant collisional stabilisation at pressures of about
1 torr (CI pressure), the lifespan of the [AB−] species formed by resonance capture should be >1 ms.
In dissociative electron attachment, the vertical Franck–Condon transition (due to electron capture) produces a
repulsive state of AB− [Fig. 11.1(c)], the dissociation of which gives rise to an atom and ion, with excess internal
and kinetic energies
i.e. AB−→ A + B−
[AB–]*
AB AB– AB
A+B •
A + B–
A + B– A+B
(a) (b)
AB
Positive electron affinity Negative electron affinity
AB+
AB
AB– A+ + B–
A+B
AB
A+B
A + B–
(c) (d)
Fig. 11.1 Potential-energy diagrams for the formation of negative ion by electron bombardment.
The repulsive state of AB− is also possible below point C by the tunnel process.
In ion-pair production, the vertical Franck–Condon transition results in an excited molecule which dissociates
into a positive and a negative ion [Fig. 11.1(d)].
AB + e → [ AB] → A+ + B− + e
∗
Excited
molecule
As already stated, the threshold electron energy for ion-pair production usually lies above 10 to 15 eV.
The specific electron energy required to generate a stable carbanion is determined from the ionisation energy
curve of the system. For p-nitrobenzyl-N-acetylalaine, the specific electron energy curve shows a maximum at
2.9 eV and decreases to zero at about 7 eV. This energy (i.e. 2.9 eV) of electrons for the production of carboxylate
anions from amino acid derivatives is very specific, a variation of 1 eV from 2.9 eV maximum, causes a substan-
tial decrease in ion current, i.e. intensity.
Negative ions are not only produced by the interaction of an electron with a molecule but can also be generated
by negative ion chemical ionisation. The various types of ion-molecular reactions observed during negative-ion
chemical ionisation are:
(i) Electron–molecule Interactions The phenomena involved for the production of negative ions by this pro-
cess has already been discussed (see negative ion in this section).
(ii) Associative Detachment Reactions Such reactions take place between an anion (Z−) and a neutral mol-
ecule (N) to give an electron.
Z− + N → ZN + e (Low energy electron)
Such low-energy electrons are used to produce stable anions.
(iii) Displacement Reactions Such reactions in the gas phase Z− + RY → ZR + Y− are equivalent of the SN2
reactions in solutions; for example the reaction of Cl− with cis- and trans-4-bromocyclohexanol in the gas phase
takes place with inversion of configuration as expected for SN2 displacement reactions.
Displacement reactions occur provided they are exothermic and there is no facile proton transfer process
from RY to Z−, e.g. the reaction between OH− ion and CH3CN takes place by proton abstraction but not by
displacement of CN−, i.e.
OH− + CH3CN → CH3O− + HCN (allowed)
OH− + CH3CN → CH3OH + CN− (not allowed)
(iv) Proton Transfer/Deprotonation Reactions The gas phase acidity scale is defined in terms of ΔHo for the
reaction
AH → A− + H+; ΔHo = ΔHacid
Here, ΔHo represents the proton affinity of A−. Since ΔHo is positive, the acidity is greater, the smaller the
ΔHAcid.
Accordingly, the deprotonation reaction
Z + YH → ZH + Y −
Base Acid Acid Base
will occur or will be exothermic provided acidity of YH is greater than that of ZH in the gas phase or the proton
affinity of Z− is greater than that of Y− in the gas phase.
Some proton-transfer reactions are of importance in negative-ion chemical ionisation mass spectrometry not
only because of their efficiency, but also because of the formation of [M—H]− ions which provide information
about the molecular anion.
(v) Association Reactions Third-order association reactions for anions are as common as for positive ions.
Just like cations, anions in the presence of polar solvents such as water also form solvated species commonly via
third-order mechanism. The magnitude of third-order rate constants for such anionic reactions is similar to that of
similar positive ion reactions including those for solvation reactions.
(vi) Charge-Exchange Reactions Such reactions can take place exothermically as well as endothermically
also. The charge exchange reaction
A + B → B + A
will be exothermic, if the electron affinity of B is greater than that of A and endothermic provided electron affinity
of B is smaller than the electron affinity of A. In addition to exothermic charge-exchange reactions, the study of
endothermic reactions as function of ionic translational energy reveals that the energy threshold for endothermic
reactions can be used to establish the difference in electron affinities of the reactions.
11.4 BASIC PRINCIPLES OF MASS SPECTROMETRY

The technique of mass spectrometry is based on the principle of ionisation of gaseous molecules under vacuum
and segregating the positively charged fragments according to their mass/charge ratio, i.e. m/e groups as a func-
tion of their relative abundance (intensity) using electrostatic field and uniform magnetic field.
11.4.1 Theory
The basic theory which laid the foundation for manufacturing the various types of magnetic deflection instru-
ments to measure the m/e ratio of each of the positively charged fragments as a function of their numbers or
kinetic energy is described here.
When ions of mass m and charge e are made to pass through a static electrostatic field of voltage V, they gain
energy equal to the work done by the electrical forces, i.e. eV. The kinetic energy of an individual ion when it
leaves the electrostatic accelerating field is thus given by
1
mv2 eV (11.5)
2
where, v is the velocity of an ion leaving the accelerating electrostatic field. Algebraic rearrangement of
Eq. (11.5) yields
2eV
v= (11.6)
m
When the accelerated ions with velocity v are subjected to a perpendicularly directed uniform magnetic field
of strength H, they experience a magnetic centripetal force given by Lorentz law
F = evH (11.7)
The magnetic field deflects the ions from their initial direction differently according to their mass to charge
ratio and thus are separated into m/e groups, each of which moves in a path with different radius of curvature r.
Since the force exerted by the magnetic field on the ion is a centripetal force, it must be equal to the centrifugal
force on the ions coming out of the accelerating field and we can readily calculate the radius of the circular path
equating mv2/r and evH, i.e.
mv2/r = evH (11.8)
Thus, the radius of the path is
mv
r= (11.9)
eH
Substituting the value of v from Eq. (11.6) into Eq. (11.9), we get
m 2eV
r= (11.10)
eH m
Algebraic rearrangement yields
m H 2r 2
= (11.11)
e 2V
For a particular value of V and H; r2 ∝ m and thus at the point of focus only the mass m+ collects and those of
higher mass will be above and below it. If we now have a tube of definite radius and a slit at the other end of it
only the masses satisfying the equation (11.10) will enter. However, by varying either the electric field or magnetic
field or both, other masses can be made to come to focus at the same point. In practice, one of them is kept con-
stant and the other is continuously varied such that masses come to focus one after the other like 1, 2, 3, ...n. The
results are recorded suitably. When electric field is kept constant and magnetic field is varied, it is called magnetic
scanning and vice versa. The electrostatic scanning is easier to design but because
1
m∝ ,
V
a wide range of masses cannot be scanned with one setting of the magnetic field. On the other hand, since
m ∝ H2, it is possible to cover masses from 12–500 or 1000 Da in one spectrum.
11.4.2 Basic Principle of a Simple Mass Spectrometer

The schematic diagram of a simple single focussing low-resolution mass spectrometer for recording the mass spectrum
of charged particles is show in Fig. 11.2. The five major component parts of the mass instrument are vacuum system,
inlet sample system, ionisation chamber/reservoir, analyser tube and magnet; and detector and recorder.
Neutral molecules,
Ionisation neutral fragments,
chamber Filament To vacuum
pump and negative ions
Neutral Molecule Variable
molecules magnet field
Sample
inlet Sample
Analyser
reservoir
+ tube
Positively charged ions

Repeller Anode deflected according
plate Negatively Variable
Charged charged magnet field to m/e
ions accelerating Ions exit
and focussing slit
plates
Collector
Amplifier
Chart Oscillo-
Computer
recorder scope
Fig. 11.2 Schematic diagram of a mass spectrometer.
(a) Vacuum System It is an essential part of the

mass spectrometer and it operates at a very low pressure
(10−6−10−7 mmHg). The low pressure is maintained in Expansion
order to avoid collisions, between the charged ions and volume Oven
the air molecules otherwise the ions will get scattered and
will not reach the collector and also secondary reactions
will occur. The high vacuum is achieved by using mercury
diffusion pump in addition to a mechanical pump or an oil Septum
diffusion pump in addition to a mechanical pump. In the Ion source
former, traces of mercury may appear and are sometimes
useful for indexing the mass numbers. While in the latter Molecular
leak
a few peaks due to hydrocarbons are obtained in the blank
which does not affect the performance of the spectrometer
but can be eliminated by the use of mass molecular sieve
trap for hydrocarbons in the source and ionising regions Diffusion
pump
of the spectrometer.
Fig. 11.3 Standard diagram of a sample inlet system.
(b) Inlet Sample System The mode of introduction of the sample into the ionisation chamber depends upon
the method of the production of charged ions. When the ions are to be generated by electron bombardment, the
sample should flow as a gas. For gaseous samples, the inlet sample system usually consists of an evacuated cham-
ber of 1–2 litres capacity at a pressure of about 100–200 mHg (Fig. 11.3). The sample is allowed to diffuse into
the ionisation chamber through a sinter, i.e. tiny orifices; the rate of diffusion must remain constant throughout
so that the relative abundance of masses in the mass spectrum may become of some sense in qualitative analysis
and is discussed in Section 11.4.2.1.
The pressure in the sample chamber is measured and it bears a constant relationship to the pressure in the ion
chamber. The sample chamber can be heated up to 400°C in order to vapourise the volatile liquids and solids. A few
micrograms of high boiling liquids or solids can directly be introduced into the ionisation chamber on the end of
a probe. In such cases, sample vapour pressure of about 10−6 mmHg at the operating temperature of the ion source
would suffice. Many substances in the molecular mass range of 300 to 1200 fall into the latter class. Moreover, in
favourable cases, useful information can be obtained from samples in the ng to pg (l0−9−10−12 g) range. The high
sensitivity of the method is one of its greatest advantages.
Note that these methods are not suitable for highly nonvolatile substances such as metals, oxides and refractory
materials since sufficient vapour cannot be developed to flow the vapour into the ion source. In such a case other
methods of ionisation like surface ionisation, spark discharge, field ionisation, etc., are used and are discussed in
detail in Section 11.6.
(c) Ionisation Chamber Ionisation chamber contains an electron impact ion source and positive ion repel-
ling and an ion-accelerating systems. The ion-repelling system acts as negative ion and electron trap also. The
chamber is maintained at a pressure of 0.005 torr and at a temperature of 200°C. The stream of neutral molecules
entering the ion chamber is ionised by the electron bombardment. Electrons are produced by heated rhenium or
thoriated indium or carbonised tungsten filament and collimated towards an anode by a small magnetic field (of
the order of 100 G) which is confined to the ionisation region only. At the time of collimation, the electrons are
subjected to a controllable electric field varying from 5–75 V.
These energetic electrons intercept the neutral molecular stream at right angles to their flow. A range from 5–14 V
is used for molecular mass determination and to study isotopically labelled compounds. The impact electrons eject
one (or occasionally two) electrons from the outermost shells. Therefore for a molecule M, we have
M + e → M +• + 2e
or occasionally
M + e → M 2 +•• + 3e
The electron may also be captured by the neutral molecule to produce a negative radical ion; the formation
probability of which is 10−2−10−3 as compared to that of M +• radical cations
M + e → M −•
When the electrons are highly energetic (50–75 eV), the fragmentation of the molecular ion (M+) results
in fragment ions of varying masses. The positively charged ions (parent or fragment or both) so produced are
repelled by the repelling electrode (positive) placed just behind the ionisation region towards the accelerating
region where they are subjected to electrostatic field of several kilovolts (400–8000 V) of energy. The entire
assembly consisting of ion source, ion-repelling electrode and ion accelerating system is known as ion gun. The
ions of masses m1, m2, m3, ..., etc., are accelerated in the accelerating region to their final velocities. The ions
leaving the accelerating region will have velocities and kinetic energies given by
1 1 1
eV = mv 12 = mv 22 = m v32 = ... (11.12)
2 2 2
(d) Mass Analyser The heart of any mass spectrometer is its analyser. The analyser tube is a semicircular
tube of 1−2 cm (internal diameter) and made of nonmagnetic material like copper or inconel and highly polished
inside. The ion beam from the ionisation chamber passes from the ion source to the collector. The ion path diam-
eter is about 15−20 cm. The entire ionisation chamber, analyser tube and collector are housed between the two
poles of a powerful magnet (4000−6000 G). The uniform and variable magnetic field is applied perpendicular
to the direction of the ion path. An ion takes about 1–5 ms to leave an electron impact source after its formation
and about 1 ms to traverse the field free region in front of the magnetic analyser. For fixed values of accelerating
electric field and the magnetic field, only ions of a particular mass reach the ion detector of the instrument, other
ions collide with the walls of the tube and collapse. If a fragment’s path matches the curvature of the analyser
tube, the fragment will pass through the tube and out the ion exit slit.
(e) Ion Detector and Recorder The ions passing through the slit at the other end of the analyser tube are
often collected either on a collector or on a photographic plate. The ion collector is linked through amplifying
circuits to a recorder. This is, in turn, coupled with a variable magnetic field, and as the strength of this field is
gradually increased, ions of increasing mass enter the collector. A collector records the relative number of frag-
ments with a particular m/e passing through the slit. The more stable the fragment, the more likely it will make
to the collector. By slowly increasing the strength of the magnetic field, fragments with progressively larger
values of m/e can be guided through the tube and comeout through the exit slit. Usually, for high-resolution
works, the ions are collected on an electrode called the electron multiplier or dynode. The ion beam strikes the
surface of a dynode of a multistage Be–Cu electron amplifier (12–20 dynodes) as shown in Fig. 11.4. The col-
lision releases a number of electrons from this surface. The electrons are accelerated to a second such surface.
The latter surface generates several electrons each time a single electron impinges on the surface. The process
is continued until a cascade of electrons arrives at the collector. In this way, a single fragment ion can produce
hundreds of electrons. Since the charge on one ion is 1.6 × 10−19 coulomb, the gain of the multiplier can be cal-
culated. A typical single spike might be 10−10 ampere in height and 10−3 second in width, so that charge equals
10 −13
10−10 × 10−3 = 10−13 coulomb. The multiplier gain is thus = 6.3 × 105. The ion beam currents which
1 6 × 10 −19
are of the order of 10−15 −10−5. A after amplification by the above technique are fed into a recorder and the
results are displayed as peaks. The heights of the peaks (which are proportional to peak areas) are proportional
to the number of ions (i.e. relative abundances) of each mass. For a doubly charged ion of mass M, the peak
will appear at a mass value of M/2. Thus the mass spectrum is a plot of number of ions of each mass against
their mass to charge ratios.
Ion beam
Each electron causes several
electron to be emitted. The
serveral paths are illustrated
at the surface Cascade of electrons
arriving at collector
Amplifier
Electrons
Electron emissive surface.

Several electrons emitted
Fig. 11.4 Schematic diagram of dynode amplifier.
The mass spectrum in the bar graph form or in the peak form is Base peak
displayed on an oscilloscope or recorded on a paper chart as shown in 100 percent
100
Fig. 11.5. The peak of highest intensity is called the base peak and is
assigned a value of 100 per cent. The intensities (height × sensitivity Peak
Percent of base peak

factor) of other peaks, including molecular ion peak, are expressed as 73 percent
percentages of the base peak.
Modern mass spectrometers are coupled with computers for storing Molecular
ion peak
and analysing data. When the peaks at the higher end of the spectrum 38 percent
are weak and the background trace is indistinct, a calibrant which gives
a large number of peaks is used. Perfluorokerosene is often used for
indexing the mass numbers.
The utility of the mass spectrometer depends upon one of its most Bar
important parameters, called the resolving power. The greater the 0
resolving power of the mass spectrometer, the better is its utility as an m/e
analytical tool. Sensitivity is the other most important parameter that is Fig. 11.5 Hypothetical mass spectrum of a
molecule. Here in mass spectrometry ‘e’ is not
used in the evaluation of mass spectral performance. the electronic charge, i.e. 0.16 × 10−18 C, but ‘e’
(i) Rate of Sample Flow to the Mass Spectrometer Since represents the charge on the ion. To avoid this
the molecular flow rate is inversely proportional to the square root confusion, after 1980, m/z is being used in place
of m/e, where z is the charge on an ion.
of molecular mass, the molecular conductance will be lower for
higher molecular compounds. The leak conductance for normal operation should be in the range 0.1−0.4 cc/s.
p
Suppose the molecular conductance (C) of a peak is specified as 0.4 cc/min for N2 at 25°C. For molecu-
lar mass of 200, the conductance will be 0.4 cc/min × 28 / 200 = 0.15 cc/min at 25°C. Since the effusion
rate is proportional to the square root of the absolute temperature (Cor = A RT / 2π M lit/s if T is in K and
R is in ergs/K/mole, Cor is the orifice conductance), then at 150°C (423 K), the conductance for a molecu-
lar mass of 200 will be 0.15 cc/min × 423 / 298 = 0.21 cc/min. This value can be used to determine the
rate of sample flow to the mass spectrometer. If 2 × 10−4 g of ethyl decanoate is injected into a 1-litre inlet
at 150°C, the inlet pressure, as calculated from PV = nRT comes out to be 2.6 × 10−2 torr (1 Torrecelli = 1
mmHg pressure).
The rate of sample flow or effusion to the mass spectrometer can be calculated from a variation of gas formula
in which the static volume V is replaced by the volume flow rate C. Thus
PC
n(moles/s) = (11.13)
RT
where, P = pressure in torr, C = conductance, lit/s, R = gas constant, 62.3 lit torr/mole K, T = absolute
temperature.
For P = 2.6 × 10−2 torr, C = 0.21 cc/min, and T = 423 K, the number of moles of ethyl decanoate (molecular
mass = 200) effusing to the mass spectrometer per second will be 2.4 × 10−10 mole/s or 4.8 × 10−8 g/s. This amount
of sample will give an intense mass spectral pattern in most mass spectrometers operating at a resolution of
500–1000. For vacuum calculations, the unit of litre torr/s is often preferred.
Accordingly, Eq. (11.13) then reduces to
Q = PC. (11.14)
at constant temperature. Thus, for the above example Q = 5.5 × 10−6 lit torr/s.
If the conductance of the peak is not known, it can be determined by measuring, the decrease in the intensity
of the spectral patterns as a function of time. As sample effuses to the mass spectrometer, the sample pressure in
the inlet system decreases exponentially according to the equation
ct
−
P Po e V
(11.15)
where, Po = initial pressure, t = 0; C = conductance, cc/s; V = volume, cc; t = time.

By algebraic rearrangement,
V
C= ln( Po / P ) (11.16)
t
The volume V includes both the 1-litre expansion volume and the manyfold volume of about 100 cc.
If in the above example, the mass 88 peak is observed to change from an initial peak height of 8200 mm deflec-
tion to 5810 mm after 30 minutes, the conductance for a molecular mass of 200 at 150°C will be
1100(cc ) 8200
C= ln = 0 21cc/s
30 × 60(s) 5810
When the many fold volume is known with sufficient accuracy (5−10 per cent), the expansion volume can
be closed off to decrease the time required to attain the same reduction in mass peak intensity or pressure. This
method of measuring and calculating sample consumption is the most convenient method of obtaining the mass
spectrometer sensitivity.
(ii) Resolving Power or Resolution of a Mass Spectrometer The resolving power, Rp, of a mass spectrom-
M1
eter is arbitrarily defined as the ratio
M 2 M1
M1
Rp = (11.17)
M 2 M1
When two peaks at masses of M1 and M2 of equal intensities are

just separated in the mass spectrum and the difference in mass num- ΔM
Relative intensity
bers gives a valley of 10 per cent between peaks of masses M1 and 100
M2 (Fig. 11.6) it is considered to be satisfactory if ΔM = M2 — M1
is not greater than unity. Thus a resolution of four hundred distin-
guishes between two peaks with m/e of 401 and 400. To resolve a
mass 238 doublet from the parent peaks of 1-methoxy-anthraqui-
tione (A) and 1,4-diaminoanthraquinone (B) taken on double focus 10 10 percent
mass spectrometer, the resolving power of the instrument should
M1 M2
be
Mass number
Fig. 11.6 Schematic representation of 10 per cent valley
238.1387 238
38.1387
387 resolution in mass spectrometry.
= ≥ 21262
238.1499 − 238.1387 0.0112
O OCH3 O NH2
O O NH2
(A) m/e: 238.1387 (B) m/e: 238.1499
The resolution of a mass spectrometer depends upon width of the ion beam as determined by slits, distribution
of kinetic energies in the beam, space change of ion beam, variations in magnetic field, variations in accelerating
voltage, poorly collimated ion beam.
The CH2—N doublet (m/e: 14.0156, 14.0031) at mass 14 will be adequately resolved if the resolving power of
14.0031
instrument is ≥ ≥ 1120.
0.0125
(iii) Mass Spectrometer Sensitivity It is the most important parameter used in the evaluation of mass spectral
performance. It should describe the output ion current received at the collector for a specified rate of sample con-
sumption. The definition should state (i) the chemical substance, (ii) the mass peak that is measured, (iii) the mass
spectrometer resolution, (iv) the quantity of the sample used per second, (v) the electron beam intensity, and (vi) the
ion current arriving at the conversion dynode of the electron amplifier. Practical units would be amps of ion beam/
nanogram/second/ampere of electron current at the resolution Rp. The compound used to establish the basic sensitivity
should qualify certain criteria, namely (i) it should be highly stable, (ii) it should be easily available, (iii) preferably it
should be a liquid to facilitate handling and have sufficient volatility to permit introduction via a 150°C inlet, (iv) a
fairly high molecular mass in the 200–400 Da range is preferred, and (v) the mass spectrum should show a significant
molecular ion (5–30 per cent of base peak).

The compound chosen for sensitivity determination should be such that it may thermally decompose prior
to electron bombardment. Cholesterol is a common example. Some manufacturers specified result with methyl
O
stearate (CH3 (CH2)16 C O CH3). Measurement of the base peak (e.g. mass 74 in methyl ester spectrum)
gives the most uniform sensitivity comparison between different instruments. Ideally, the base peak should be
used to specify the basic instrument sensitivity and the ratio, base peak , should be given to indicate instrument
parent peak
variations. The term amp/torr is particularly useful for residual gas analysis. In essence, there are two aspects of
mass spectrometry, that must be considered in defining sensitivity. One is the basic instrument sensitivity that
gives the amount of current arriving at the collector for a specified sample consumption rate. The second is the
method of handling the sample which thus determines the efficiency of sample utilisation, this aspect is a chemi-
cal problem dealing with the mass transfer of small quantities of chemicals.
11.5 VARIOUS FORMS OF MASS SPECTROMETRY ON THE BASIS

OF METHODS OF SEPARATION OF IONS
Depending on the method of separation of charged ions produced by electron bombardment, according to
their mass-to-charge ratios in the mass spectrometer, spectrometry has broadly been divided into the following
classes:
11.5.1 Magnetic Field Deflection Mass Spectrometry

(a) Single Focussing Magnetic Field Deflection Mass Spectrometry (SFMDMS) In a single focussing
magnetic deflection mass spectrometry, the positive 180˚ magnetic sector
ions are separated by uniform and variable magnetic
field according to their mass to charge ratios. The first
electron impact ionisation and single focussing mag-
netic deflection mass spectrometer was developed by
A J Dempster in 1918. Such type of electron impact
sources (i.e. electron from a hot-wire filament) are still
very widely used in modern spectrometers. The sche-
Ion beam
matic diagram of a 180° single focussing Dempster
mass spectrometer is show in Fig. 11.7; the theory and
working of which has already been explained in sec-
Detector
tions 11.4.1 and 11.4.2 respectively. Ion source
(b) Double Focussing (Electrostatic and Mag- Fig. 11.7 Schematic diagram of a single focussing mass spectrometer.
netic Fields) Mass Spectrometry (DFMS) The basis of DFMS is that the ions are first sorted out according
to their energies by applying radial electric field perpendicular to the path of the ions and subsequently according
to their mass-to-charge ratios by applying magnetic field again perpendicular to the path of the ions.
The ions at the time of generation have different energies (because of Boltzmann kinetic energy distribution)
which would affect resolution. However, in a magnetic single focussing spectrometry, it is assumed that all the
ions have same kinetic energy and the ions are separated according to their mass to charge ratios. Due to this
assumption, the magnetic single focussing spectrometers have low resolution generally between 200–1000. This
ΔV
can be overcome by employing a very high accelerating potential so that becomes very small. But a better
V
way would be to apply a radial electrostatic field, i.e. the voltage is applied across the plates of a cylindrical con-
denser. When the ions pass through such a radial field, the centripetal force of the electric field is balanced by the
centrifugal force of the ion, i.e.
mv2 mv2
= eV or r = (11.18)
r eV
Thus, for a fixed value of V, the ions of same energy will follow the circular path of radius r. The radius of
curvature of ions deflected in an electrostatic field is given by
2V (11.19)
r=
V
where, V is in volts/cm. Thus ions with energy V + ΔV and V − ΔV will be separated.
Subsequently, magnetic field is applied on such ions to achieve double focussing. In a magnetic field analyser,
ions are separated by differences in their m/e ratios as they move in a circular path towards a collector in order of
increasing mass to charge ratio, i.e.
m H 2r 2
= (Eq. 11.11)
e 2V
where the terms and symbols have their usual meanings.
High mass resolution double-focussing spectrometers were developed for the purpose of determining the exact
atomic masses of the elements. One type of double focussing instrument was developed in 1930s by Josef Mattuch and
Richard F F Herzog and another was developed by E G Johnson and A O Nier in 1953. Both the mass spectrometers are
based on the same principle but differ in their magnetic and electric field geometries. The block diagram for the former
is shown in Fig. 11.8 and that of the latter in Fig. 11.9.
(c) Cycloidal Focussing Mass Spectrometry In this type of mass spectrometry, the ions are passed
through a homogeneous electric field at right angles to and superimposed on a magnetic field. The ions passing
Ion
source Electrostatic
+ field
Detector
32˚50
´ Magnetic
field
Fig. 11.8 Mattauch–Herzog geometry for double-focussing mass spectrometer.
Electrostatic
field Magnetic field
+
Ion
beam
–
90˚
Detector
Source
Fig. 11.9 Nier–Johnson geometry for double focussing mass spectrometer.
Medium energy
High energy
Low energy
Recorder
Ion source Amplifier

Collector
Fig. 11.10 Schematic diagram of a cycloidal mass spectrometer.
through such crossed magnetic and electric fields generate a cycloidal path as shown in Fig. 11.10. As either
of the two crossed fields is varied, ions of identical m/e ratio focus on the fixed collector slit even though not
of equivalent energy. For ions focussed in the cycloidal analyser,
kH 2
m /e = (11.20)
2πV
where k is a constant. Here, the small radius of curvature allows the use of a smaller magnet without sacrificing
resolution. The resolution is at least 1000 in the mass range extending from 10 to 2000.
11.5.2 Ion, Resonant/Cyclotron Resonance, Mass Spectrometry (IRMS/ICRMS)

ICRMS was developed by J A Hipple, H Sommer and H A Thomas in 1949 and it made inroads into chemistry in
the middle to late 60s’. The basic principle of ICRMS is that when ions are subjected to simultaneous mutually
perpendicular magnetic and radio frequency fields, they follow spiral paths. The ions are collected sequentially
either by scanning the radio frequency of the electric field or by magnetic field. This technique is particularly
useful for the characterization of ion–molecule reactions. In 1974, Melvin B Comisarow and Alan G Marshall
revolutionised ICR by developing Fourier transform ICR mass spectrometry (FTICRMS). The major advantage
of FTICRMS is that it allows many different ions to be determined at once rather than one at a time. This tech-
nique has also the higher mass resolution than that of any other type of mass spectrometric technique.
The block diagram of a typical ICR mass spectrometer often called the Omegatron is shown in Fig. 11.11.
When only magnetic field is operative, the centripetal force of the magnetic field is balanced by the centrifugal
force acting on the ion, i.e.
Magnetic field parallel to

electron beam
Path of
electron-beam Electron
trap
Filament +
rf Plates
Path of Ions-
at resonance Ion collector
Ion
analyser
Fig. 11.11 A schematic diagram of an omegatron.

mv 2
= Hev (Eq. 11.8)
r
v
or = H (e / m ) = w = 2pvv (11.21)
r
v ω
where, ω = is the angular velocity of the ion and v = is the frequency of revolution in Hz.
r 2π
Rearrangement of Eq. (11.21) yields
H
m /e = (11.22)
2pn
Thus, the magnetic field causes the ion to move in a circular path with angular velocity w and frequency v.
When both the mutually perpendicular fields simultaneously operate on the ions, they follow the spiral path at
right angles to both the fields. The frequency v is now called the cyclotron frequency of the ion and is expressed
as the number of complete spirals/cycloids per second and is given by
⎛e⎞ H
vc = ⎜ ⎟ , (11.23)
⎝ m ⎠ 2π
H
or vc 9.16 × 103 (e / m )
19 ,
2π
when m is in amu, and H is in gauss, the cyclotron frequency will be in Hz which is independent of radius r and
speed v. When the radio frequency of the rapidly applied oscillating electric field matches with that of the cyclotron
frequency of ions, the ions will gain energy, consequently only ions of an m/e ratio in resonance with the oscillating
radio frequency field will be successively accelerated and follow the spiral path of ever increasing radius until they
strike the ion collector. The maximum kinetic energy of the ions when they leave the cyclotron is given by
1 1 e 2 2 2
Kinetic energy = 2
m v max = ( ) H rmax (11.24)
2 2 m
Ions with other m/e ratios, i.e. nonresonant ions adopt smaller circular paths near the axis of the electron impact
beam. The resonant frequencies for ions over a mass range of 1–200 amu with magnetic field of 15000 G range
from 50 kHz to about 25 MHz. The energy that is absorbed by the ions is measured by an oscillator detector rather
than collecting the ions in an ion collector. A mass spectrum is thus obtained by plotting the energy absorbed by
the ions against either the rate of scan of the magnetic field at constant radio frequency of the electric field or the
rate of change of the electric field frequency at constant magnetic field.
The mass resolution of an Omegatron is 700 over a mass range of 1–280 amu and is 50 over a mass range
of 2–50 amu.
When a positive voltage is applied to the side plates of an IRMS, the positive ions motion down to the
centre of the analyser and are said to be trapped while the negative ions and scattered electrons are expelled
from the chamber. For trapping of negative ions, the trapping potential is simply reversed. Since the drift
velocities of ions are of the order of 500 cm/s, the residing time of ions is over a range of 0.5–25 ms, and
path lengths for thermal velocity ions are over a range of 1–50 m. This is 1000 times longer than in a sector
mass spectrometer. This very feature of the IRMS makes it useful for the study of ion–molecule reactions
at low pressures. It is also used to measure the density of molecular nitrogen in the upper atmosphere via
rocket flights.
Problem 11.1: The ion resonant mass spectrum of molecular ion of O2 shows a peak at m/e: 32. Calculate the
cyclotron frequency of the molecular ion if the strength of the applied magnetic field is 15000 G.
Solution Substituting the data given in the problem into Eq. (11.23), we get
⎛ 1 ⎞ 15000
v c = 19.16 × 103 ⎜ ⎟ × Hz
⎝ 32 ⎠ 2 × 3.14
= 1.43 × 106 Hz = 1.43 MHz

11.5.3 Time of Flight Mass Spectrometry (TOFMS)

The concepts of TOFMS were proposed in 1946 by William E Stephens. This technique does not use the magnetic
field at all. In this technique, the original ions are separated into ‘wafers’ of ions by the difference in their veloci-
ties as they move in a straight path in a field-free region towards a collector according to their masses.
The block diagram of a TOFMS for recording the mass spectrum of a sample is shown in the self-explanatory
Fig. 11.12. Ions are produced in the ionisation region by one ms pulse of the bombarding electrons having
preselected energy in the 0–100 eV range, typically 70 eV. The ions so produced are accelerated by an electric
potential. The ion focus grid is subjected to an accelerating potential of 100 V in the form of a voltage pulse of
negative charge having repeating frequency of few thousands/second; the lifespan of voltage pulse is 1 ms or less.
This positive pulse accelerates the ions through the grid where they are picked up by the field of the ion energy
grid which is positioned at a distance of less than 2 cm from the ion pulse grid. After successive accelerations,
all the ions acquire a drift energy of 2700 eV (or 2800 eV ) in a distance of less than 2 cm and hence at the ion
energy grid. These ions then move in a straight path of length 100 cm (= 1 m) in a field-free region towards a
magnetic electron multiplier detector which collects and amplifies the ion signals. Although the energy of all the
ions is the same, they drift through the field-free region at different velocities ⎡⎣ v = 2 eV/m
V ⎤⎦ and thus arrive at
the detector end one after another according to their masses. The arrival time between two successive ions can
be ≤ 10−7 s. The mass spectrum is displayed on an oscilloscope or recorded on a paper chart. Modern TOFMSs
are coupled with computers for storing and analysing data.
The TOF, t(in microseconds) of ions through a distance L (in cm) is given by the expression
⎛ m⎞ ⎛ 1 ⎞
t L ⎜ ⎟⎜ (11.25)
⎝ e ⎠ ⎝ 2V ⎟⎠
Here, V is in volts. A typical flight time for a one-metre drift tube is in the rage of 1–30 ms.
To vacuun
pump
Filament Potential selector

0 to 100 V
grids
Ionisation electron beam 100 to 2800 eV
(continous or pulse) Oscilloscope
1 1
Molecule
Sample 2 Magnetic
Chart
in electron
2cm 3 recorder
3 multiplier
Repeller Moleculer
electrode ion beam Computer
+
Ion
Electron Field free/drift region
acceleration
trap (ions are separated in
region
time by their m/e values)
Fig. 11.12 A schematic diagram of a time of flight spectrometer.
The difference in the flight time for ions of different masses can be calculated using Eq. (11.25) in the form
Δt = ( − ) = L2 / 2eV ( m 2 − m1 ) (11.26)
In this method, ions cannot be continuously produced; otherwise there will be mix up of masses. Hence, ion
bunches are pulsed. The analysis is very rapid and an entire mass over a range of 0–1000 can be scanned in 50 ms;
the frequency of repetition of spectrum may be 10,000–50,000 times/second; but 20,000 is commonly used. The
resolution is about 1 part in 400.
This technique is very useful for isotope ratio studies, kinetic studies of fast reactions, direct analysis of efflu-
ent peaks from a gas chromatography, and analysis of samples volatilised by a single pulse from a laser. Time of
flight devices are now largely used with sophisticated ionising methods such as FAB, etc., for high mass measure-
ments including large biomolecules.
Problem 11.2: In a TOFMS of a mixture of CO2 and CH4 there are observed lines corresponding to the
following values of m/e : 8, 13, 14, 15, 16, 28 and 44. What transit times might these lines correspond to if
the drift path length is 100 cm and the accelerating voltage is 2800 V?
Solution In order to determine the TOFs’ for the ions given in the problem, we proceed as follows:
The TOF of an ion is given by
⎛ m⎞ ⎛ 1 ⎞
t L ⎜ ⎟⎜ (11.25)
⎝ e ⎠ ⎝ 2V ⎟⎠
or = k (m / e ) k = L (1/2V ); (11.26)
Since the quantities L and V are constants in the problem, the value of parameter k needs be determined first.
1
Thus, k = 100 = 100 1.7857 × 10 −4 = 100 × 1.336 × 10−2 = 1.336
2 × 2800
and Eq. (11.25) reduces to
t 1.336 m / e
Substituting the respective values of m/e into the above equation, we obtain
t(m/e : 8) = 1.336 8 ms = 1.336 × 2.828 ms = 3.78 ms
Proceeding as above, we obtain

t(m/e : 13) = 4.81 ms; t(m/e : 14) = 4.99 ms, t(m/e : 15) = 5.17 ms;
t(mle : 16) = 5.34 ms, t(m/e : 28) = 7.06 ms and t(m/e : 44) = 8.86 ms,
11.5.4 Radio Frequency Mass Spectrometry (RMS)

In radio frequency mass spectrometry, the ions can be resolved according to their m/e ratio without the need of a
heavy magnet. When electrostatically accelerated ions with uniform energy are subjected to simultaneous radio
frequency field and electrostatic field of potential, V, they are sorted out according to their velocities provided
v = sv (11.27)
where v is the velocity of ions, v is the frequency of the radio frequency field (in the order of 10–100 MHz) and
s is the spacing between the adjacent grids.
Also we know that
1 2 1 2 1 2
eV = mv1 mv2 mv3 (Eq. 11.12)
2 2 2
m 2V
= (11.28)
e s 2 v2
m 0.266 V
or in cgs units = 2 2 (11.29)
e s v
Thus, the mass spectrum can be scanned by varying the radio frequency (v) while keeping the potential V of
the electrostatic field constant.
Based on this principle, one type of radio frequency mass spectrometer was developed by W H Bennet in 1950,
the block diagram of which has been shown in Fig. 11.13.
Problem 11.3: What is the radio frequency range of a radio frequency mass spectrometer analyser for a singly
charged nitrogen molecule if the accelerating voltage varies from 400–4000 volts and the spacing between the
adjacent grids in the analyser is 1 cm?
Accelerating
Energy detector
focussing
grids
electrodes
rf Ion
repeller
Filament
Sample
in
Ion
collector Recorder
Molecule
+
Electron
trap
Molecular
ion-beam
Radio frequency
analyser
Fig. 11.13 A schematic diagram of a radio frequency mass spectrometer. All grids marked x are at the same potential.
Solution Algebraic rearrangement of Eq. (11.29) yields
1
v= 0.266V (e / m ) MHz
s
Substituting the given data into this expression we get
0.266 × 400
v( )= MHz = 4.45 MHz
28
0.266 × 4000
and v( )= MHz = 6.16 MHz
28
Thus, the radio frequency range for the accelerating voltage over the range of 400–4000 volts is from
4.45–6.16 MHz.
Problem 11.4: The radio frequency mass spectrum of CO2 shows a peak at m/e: 44. Calculate the accelerating
voltage of the analyser if its radio frequency is 5 MHz and the spacing between the adjacent grids in the analyser
is 1 cm.
Solution Algebraic rearrangement of Eq. (11.28) gives
1
V ( m /e)(s ) 2 volts
2
Substituting the given data into the rearranged expression, we obtain
1
V = ( 44)(1 × 5) 2 volts = 50 volts
2
11.5.5 Quadrupole Mass Spectrometry (QPMS)

QPMS was first developed in the mid 1950s’ by the group of Professor Wolfgang Paul of the University
of Bonn (where once Kekule and Hertz also made great discoveries) in Germany and is of two types,
namely quadrupole mass filter-mass spectrometry (QPMF-MS) and quadrupole ion trap-mass spectrometry
(QPIT-MS). In the former, a two dimensional quadrupolar fields generated by both a direct current (dc)
voltage (Vdc) and a radio frequency (rf) voltage (Vrf) are used to filter and mass analyse the ions while in the
latter a three dimensional quadrupolar radio frequency field is used to trap and mass analyse the ions in the
same place. Just like time of flight, and radio frequency mass spectrometric techniques and unlike single
and double-focussing mass spectrometry there is no need of heavy magnet to sort out ions according to their
mass to charge ratios.
(a) Quadrupole Mass Filter Spectrometry (QPMF-MS) In this technique, a two-dimensional

quadrupole electric field comprising both radio frequency and direct current components are used to filter
out ions according to their m/e ratios. Mass selection is achieved by varying each of the radio frequency and
direct current frequency while keeping their ratios accurately constant. Although the techniques are not as
accurate as double-focussing mass spectrometric techniques, but they being faster are widely used as GC and
LC-detectors these days.
The heart of 2-D quadrupole mass filter mass spectrometer is the quadrupole mass filter, the block diagram
of which is shown in Fig. 11.14(a) and that of quadrupole mass spectrometer in Fig. 11.14(b). It consists of four
metallic rods. Typical dimensions of the quartz rods coated with platinum in the mass quadrupole are about 12
mm diameter and 25 cm long, operating with 3 MHz rf and 10 V dc. The diagonally opposite rods are joined
together electrically, the two poles to opposite poles of a dc source and simultaneously to a radio frequency oscil-
lator. These voltages create a hyperbolic field between the rods for any specified combination of radio frequency
and dc voltages.
In such a geometry of the quadrupole, the interaction between both a direct current voltage and a radio fre-
quency voltage causes complex oscillations in the X- and Y-directions but a stable oscillation in the direction
which is perpendicular to the plane of the paper. The stable oscillations sort out the ions according to their mass
to charge ratios as they shoot down with constant velocities from one end of the quadrupole to the other end (i.e.
along Z-axis) without striking the poles. All the other ions experiencing complex oscillations along the X- and
Y- directions will strike the poles and collapse. Mass spectrum is obtained by varying each of the radio frequency
and direct frequency while keeping their ratios constant. The scanning time varies from 500 ms to 30 minutes. The
filtered out ions are amplified and the spectrum is displayed either on an oscilloscope or paper chart. The mass
spectral data can also be stored in a digital computer for analysis. The mass resolution is 500 over a mass range of
1–500 and 100 over a mass range of 1–120 amu.
Smaller quadrupole mass spectrometers are used as residual gas analysers and simpler monopole mass
spectrometers employ only a single rod.
(b) Quadrupole Ion Trap-Mass Spectrometry (QPIT-MS) In this technique, a three dimensional rotationally
symmetric quadrupole radio frequency field is used to trap and mass analyse the ions in the same place. The origina-
Y
Z (Ion axis)
X
Vrf + Vdc
Vrf + Vdc
(a)
+
– – End view Collector
+
Anode
Ion beam Rod system

Filament Chart recorder
(b)
Fig. 11.14 A schematic diagram of the (a) geometry of the rods/quadrupole mass filter in a quadrupole mass spectrometer, and
(b) quadrupole mass spectrometer. The Z-axis is perpendicular to the plane of the paper.
Filament
Gate electrode
+ +
+ +
+
+ +
+ +
GC effluent
Magnet
electron
multiplier
Ion
analyser
Fig. 11.15 A schematic diagram of an ion trap mass spectrometer.
tor of QP-MS, Professor Wolfgang Paul, shared one half of the 1989 Nobel Prize in physics with Hans G Dehmelt
for the development of ion trap technique and the other half was awarded to Norman F Ramsey for the invention of
the separated oscillatory fields method and its use in the hydrogen maser and other atomic clocks.
The commercial version of quadrupole ion trap mass spectrometer based on the design of George C Stafford
and co-workers was introduced in 1983 as a GC detector, the block diagram of which is shown in Fig. 11.15. The
heart of this spectrometer is the quadrupole ion trap that consists of three electrodes connected together to a radio
frequency circuit. One electrode, known as the gate electrode, is a doughnut-shaped piece of metal that horizon-
tally encircles the chamber while the other two are hemispherical caps on the top and the bottom of the chamber.
These electrodes generate a three-dimensional rotationally symmetric field to trap and analyse the ions according
to their masses in the same place. Mass scanning is carried out by varying the radio frequency voltage, ions of
increasing mass-to-charge ratios successively become unstable as the radio frequency voltage is increased. Such a
type of mass scanning is called the ‘Staffords’ mass-selective instability scanning. The separated ions according to
their m/e ratios are fed to electron multiplier, i.e. an ion signal collector and amplifier. The spectrum is displayed
either on an oscilloscope or a paper chart. The mass spectral data can also be stored in a digital computer for
analysis. The scanning times of mass spectra of ions over a mass range of 20–650 amu varies from one tenth of a
second to few second. Although QIT-MS at the embroynic stage were introduced as GC-detectors, but these days,
they are not only used as GC-detectors but also as LC-MS detectors and stand alone as mass spectrometers.
11.6 VARIOUS FORMS OF MASS SPECTROMETRY ON THE BASIS

OF IONISATION PROCESSES OTHER THAN EIMS
Many times the molecules do not show molecular ions in electron impact mass spectrometry because of which
it becomes impossible to determine the exact molecular masses of such systems by this technique. Also many
nonvolatile compounds such as amino acids, peptides, proteins, sugars, etc., cannot be introduced as such into
the mass spectrometer. They have to be converted into some reasonably volatile derivatives. The other major
drawback of EIMS is that the electron bombardment destroys the small bio-molecules. Such problems encoun-
tered in EIMS have been solved to a large extent with the emergence of new processes of ionisations such as
chemical ionisation, desorption ionisation, i.e. fast atom bombardment ionisation; electrospray ionisation, etc.
Thus, just like EIMS, other forms of mass spectrometry have also been named after the ionisation processes.
11.6.1 Chemical Ionisation Mass Spectrometry (CIMS)

Chemical ionisation is a bimolecular process in which ionisation occurs as a result of ion-molecule reactions.
It was first observed in 1913 by J J Thomson in hydrogen gas but he did not characterise the process at that
time. CIMS was first characterised in the mid 1960s by Frank H Field and M S Burnby Munson. CIMS is a soft
ionisation technique in which volatilised analyte molecules are ionised by reaction with the secondary ions of the
reagent gas such as methane, isobutane, ammonia, etc., which is ionised by the impact of electrons. The sample is
introduced at about one torr pressure with a carrier gas. The molecular ions produced by the CI process appear at
m/e : M + 1 and MH+ ions and are known as quasi-molecular ions and often appear as prominent or base peak in
the mass spectrum. They possess less internal energy compared to the molecular ions produced by EI and hence
undergo less fragmentation, i.e. generate fewer fragment ions compared to EIMS. Moreover, the nature of the CI
process is such that the product ions formed contain even number of electrons.
Proton-releasing tendency of the secondary ions of the carrier gases follows the order: CH5+ < C4H9+ < NH4+. Thus
by the choice of the carrier gas, we can control the tendency of CI produced MH+ ion to yield fragment ions. NH4+
does not protonate same molecules and if it protonates, the resultant MH+ ion produced will have insufficient amount
of internal energy for fragmentation. Consequently, MH+ ion will appear as a base or prominent peak in the resultant
mass spectrum. Thus, ammonia CI is an excellent technique for the determination of molecular masses of volatile
compounds. The mechanism of production of MH+ ions when methane is used as a carrier gas is given here.
4 + (300 eV)
V)) C +•
CH 4 + 2e ⎫
⎪
CH 4+• → CH 3+ + H • ⎪
⎪
⎬ Primary ion production process
CH+•
4 → CH 2+ + 2H • ⎪
⎪
CH+•
4 → C+ + 4 H • ⎪⎭
CH +4 • + CH 4 → CH 5+ +CH 3
CH 3+ +CH 4 C2 H 5+ +H 2
⎫
C2H4+ + H2
⎪
CH+2 + CH4 ⎪
⎪
⎬ Secondary ion production process
+C2H3 + H2 + H
+
⎪
CH4 + C2H5+ → C3H5+ + 2H2 ⎪
⎪⎭
Small relative concentrations of C3H 7+ , C2 H +2 , C3H 3+ , C3H 4+ , and C4 H 9+ are formed, but they do not make
significant contribution to the total ionisation.
M+CH 5+ → MH + ( + ) + + CH 4 ⎫⎪ ⎫
⎬ Proton transfer
t ⎪
M + C 2 H 5+ → MH + or( M + ) + + C2 H 4 ⎪⎭ ⎪
⎪
+• ⎪
MH → H + M
+
⎪ Chemical ionisation process; CH 5
+
⎪
M + C2 H 5+ → ⎡⎣ M + C2 H 5+ ⎤⎦ ⎪
⎪ acts as a Bronsteed acid , and
⎪
M + C3 H 5 → ⎡⎣ M + C3 H 5 ⎤⎦
+ +
⎬
⎪ +
Hydride, i.e. H abstraction charge exchange ⎪ C 2 H 5 as a Lewis acid to produce

⎪
⎪
M + C2 H 5+ → ⎡( M − l ) + C2 H 6 ⎤
+
⎣ ⎦ ⎪ ions from thhe substrate M
⎪
M+C2 H 5 → ⎡⎣ M + C2 H 4 H ⎤⎦
+ + − ⎪
⎪
⎪
⎭
However, proton transfer may still occur from C2 H 5+ to stronger bases (i.e. C2 H 5+ + BH → BH +2 + C2H4) and this
behaviour has been observed for water and ammonia. Hydride transfer reactions, i.e. (B+ + C2H6 formations) in
which C2 H 5+ acts as a Lewis acids are also observed.
In the presence of a good proton acceptor, the ions CH5+ and C2H5+ act as Bronsted acids and protonate the
sample molecule:
CH5++ BH → BH2+ + CH4
C2H5+ + BH→ BH2+ + C2H4
C2H5+ + BH→ B+ + C2H6 (Hydride transfer)

Here, B represents any organic (possibly inorganic group). Since CH4 is a weak Bronsted base, CH 5+ is a strong
Bronsted acid and proton-transfer reactions to stronger bases than CH4 will occur. These reactions being exothermic,
BH +2 ions produced above will dissociate to form other ions.
B+ + H2
BH2+
A+i + Bi
where A i+ is one of the possible fragmentations. C2H4 is a stronger Bronsted base than CH4, and hence C2 H 5+ is
a weaker Bronsted base than CH 5+ .
These reactions are typical of those observed for alcohols, aldehydes, esters, etc., and also for many biochemi-
cal compounds typically encountered in recent chemical ionisation applications.
If the sample material is not a good proton acceptor, the chemical ionisation process will occur as a hydride ion
abstraction or as a dissociative proton transfer.
e.g. the CI spectra of n – hexadecane (Methane reactant, PCH4 = 1.0 torn) are consistent with our previous
discussion.
⎡ + ⎡
CH5+ + n − C16 H34 → CH4 + ⎢⎣ C16 H35⎢⎣
+ +
C16 H33 + H2 C16−n H33−2
C2 H 5+ + n C16 H 34 → C16 H 33
+
C2 H6
The most striking difference between the CI and EI mass spectra of n – hexadecane is much larger relative intensity
+
of the chemical ionisation C16 H 33 (m/e : 225) ion (relative intensity 36 percent) than any electron impact C16 ion.
Additional fragmentation occurs to give a mass spectral pattern that is similar in appearance to the electron
impact spectrum of a hydrocarbon, but the abundance of the quasi-parent ion is greatly increased relative to the
fragment ions. The increased relative abundance of the quasi-molecular ion has proven to be of great value in
many studies, particularly with relatively complex bioorganic molecules. CIMS can be used to obtain the mass
spectra of low molecular mass (200 daltons) molecules having a large number of polar functionalities or high
molecular mass compounds (800–10,000 Da).
Typical peaks in the CI (reactant gas: CH4, PCH4 = 1 torr) and EI mass spectra of some molecular systems are
compared in Table 11.1.
Table 11.1 A comparison of CI and EI mass spectra of some amines.
m/e CI EI Remarks
Intensity Intensity
142 19.9 29.6 (n-C4H9)2 NCH2+
156 2.5 0.3 (n-C4H9)2 NCH2CH2+
170 3.5 0.0 (n-C4H9)2 NCH2CH2CH2+
184 23.1 0.0 (n-C4H9)2NC+HC3H7, (M − l)+, hydride abstraction
185 10.8 1.6 (n-C4H9)3N+, M+, charge transfer
186 21.1 0.0 (n-C4H9)3 NH+, (M + 1)+ proton transfer
214 1.7 0.0 (n-C4H9)3 NC2H5+, addition
The (M+ 1)+ and (M − 1)+ ions are caused by proton transfer and hydride-transfer reactions. The ions at m/e:
214 and m/e: 186 indicate that the electrophilic attack takes place on the nitrogen atom of the molecule while the
ion at m/e: 184 shows that attack occurs on the alkyl radicals resulting in the abstraction of H−. The peaks at m/e:
142, 156 and 170 result due to the following types of reactions.
+
CI: (n − C4H9)3N + 5 → (n − C4H9)2N = CH +2 + CH4 + C3H8
•
EI: (n − C4H9)3N + e → (n − C4H9)2N = C+ H2+ C3 H 7 + 2e
Since the ionisation energy of C2 H 5+ (8.3 − 8.8 eV) is large compared to that of tri-n-butylamine (7–8 eV),
the large relative abundance of parent ion, i.e. = 7.7 per cent of the total ionisation in CI, is due to the transfer of
charge from ethyl ion to the amine molecule.
(n − C4H9)3 N + C2H5+ → (n − C4H9) 3N+ + C2H5
The meaningful difference between ionisation potentials of primary (8.5 eV) and tertiary (7–8 eV) aliphatic
amines enables to distinguish between isomers by the CI technique.
COOH
N
Similarly, M+ ions may be absent or appear as a negligibaly low intensity peak in EIMS of but a
strong peak corresponding to quasi-molecular ion appears in the CIMS of the molecule, H
+
COOH + CH5 COOH + CH4
+
N N
H H2
(M+1)+ m/e: 115
A comparison of CI and EI mass spectra of 3-chloro-l, 2-epoxypropane (Table 11.2), ClCH2 CH2 CH2 shows
O
that virtually no parent ion is formed by the EI technique but an appreciable amount of (M + 1)+ ion is formed in
the CI system.
Table 11.2 A comparison of CI and EI mass spectra of 3-chloro-l, 2-epoxypropane.
m/e CI EI Assignment
57 38.7 17.6 C3H5O+ ( M + 1−HC1 )+

75 0.8 35
ClCH2 − CH = CH+, ( M + l − H2O)+
77 0.3 37
ClCH2 − CH = CH+, ( M + l − H2O )+
91 0.5 0.03 ( M − l )+, Hydride, i.e H− abstraction
92 0.0 0.06 M+
93 31.3 ( M + l )+
95 10.0 ( M + l )+
The fact that the oxygen is held by two bonds in a cyclic structure (i) results in stability of the protonated
molecule, and (ii) m/e: 93 and 95; contrasted with loss of water molecule resulting from protonation of the OH
group in higher molecular mass alcohols.
–H2O
CH2 Cl CH— CH2 CH2 Cl CH— CH2 (M + 1 — H2O)+
+ m/e: 75, 77
O O
H
(i) (ii)
m/e: 92 m/e: 93, 95
The attack at Cl also involves proton transfer but with immediate loss of HCl to form C3H5O+ (m/e: 57).
Let us now examine the CI and EI mass spectra of n-heptyl-propionate (C2H5COOC7H15) listed below in
Table 11.3.
The addition of ethyl and allyl ions to the ester produces four different ions in the spectrum as follows:
C2H5+ + C2H5 COOC7H15
∗
O C2H5+
C2H5C OC7H15
Table 11.3 A comparison of CI and EI mass spectra of n-heptyl propionate.
m/e CI EI Assignment
57 15.3 20.5 C3H5O+

75 38.0 7.2 C2H5COOH2+
97 6.3 0.5 C7H13+
H
103 7.7 0.0 C 2 H 5CO OC 2 H 5 + , alkyl exchange
H
115 0.9 0.2 C 2 H 5COOC3H 5 + , alkyl exchange
171 0.9 0.00 (M − 1)+, hydride (H−) abstraction

172 0.0 0.00 M+
173 12.6 (M + 1)+, proton transfer
201 0.8 C2H5C (OC2H5)(OC7H15)+, ethyl addition
213 1.0 C2H5C(OC3H5)(OC7H15)+, allyl addition
The dotted lines in the structural formula indicate that the ionic attack may be on either oxygen atom. This
transient intermediate can react forward in two ways:
(i) Stabilisation by collision with methane, i.e.
∗
O C2H5+ O C2H5+
+ CH4
C2H5C OC7H15 C2H5C OC7H15
m/e = 201
(ii) Elimination of C7H14 to form a protonated ethylpropionate ion (alkyl exchange reaction), i.e.
C5H11
H OH
O C H
C2H5 C O C2H5 + 1 C7H14
C2H5C + CH2 +
O m/e: 103
C2H5
The allyl ion also reacts in an analogous manner, i.e.

+
H
C2H5COOC7H15 + CH5+ C2H5COOC7H15 + CH4
m/e = 173
C2H5COOH2+ + 1 – C7H14
H C5H11 m/e : 75
O C
H
C2H5 C + CH2
O
H
C7H15+ + C2H5COOH
m/e : 99
C2H5CO+ + C7H15OH
m/e : 57 C7H13+ + H2
m/e : 97
Quantitative analysis of a mixture of known compounds can also be carried out by CIMS from the relative
sensitivities of the compounds under analysis. The relative sensitivities of the compounds in question are
estimated from the mixture of known composition. These values are then used for the analysis of unknown
mixtures of these compounds. Analysis of mixtures of n-dodecane, n-tetradecane, and n-hexadecane have been
carried out by this technique and satisfactory results have been obtained, n−C12H26: 24.3 (25.0), n−C14H30: 50.4
(50.0), n−C16H34: 25.2 (25.0). The quantities within parenthesis are the expected values.
In conclusion, the spectra produced by CIS are very different from conventional EIS and are sometimes
much more useful for both qualitative and quantitative analysis, particularly of high molecular mass and
polyfunctional compounds.
11.6.2 Desorption Ionisation Mass Spectrometry (DIMS)

Analysis of high melting inorganic solids such as metals, oxides, refractory materials, etc., and of large polar
nonvolatile organic molecules of interest in biology and medicine, i.e. protein and antibiotics, is not possible
by classical mass spectrometry that is based on the ionisation of the volatile substances in the gaseous phase by
electrons as in electron impact mass spectrometry, photons as in photon impact mass spectrometry and chemical
reactions as in chemical ionisation mass spectrometry. The nonvolatile high molecular mass organic biological
molecules can’t be vapourised without extensive, even catastrophic decomposition. However, such analytes can
be mass analysed by desorption mass spectrometry, also known as soft ionisation mass spectrometry, that is based
on the extraction of ions from a substrate surface either by the use of strong electrostatic fields or by particles such
as atoms or ions or by the use of radiations, i.e. lasers.
The electrostatic desorption ionisation techniques, viz., Field Ionisation Mass Spectrometry (FIMS), Surface
Ionisation Mass Spectrometry (SUIMS), Electrohydrodynamic Ionisation Mass Spectrometry (EHIMS), Spark
Source Ionisation Mass Spectrometry; (SSIMS) except Electrospray Ionisation Mass Spectrometry (ESIMS) is
usually preferred for the analysis of high melting inorganic substances, while the particle or radiation desorption
ionisation spectrometry (as well as ESIMS) are usually used for the analysis of high molecular mass nonvolatile
large biological molecules. The particle or radiation desorption ionisation mass spectrometry is also known as
violent ionisation or energy sudden ionisation mass spectrometry. The energy sudden ionisation techniques, viz.,
Plasma Desorption Mass Spectrometry (PDMS), Fast Atom Bombardment Mass Spectrometry (FABMS), sec-
ondary ion/fast ion bombardment mass spectrometry (SIMS–FIBMS) and Laser Ionisation Mass Spectrometry
(LIMS) have been able to produce intact ions from remarkably large molecules. In these methods, particles or
radiations with high translational energies are used for ionisation of molecules from the surface of the analyte.
The ionisation occurs via the translational modes of the sample molecule rather than the vibrational modes of
the sample, consequently less thermal destruction of the ionised molecules occurs since the analyte molecule
leaves its solid or liquid environment within a time of the order of 10−12s. Efficient ionisation has been observed
when the analyte is dispersed in a layer of suitable matrix of high boiling polar liquid rather than on a bare solid
surface of the analyte. The sample should be more hydrophobic than the matrix so that it will occupy the matrix/
vacuum interface. The polar solvent promotes ionisation and allows diffusion of fresh sample to the surface. The
lifespan of the ions thus produced is about 20−30 minutes compared to a few seconds for ions produced from
solid samples. The product ion currents are usually very small and except in the case of laser desorption, decrease
rapidly with increasing molecular mass of the sample. When the ions are very large, their detection with multipli-
ers (dynodes) requires post acceleration that are sometimes awkwardly high. Furthermore, the ions often have
high levels of internal excitation that can cause significant peak broadening due to pre-dissociation.
The mass spectra usually exhibit prominent peaks corresponding to the adduct ions such as (M + H)+ compared
to M+ ions which are sometimes not observed at all. The adduct ions such as (M + Na)+or (M + K)+or (M + NH4)+
also appear in the spectrum either due to the presence of salt impurities such as NaCl or KCl or NH4C1 in the
sample or upon addition of such salts to the analyte. Adduct ions due to the solvent such as [(solvent)nH]+ are also
prominent but do not complicate the spectrum and may be used for indexing the mass numbers.
The combination of DI, such as SIMS, LD and MS/MS, allows surface analysis of heterogenous materials with
both sensitivity (of a surface) and selectivity (of a compound). In addition, it improves the analysis of bulk samples and
allows species generated directly by desorption or by ion/molecule reactions in the ion source to be characterised.
The selectivity of DI to both precharged moieties as well as surface has been used for derivatisation of samples
chemically so as to produce ionic species at the surface. This can be done in situ and has been found effective in
enhancing sensitivity and specificity of detection.
(a) Spark Source/Spark Source Ionisation-Mass Spectrometry (SS/SSI-MS) The potential value of SSMS
was first demonstrated by Dempster around 1920s, but the technique came of age in 1950s when N B Hannay used
the SSM spectrometer for the analysis of semiconductors. A schematic diagram of a spark source mass instrument
To vacuum
pump
Accelerating and
foucssing grids
Ion
collector
Ion
analyser
Sample Molecular
electrodes ion beam
Fig. 11.16 A schematic diagram of a spark source mass spectrometer.
is shown in Fig. 11.16. In SSMS, the radio frequency (800 kHz) spark of high intensity using potential field of
greater than 100 kV is produced between the two sample electrodes provided the sample is an ionic compound.
Nonconducting samples can be mixed with some conducting material and converted into an electrode. The amount
of sample required is less than 10 mg. Pulse length and repetition rate are controlled for selective volatilisation and
to avoid overheating of the sample. The spark vapourises and ionises the sample. This method often leads to multi-
ple ionisation like M++ and ions of high kinetic energy. The ions produced in the ionisation chamber are accelerated
and fed to the mass analyser. For better resolution, double focussing instrument (especially with Mattuch–Herzog
geometry) is used in SSMS.
The most important advantages of SSMS as an analytical tool are its high detection sensitivity and nonselectiv-
ity for all elements (including impurities as well) in the sample.
The technique is used for trace analysis of wide range of sample types, for example, high melting solids such
as metals, oxides, refractory materials, etc.
(b) Surface Ionisation (SUI)/Filament Coating Mass Spectrometry (SUI/FC-MS) Just like SSMS,
FCMS is also used for the analysis of inorganic materials. In FCMS, a very small amount of the material, i.e. a
few micrograms is coated on a ribbon (tungsten) filament which also generates electrons. At a very high tempera-
ture (2000°C), of the filament, the sample vaporises and also ionises. For better efficiency, different numbers of
filaments and configurations are used for which a triple filament is the most popular one. The filament is placed
in the ionisation chamber and the ions are fed to the mass analyser after acceleration.
The probability of volatilisation of sample from a surface as a positive ion is a function of the ionisation
potential (IP) of the sample and the work function (j) of the filament and is expressed as
m° ⎛ IP − ϕ ⎞
∝ exp ⎜ ⎟ (11.30)
m +
⎝ kT T ⎠
where, m° is the number of neutral molecules and m+ is the number of positive ions.
This technique is very useful for analysis of inorganic compounds that have low ionisation potentials (3−6 eV)
but is hard for organic compounds with ionisation potentials over the range of 7−13 eV. The method is especially
useful for determining the isotopic abundance in inorganic compounds for geochemical applications such as age
of rocks and minerals.
(c) Field Desorption/Field Ionisation Mass Spectrometry (FDMS/FIMS) This technique provides
sample ionisation at relatively low energy with resultant reduced fragmentation and increased relative
abundance of the parent ion. Desorption of adsorbed species under the action of very high electrostatic
fields (of the order of 108 V/cm) was first observed by E W Muller in 1956 and by M G Inghram and R
Gomer in 1955. But field ionisation mass spectrometry of organic molecules was first introduced, i.e.
put into practice by H D Beckey in 1969. In this technique, a few micrograms of a sample is applied to a
fine wire of a few micrometers diameter on whose surface is disposed an array of sharp needles or ‘whis-
kers’. The wire is placed in a vacuum and is maintained at + 8 kV (for positive ions) or −8 kV (for negative
ions) and can be heated (Fig. 11.17). The field at the tips of the needles can be as high as 108 V/cm. The
analyte molecules desorb as M+ ions from the tips of the nee- To vacuum
pump
dles by Coulombic repulsion where the field strength is very Sample
high. The energy available for field ionisation and subsequent
+ 8 kv
excitation of a molecule is generally about 12−13 eV. Since
organic molecular ionisation potentials range from 7−13 eV, Moleculer
ion beam Ion
many molecules will have very little excess energy in the par- analyser
ent ion to cause fragmentation. Frequently, a quasi-molecular
ion MH+ is observed due to a surface reaction of the sample
with adsorbed water on the blade or emitter. A significant ion
peak is observed in the heptanal [CH3(CH2)5CHO] spectrum Accelerating and
Wire
at a mass of 115 (C7H14OH+). focussing electrodes
Although highly nonvolatile substances can be converted into Fig. 11.17 A schematic diagram of a field desorption
ions but sample preparation is very tedious. It is very difficult to find mass spectrometer.
and maintain the combination of temperature and voltage suitable for a particular species. Further, since the desorbed
ions have very high energy, relatively high-resolution expansive magnetic sector analysers are used for their analysis.
This technique proved to be valuable for studying surface phenomena such as adsorbed species and the results
of chemical reactions on surfaces. Since only a very small amount of internal energy is excited in the molecule
by this process, few or no fragment ions are formed. This enables to determine the molecular mass and with a
high-resolution spectrometer, the empirical formula for labile compounds; the amount of component parts of
a mixture of compounds.
(d) Electrohydrodynamic Ionisation Mass Spectrometry (EHMS) EHMS was first introduced by D S
Simons, B N Colby and C A Evans, Jr. in 1974. In EHMS, the sample is dissolved in a low vapour pressure
nonvolatile liquid such as glycerol and injected into an evacuated chamber through a small capillary which is
maintained at high voltage. The low vapour pressure nonvolatile liquid is preferred, otherwise it will freeze dry
from rapid evaporation into vacuum. Solute ions along with molecules and clusters of solvent are desorbed from
the emerging liquid by the high field at its surface and are mass analysed. The drawback of this technique is that
the desorbed ions are usually solvated with one or the other molecule of the solvent. As in FDMS, the high ion
energies in EHMS require expansive magnetic sector analysers. Moreover, few liquids that have low vapour pres-
sure are good solvents.
(e) Electrospray Ionisation Mass Spectrometry (ESI-MS) Electrospray ionisation has emerged as a
powerful tool for producing intact ions from large and complex species in solution. In ESI-MS, highly charged
droplets of an analyte in a suitable solvent are dispersed from a capillary or needle in an electric field and evapo-
rated in flowing gas such as N2 or CO2 called the bath gas at atmospheric pressure, and the resulting ions are drawn
into a mass spectrometer inlet. Electric field at the tip of the liquid orifice (i.e. capillary or needle) should be of
the order of 40,000 V/cm magnitude and the liquid should have an electrical conductivity in the range 10−9−10−8
ohm−1 cm−1. The number of unit charges per drop may run as high as 6 × 106 for a drop 34 m in radius. ESI-MS
differs from FD- and EH-MS in that desorption is from small charged droplets of solution into an ambient bath
gas instead of into vacuum.
The initial experiments on electrospray ionisation were carried out in 1917 by John Zeleny, but the idea
of using electrospray dispersion of an analyte solution in a bath gas (N2) to produce solute ions for mass
analysis including charge residue model (CRM) was originally developed in 1968 by Malcolm Dole and
co-workers (L L Mack, R L Hines, R C Mobley, L D Ferguson and M B Alice). A well-defined breakthrough
of the ESI-MS in its applications to identify large biomolecules, i.e. polypeptides and proteins of molecular
mass 40 kDa was made in 1988 by M Mann, C K Meng and J B Fenn.
Charge Residue Model According to CRM, the charge on each droplet of an analyte solution is given by
q2 = a(3Vg e)(or q = [a(3Vg e)]1/2 (11.31)
where q is the charge on the droplet, V its volume, g its surface tension and e its dielectric constant. Evaporation
of solvent from each droplet causes a decrease in their diameters. Consequently, the charge density on the
surface of each droplet increases until the so called Rayleigh limit is reached at which the Coulomb repulsion
becomes of the order as the surface tension. The resulting instability, sometimes, called a ‘Coulomb explosion’
tears the droplet apart, producing charged daughter droplets that also evaporate. It has been shown that the most
stable state is obtained if the droplet having a = 4 breaks down into four equal daughter droplets provided the
charges remain on the parent droplet as the droplet evaporates. This sequence of events repeats until the radius of
curvature of the daughter droplet becomes small enough that the +

+
+
+ +
field due to the surface charge density is strong enough to desorb +
+ +
ions from the droplet into the ambient gas. The desorbing ions Evaporation + +
include cations (or anions) to which are attached neutral solvent + Initial drop +
+
+
or solute species; thus generating so called quasi-molecular ions + +
+
+ +
for mass analysis. The schematic diagram for formation of quasi- +
molecular ions from the parent charged droplets has been shown
in Fig. 11.18. Desolvation in the
self generated
It is to be borne in mind that the electrospray of the dilute + + +
+ +
electric field +
solution of an analyte is to be carried out in vacuum other- + + +
+ + + Quasi-
wise the drops would not evaporate within the required fraction molecularion
of a second because of the strong evaporation cooling of the
drop. The fall in temperature of the drop due to evaporation in Fig. 11.18 Schematic diagram for the formation of
vacuum, provided no heat is transferred to the drop, is given by quasi-molecular ions from the parent charged droplets.
the expression
ΔT = T − To= (ΔHv/Cp )ln x (11.32)
where, T is the temperature to which the residue drop falls from the initial temperature To , ΔHv is the heat of
vaporisation of the liquid and Cp is the heat capacity of the liquid, x is the fraction of the drop remaining at
temperature T. When half of the drop has evaporated, x = 0.5, the drop in temperature of the analyte droplet for
benzene as solvent would be 160°C. Thus in order to maintain a high rate of evaporation, heat must be transferred
to the drop. This is done by evaporating the drop in flowing nitrogen gas at atmospheric pressure.
A block diagram of an ESI apparatus has been shown in Fig. 11.19. The dilute solution of an analyte to be
ionised is introduced into the evaporation chamber of plastic film through a hypodermic needle which is main-
tained at a few kilovolts relative to the walls of the chamber. The high voltage of the needle varies between 0
and 30 kV. The operating voltage needle is usually 10 kV negative with −3 kV on the first collimating plate and
−1.4 kV on the second plate. The distance between collimating plates and the needle and supersonic nozzle can
easily be changed. The resulting field at the needle tip charges the surface of the emerging liquid, dispersing it by
Coulomb forces into a fine spray of charged droplets. A concurrent flow of bath gas (N2) at one atmospheric pres-
sure (pressure, temperature, rate of flow and direction of flow of the gas depend on the purpose of studies and the
design of a particular ESI apparatus) enhances the evaporation of the charged droplets. The solvent vapours from
the evaporating droplets along with any other uncharged material are swept away by the flow of the bath gas. The
stream of quasi-molecular ions are collimated by electrostatic plates to the supersonic nozzle-skimmer system.
Some of these ions emerge at the exit end of the nozzle as a supersonic free jet in the first of the two vacuum
chambers, i.e. nozzle and skimmer chambers. The nozzle chamber is pumped by a mechanical pump (I) while
the skimmer chamber is pumped by a cold trapped diffusion pump backed by a mechanical pump (II). When the
pressure in the evaporation chamber is atmospheric, the pressure in the nozzle chamber is 0.13 torr and in the
skimmer chamber is 1 × 10− 4 torr. The beam of skimmed ions are directed by electrostatic lenses to the MS for
mass analyses.
N2 gas
out Skimmer
Quadrupole
Needle Plates Nozzle mass spectrometer
N2 gas in
liqiud in
Pump Pump
Spary chamber
(I) (II)
Electrostatic
lenses
Fig. 11.19 Schematic diagram of apparatus for the production and detection of macroions.
The supersonic nozzle and skimmer system has the following advantages: (i) Studies can be carried out at
much higher pressures in the evaporation chamber since a two-stage reduction of pressure is operative in the
system (ii) the intensity of the final beam of ions is much higher (eleven times or more) compared to that of
conventional oven beam, (iii) in case of beam of mixed gases, i.e. seeded beams, the heavier molecules are more
concentrated in the beam, and (iv) considerable velocity monochromisation takes place.
The ESI-quadrupole mass spectrum of a mixture of seven quaternary ammonium salts which cannot be
vapourised without decomposition in a mixture of methanol–water (50:50) shows no evidence of either fragmen-
tation or interference between species. The spectrum of the mixture gives peaks at
Components of the mixture m/e (ªPeak height in mm)
(CH3)4N+ 74 (3)
(CH3)3C10H21N+ 200 (3)
(CH3)3C12H25N+ 228 (22)
(C4H9)4 N+ 242 (4)
(C4H9)4 P+ 259 (13)
(CH3)3C16H33N+ 284 (4)
(CH3)3Cl8H37N+ 312 (7)
The ESI-quadrupole mass spectrum of the antibiotic cyclic decapeptide gramicidin S(A): in a methanol–
water mixture (50 : 50) shows peak corresponding to the singly and doubly charged parent ions respectively at
m/e :1141 (2 mm) and 571:(41 mm). The average value of molecular mass of the compound based on the natu-
ral abundance of isotopes has been found to be 1141.5. The relative intensities of these peaks depend not only
upon the composition of the solution but also on the concentration of the analyte. The intensity of the doubly
charged ion grows enormously while that of the singly charged ion disappears at very low concentration. The
relative magnitudes of intensities of the singly and doubly charged ions also depend on the structure. At con-
centrations for which the peptide cyclosporin (made up of eleven amino acids; one of which is not a normal
acid but has been found to be a b-hydroxy, singly unsaturated amino acid and is derivative of threonine) whose
size is similar to that of gramicidin S(A) shows peak corresponding to singly charged ion while gramicidin
S produces doubly charged ions. The smaller peptide bradykinin (B) exhibits a strong peak corresponding to
triply charged ion as expected since gramicidin S has two basic amino acids in its ring while the slightly smaller
sized bradykinin has basic residues at both ends of the sequence.
Amino-acid sequence data in proteins can be obtained by mass spectrometry, i.e. ESI-tandem MS using even
nanogram material of the sample. The mass analysis of proteins with molecular masses up to at least 130,000 can
be carried out by ESI-mass spectrometry. The marriage of ES with FT ICRR to provide high resolution proved
to be very useful in the analysis of very large macro ions, i.e. macromolecular ions with molecular mass greater
than 130,000.
Val Orn Leu Phe Pro
Pro Phe Leu Orn Val
(A) : Gramicidin. S
Arg—Pro—Pro—Gly—Phe—Ser—Pro—Phe—Arg
(B) : Bradykinin
(f) Plasma Desorption Mass Spectrometry (PDMS) In PDMS, very high energy ions are used to desorb and
ionise molecules in a solid-film sample. It was developed in 1976 by R D Macfarlane and D F Torgerson and was
the first method to demonstrate feasibility for studying high molecular mass proteins and complex antibiotics.
Desorption and ionisation of a solid film of analyte on nickel foil (10−3 mm thickness) is carried out by the
impact of a fission product of a radioactive isotope, usually californium-252 (Fig. 11.20). Each fission event of
252
Cf gives rise to two fragments 142 Ba18+ and 106 Tc 22+ with kinetic energies of 79 and 104 MeV respectively.
These fission fragments when pass through the sample, raise its temperature to about 104 K. Consequently, at such
high temperature, desorption and ionisation of the molecules of the sample takes place, i.e. the analyte exists in
Radioactive
isotope + –
or
Sample film
Ni foil (10–3 mm)

Fig. 11.20 A schematic diagram showing the impact of a fission product of a radioactive isotope on a solid sample film on nickel
foil (10-3 mm).
the plasma state (gaseous state with equal number of positive and negative ions). The ions are accelerated and
fed to the mass analyser. PD produces better molecular ion signals than FAB and FIB over the molecular mass
range of 10,000–20,000 Da. The high molecular mass of ions that have been produced by this technique in 1989
by G Jonsson et al. is 45,000. The liquid matrix used in PD spectrometry is nitrocellulose.
(g) Secondary Ion/Fast Ion Bombardment Mass Spectrometry (SIMS/FIBMS) In SIMS, a beam of
secondary ions is used to desorb and ionise the molecules from the solid or liquid surface of an analyte. The
former technique was first developed as an analytical method in 1950s by Richard E Honig while liquid SIMS was
developed in 1981 by M Barber and co-workers.
An incident beam of high-energy ions, such as 40 keV Cs+ is used to bombard the sample layer obtained by
dispersing the analyte in thioglycerol, glycerol, thioglycerol/diglycerol (1:1). More hydrophobic matrices used are
tetragol [HO(CH2CH2O)4H] and tetracol [HO((CH2)(CH2)(CH2) (CH2)O)nH]. The desorbed ions after acceleration
are fed to the mass analyser. Just like PDMS, MH+ is usually the most abundant ion in the spectra of large molecules.
The high molecular mass obtained using this technique was 24,000 Da by M Barber and B N Green in 1987.
(h) Fast Atom Bombardment Mass Spectrometry (FAB-MS) A block diagram of fast atom bombardment
mass spectrometer is shown in Fig. 11.21. In FAB-MS, beams of neutral atoms of Xe or Ar having high transla-
tional energy are used to desorb and ionize compounds gently from the surface of a solid or liquid matrix, making
it possible to obtain spectra of large, involatile polar organic molecules such as peptides, amino acids, etc. Fast
atom bombardment of solids as a new ion source for mass spectrometry was first introduced in 1981 by M Barber,
R S Bordoli, R D Sedgwick and A N Tyler. In matrix assisted FAB-MS, the →
q → matrix used are the same as
liquid
those used in fast ion bombardment mass spectrometry. The fast atoms of Xe (or Ar ) are produced first by ionis-
ing Xe atoms with electrons to give xenon ‘radical cations, i.e. Xe + e → Xe +• + 2e. The radical cations are then
accelerated to 6–10 keV to obtain radical cations Xe +• of high translational energy. The accelerated radical cations
so produced when passed through xenon give rise to Xe atoms with high translational energy.
Fast atom
gun
Fast
atom beam
Surface of a
sample matrix
+ + + Ion
Probe analyser
Ion beam
Metal probe
tip
Fig. 11.21 A block diagram of a fast atom bombardment mass spectrometer.
Xe + e → Xe +•+ 2e (electron impact)
Xe +• → Xe (acceleration to 6–10 keV)

+•
→ → →
Xe +• +Xe → Xe + Xe +• (high-energy Xe atoms)
The highest molecular mass of ions of polar analytes that have been analysed with FAB-MS is 24000 by
M Barber and B N Green in 1989.
(i) Laser/Soft Laser Desorption Ionisation Mass Spectrometry (LIMS/SLIMS) A soft laser beam is used to
desorb and ionise molecules from the surface of the analyte. It was developed in the late 1970s by Matten A Post
Thumus, Peiet. Kistemaker and Henk L C Meuzelaar while matrix-assisted laser desorption deionisation mass spec-
trometry (MALDIMS) was developed in 1985 by Franz Hillenkamp and Michael Karas and independently by Koichi
Tanaka and co-workers. In MALDI, different solid or liquid matrices containing a highly ultraviolet substance with
absorption maximum matched to the wavelength of the laser pulse are used. For example, a nitrogen laser beam which
has a wavelength of 337 nm, which is not absorbed by the aromatic amino acids in proteins and peptides, is used to
avoid fragmentation. YAG laser at 266 nm is also used. The organic acids such as nicotinic acid, a-cyano-4-hydroxy-
cinnamic acid, etc., are used as matrices. In matrix-free desorption, porous silicon surface is used to trap the analyte
molecule because of its high absorptivity in the ultraviolet, acts as an energy receptacle for the laser radiation. Laser
desorption ionisation on silicon (LDIOS)-TOFMS can handle a broad range of compounds including carbohydrates,
peptides, glycolipids, natural products and small drug molecules in the mass range of 150–12000 Da with little or no
fragmentation, e.g. 337 nm N2 laser mass spectra of des-arg-bradykinin show peaks at m/e: 905 (MH+) in absence
of any salt; 905 (MH+), 927 (MNa+) in presence of NaCl and 905 (MH+), 943 (MK+) in presence of K3PO4 buffer
solution; of antiviral win drug (I) exhibit peaks at m/e: 357 (MH+), fragment peaks; 299, 180, 177, 121, of N-octyl-
b-D-glucopyranoside exhibit peaks at m/e: 293 (MH+), 315(MNa+) and 23 (Na+ ); and of mixture of caffeine (C8H10-
N4O2), win drug and reserpine (C33H40N2O9) exhibit peaks at 196 (MH2++), 357 (MH+) and 609 (MH+) respectively.
CH3
N
O O
O N
(I) CH3
The highest molecular mass of ions that have been produced with LD is 2,10,000 by Karas and Hillenkamp
in 1988.
To summarise, some of the typical desorption ionisation mass spectral techniques and the energisation of a
surface by various types of sources for the ejection of charged particles are shown in Fig. 11.22.
LD SIMS FAB PD
Ar0
Ar+
Laser Energetic Energetic Fission
beam ion atom fragment
Laser desorption Secondary Fast atom Plasma

ions mass bombbardment desorption
spectrometry
Primary beam Secondary ions
in
out
Sample Implantation Energy Desorption

surface transfer
Fig. 11.22 Energisation of a surface by a variety of means results in ejection of charged species which
give a mass spectrum of the solid.
11.7 ISOTOPE RATIO MASS SPECTROMETRY

Isotope ratio mass spectrometry deals with the direct measurement of ratio of isotopes, e.g.12CO2,13 CO2;
1
HOH, 3HOH, etc. A trace material can be detected even after great dilution. In an isotope ratio mass spectrometer,
the ion currents corresponding to the two isotopic ion beams are measured simultaneously. The ratio of the two
currents will give the ratio of the isotopic species. Isotopically labelled tracers are used to study the functions of
the body. The age dating of minerals and rocks is based on the rate of decay of radioactive nuclides. Thus if the
decay rate of 238U to 206Pb, 40K to 40Ar and 87Rb to 87Sr is known, and the ratio of the isotopes of one of these pairs
is measured then the age of minerals and geological deposits can be estimated.
11.8 TANDEM MASS SPECTROMETRY (TANDEM MS)

With tandem-MS technique, it is possible to zero in on molecules of a particular compound or on groups of
molecules which share a chemical trait. In tandem MS, i.e. MS-MS, a precursor ion usually M+ or MH+ is mass
selected and fragmented by ‘collision-induced dissociation (CID), also known as ‘collisionally activated dissocia-
tion (CAD)’ , followed by mass analysis of the resulting product ions. Heavier collision gas such as Ar or Xe is
used for low-energy collisions whereas He is used for high-energy collisions.
The collision-induced dissociation method was introduced in 1968 by Keith R Jennings. The technique requires
two mass analysers in series (or a single mass analyser that can be used sequently) to analyse the precursor and
product ions and a schematic diagram of tandem-MS technique is presented in Fig. 11.23. This technique provides
structural information by establishing relationships between precursor ions and their fragment products.
The combination of soft ionisation methods with collision induced provides power to tandem-MS in the analysis
of mixture. One of the most popular types of MS instrument is the triple quadrupole mass spectrometer which was
invented by Richard A Yost and Christie G Enke in 1978.
Tandem-MS technique provides evidence for the formation of metal–ligand bonds in the solvated metal ions
and new metal–metal bonds in metal complexes in laser desorption process. For example, LD-MS/MS spectrum
of pure silver foil in the presence
+
of vapours of ethanol
+
and isopropanol shows peaks at m/e : 107, 153 and 167 and
are attributed to 107Ag+, Ag … O Et and Ag … O Pr respectively. The kinetics of desolvation reaction provides
H H
direct information on the relative solvating power of the various organic molecules attached to the metal centre.
C2H5OHAg+
C2H5OH…Ag+…HOC3H7
Ag+
CH = NOH
N
C3H7 OHAg+
(II)
LD-MS/MS spectrum of Co2L3 (M = 377) where L represents the acetylacetone ligand (CH3COCH2COCH3,
molecular mass = 100) shows peaks corresponding to the Co2L3+ (377+), CoL3+ (359+), Co2L2O+ (334+), Co2L2+
(318+), Co2LO+ (234+), Co2L+ (218+) and CoL+ (159+) ions. The data show a strong tendency for both metal atoms
to remain associated in the daughter ions formed from a mass selected parent ion of Co2L3.
LD-MS/MS spectrum of zwitter ion (CH3)3N+(CH2)5CO2− in varying amounts of HCl indicates that the intensity
of the MH+ peak (m/e: 174) grows while that of peak at m/e: 128 decreases as the concentration of acid increases.
This suggests that the intensity of the molecular ion peak of zwitter ion can be enhanced by converting the zwitter
ion to molecule with a net charge.
The formation and characterisation of derivatives of amines by quaternisation in the gas phase can be car-
ried out by LD/SI—MS/MS technique. For example, pyridine aldoxime (II) and ethyliodide adsorbed on a
silver surface when sputterted into the gasphase by secondary ionisation and analysed by MS/MS technique
shows peaks at m/e : 179+, 119+ and 109+, 107+. They are attributed to the formation of diethyl analog of pyri-
dine aldoxime and its decomposition products. The peaks at m/e: 109+ and 107+ are attributed to silver ions.
The last two examples prove the selectivity of DI to both precharged moieties, as well as surfaces.
Positive ion ESI-CID-MS/MS or CID-MS/MS or ESI-MS/MS and other soft ionisation MS/MS techniques
proved to be very useful for identification and sequencing of polysaccharides andtheir derivatives. The nature of
precursor ion depends upon the type of derivatisation. Thus the molecular mass related ions such as (M + Na)+,
[M + H]+, [M' + Na]+, etc. (where M represents the free acid form and M′ represents the full sodium salt) act
as precursor ions. As an example, the basic structures of trisaccharides and their fragmentation patterns based
on ESI-CID-MS/MS are given in schemes (1) and (2) respectively.
Ionisation
Source
M2
M4 M1
M3
Mass spectrum
Mass analyser I
m/e M
Fragmentation
Reaction or
collision reagion
Mass specific detector

Mass analyser II
M2 − m3
M2 − m2 M2 − m4
MS/MS Spectrum
M2 − m1 M2 − m5
Multiplier
M2
m/e
Fig. 11.23 A schematic representation of Tandem-MS technique.
OH NHAc OH
OH O
HO O OR NaO3SO O
O O
O O O O
NHAc
HO OH OH OH
HO
O O
H3C H 3C OH
OH OH
HO HO
Lea 1
OH OH
OH O OSO3Na O
HO O NaO3SO O O
O OR O
O O
NHAc NHAc
HO OH HO OH
O NaO3SO O
H 3C OH OH
OH OH
HO HO
Lex 2
OSO3Na OH
O
NaO3SO O
O O
O NHAc
OH
NaO3SO
H 3C O
OH
OH
HO 3
Scheme 1. Basic structures of Lea- and lex-type trisaccharides (left) and mono-, di-, and trisulphated
Le -trisaccharide propyl glycosides (1-3) (right).
x
286 290
–(Fuc·H2O) –(Fuc·H2O)
432 (436) 576

CH2OH
−NaHSO4
O CH2OH
HO +
O (+Na)
OSO3Na OC3H7
O m/e : 696
−SO3
HO
287 O O 616
NHAc
CH3
HO 430
HO 550
−NaHSO4
OH −SO3
470
1
286 290
–(Fuc·H2O) –(Fuc·H2O)
432 (436) 678

CH2OH
−NaHOS4
O CH2OH
HO +
O (+Na)
OSO3Na OC3H7
O m/e : 798
−Na2S2O7
NaO3SO
389 O O 576
−NaHSO4 NHAc
CH3 −NaHSO4
269 HO 412
532
HO 652
−NaHSO4
OH −Na2S2O7
430
2
286 290
−(Fuc·H2O) −(Fuc·H2O)
432 (436)
CH2OH
O
NaO3SO CH2OH +
O (+Na)
OSO3Na OC3H7 m/e : 900
O
−NaHSO4
NaO3SO
O O 780
−Na2S2O7 491
269 CH3 NHAc
−SO3 −NaHSO4
HO 634 514
411 HO 754
−Na2SO3 −NaHSO4
OH
285
3
Scheme 2. Fragmentation patterns of the positive ion ESI C1D-MS/MS of sulphated Lex-trisaccharides having
the respective (M′ +-Na)+ as precursor ion.
11.9 GLC MASS SPECTROMETRY

GLC separates a mixture either of original volatile components or their volatile derivatives but does not characterise
the components. One can infer indirectly by comparing the retention times with the possible known compounds. It
can play a valuable role in combination with any other instrumental technique which can accept gaseous or volatile
liquid samples and which is compatible in speed. The most important are infrared spectrometry, mass spectrometry,
etc. But such a marriage of GLC with MS proved to be very fruitful. GLC chromatograph is directly coupled to a
mass spectrometer and the mass spectrum of each separated component is obtained as it emerges from the GLC.
The effluent from a gas chromatograph is usually a pure substance in small quantity (10−12−10−3 g) in the gas phase
and the identification of such a pure substance in very small quantity (10−9−10−12 g) is thus possible from its mass
spectrum. Such different types of gas liquid chromatograph-mass spectrometers are widely used these days for the
separation and characterisation of the components of a mixture of volatile substances. To summarise, GLC-mass
spectrometry deals respectively with the separation and characterisation of the components of mixture of volatile
substances with GLC chromatograph and mass spectrometer combination. The gas chromatograph separates the
components of a mixture, while the mass spectrometer then imparts structural information about each one.
The direct coupling of GC and TOFMs was achieved in the mid 1950s by Roland S Gohlke and F W McLafferty
in collaboration with W C Wiley, Maclaren and Dan Harrington. At about the same time, GC was coupled to a
magnetic sector spectrometer by Joseph C Holmes and Frank A Morrell. But the great utility of modern GC-MS
was made possible by the advent in 1960’s of carrier gas separators also known as sample enriching devices
that removed the GC carrier gas (i.e. He) prior to introduction of a sample into the high vacuum mass spec-
trometer. Schematic diagram of a typical computerised gas chromatograph/mass spectrometer is shown in a
self-explanatory, Fig. 11.24 (a) and (b). Separators were developed independently by Einar Stenhagen; Kagnar
Ryhage and by Watson and K Biemann. Separators designed for the enrichment of concentration of sample in the
carrier gas are based on (i) the faster diffusion of carrier gas through the membrane, and (ii) preferential passage
of sample molecules through a membrane called the semipermeable membrane.
Ryhage introduced the effluent of the GLC as a jet striking against a metal surface which had two, very closely
spaced tiny holes. The sample passed through the holes but the bulk of the lighter carrier gas was pumped out.
This led to enrichment and the same process was repeated again when a sample sufficiently enriched to give a
good mass spectrum was obtained. A schematic diagram of a two-stage membrane separator based on the model
of Watson and Biemann separator is shown in Fig. 11.25.
Electron beam
source
Sample Focussing Ion sorting
injection port lenses region Detector
Carrier
gas inlet Capillary
Data
system
GC coloumn
Gas chromatograph
oven Ionisation
GC coloumn
occurs here
Transfer
line
(a)
Digital
GCMS data
Has peak No
occurred?
Intensity of
peak
Time of
peak
Mass of
peak
Store
(b)
Fig. 11.24 (a) Schematic diagram of a typical capillary gas chromatograph/mass spectrometer, and
(b) a conceptual representation of computer processing of digitised GCMS data.
Membrane or
fritted glass
tube
Carrier
gas in
To ion
source
From
GLC
To pump
Fig. 11.25 Schematic diagram of a two-stage membrane separator based on Watson and Biemann model.
As the sample carrier gas mixture passes through the glass fritted disc, the lighter gas diffuses out faster and
the sample gets enriched. The whole system can be heated up to 400°C. The quantity of any gas going through
the porous glass is given by
Q k p M (11.33)
where, p is the partial pressure of the component, M is the molecular mass, and k is a constant that depends on
the conductance of the porous tube. The pressure on the outside of the porous tube is assumed to be sufficiently
reduced so that back effusion is negligible. The ratio of the sample Qs to the quantity of helium carrier gas VHe that
goes through the frit, is
Q p M He
(11.34)
V He ri p He M s
VHe is expressed in units of cm3 atm/min.

For molecular flow to occur through the porous tube, the mean free path of the gas must be large in comparison
with the diameter of the pores. The average pore size in ultrafine sintered glass is about 10− 4 cm.
5 10 −3
Thus, the allowed pressure, corresponding to L = 10-4 cm; (L = cm, where p is the pressure in torr;
p
1 torr = 1 mm of Hg) is 50 torr. In practice, 1–10 torr is preferred to avoid adverse effects from the pore-size
distribution. Silver membranes instead of glass-fritted disc, have also been used. However, if a Teflon membrane
(up to 300°C) is used instead of the fritted disc then only the organic substrate diffuses into the mass spectrom-
eter. Silicon rubber membranes, which act as semipermeable membranes for organic vapours and reject the
carrier gas, are also used. By such techniques, good mass spectra have been obtained for compounds as large
C20−C24 fatty acid esters. It is to be noted that one separator should perform over the entire range of GCMS anal-
ysis. The effectiveness of every separator depends on the chromatograph (flow conditions), the chemical sample
(temperature), and the mass spectrometer (pumping speed). Time of flight, or quadrupole mass filter or quadru-
pole ion trap spectrometers are ideal for coupling with gas liquid chromatographs as they are fast compared to
double focussing mass spectrometers. Scan times being faster (i.e. fraction of second is required to record each
spectrum), and so several mass spectra can be obtained during the elution of a single peak from the GLC unit.
Such mass scans at different points of a GLC peak enable to check its homogeneity. Thus, partially overlapping
chromatographic peaks can be resolved.
High-pressure or performance liquid chromatography-mass spectrometry (HPLC-MS) is used for compounds
which are not amenable to GLC-MS analysis such as nonvolatile, thermally labile and high-molecular-mass polar
compounds. Pumping pressures of several hundred atmospheres are required to achieve reasonable flow rates with
modern packings which are made up of particles with diameters of 10 mm or less.
Applications of modern GLC-MS include identification of optical isomers, environmental analysis, drug testing,
e.g. the identification of steroids and drug metabolites in urine and blood, pharmacological studies, etc.
In conclusion we can say that chromatographic methods normally mated to mass spectrometry are liquid
chromatography, however, particularly for desorption ionisation, it is possible to interface thin layer, and
paper chromatography and electrophoresis directly impinging the ionisation beam on the surface of the
chromatogram, i.e. direct examination of chromatograms to characterise individual spots can be carried out.
The primary ion, atom or laser beam is simply directed on the spot and a mass spectrum is recorded.
11.10 MERITS AND DEMERITS OF VARIOUS TYPES

OF IONISATION—MASS SPECTROMETRY
The advantages and disadvantages of various types of ionisation techniques are related to the volatility and non-
observance of the molecular ion peaks of the compounds. The volatility problem is mainly encountered in case of
natural products (in vivo and vitro), i.e. amino acids, peptides, proteins, steroids, triglycerides, sugars, etc., because
of their higher melting and boiling points, i.e. low vapour pressures. The merits and demerits of these techniques are
elaborated here by taking examples of amino acids and steroids. The major peaks in the mass spectrum of leucine
(molecular mass = 131) obtained by EI-MS, CI (isobutane, 200°C) and FD techniques are compared in Table 11.4.
Table 11.4 A comparison of EI, CI and FD mass spectra of leucine.

Technique m/e (peak highest in mm × 10)
EI 131(1), 86(40), 74(13), 44(26), 30(23)
CI 132(39)
FD 132(25), 87(39), 86(29), 60(11), 45(31), 31(22)
The data reveal that EI spectra of amino acids or their derivatives, i.e. ester give weak or nonsignificant molecular
ion peaks, but non-EI spectra give either molecular or quasi-molecular ion peaks. The weak molecular ions in the
EI spectra arise because amino acids easily lose their carboxyl group and the esters easily lose their carbo-alkoxyl
group upon electron impact.
COOH
+ •
R–CH R CH NH2 + COOH
NH2
• +
m/e : 86(40)
m/e : 131(1)
or
COR⬘
+ •
R–CH R CH NH2 + COR⬘
NH2
• +
m/e : 131-(44 + R⬘)
Here R = (CH3)2CHCH2
The FD spectrum exhibits an MH+ (m/e : 132) peak that fragments as follows to give peak at m/e: 86.
•+
COOH •
R CH R CH NH3 + COOH
NH3
+
m/e : 87
+ •
RCH N H2 + H
m/e : 86
The important peaks in the spectrum of cholesten-5-ene-3, 16, 22, 26-tetrol (molecular mass = 434) a polyhydroxy
steroid (A) obtained by different M-S are EI (m/e (peak height in mm × 10)) : 434 (non-existent), 318 (2), 300 (5),
285 (4.8), 267 (6), 145 (7), 99 (39), CI (isobutane, 200°C) : 417 (16), 399 (32), 381 (13), 373 (non-existent), 283
(25), 271 (23), 255 (12), 99 (39) and FD : 434 (38).
OH
22 24
21 20 23 27
25
18 17
OH 26
12 OH
19 11 13 16
1
15
2 9 14
10 8
3 5
7
HO
4 6
(A): Cholesten-5-ene-3, 16,22,26-tetrol
EI spectrum exhibits a weak or non-existent molecular ion peak. This shows that dehydration is partly caused
by heating. In CI-MS, the absence of protonated molecular ion peak reveals that the dehydration of the protonated
ion occurs readily (i.e. m/e: 435 ⎯− H O → 417). Facile dehydration is not observed at all in the FD spectrum that
shows only a molecular ion peak at m/e : 434.
To summarise, it is concluded that the merits and demerits of a particular type of ionisation mass spectrometry
are not only governed by the volatility and non-observance of molecular ions of the compounds but also upon the
sensitivity and specificity of the ion-mass spectrometry.
11.11 INTENSITIES OF THE SIGNALS IN THE MASS SPECTRUM

Intensity of the signal is a measure of the relative abundance of ions which is given as the per cent intensity of a
given peak relative to the most intense peak in the mass spectrum. The most intense peak in the spectrum is also
called the base peak and is not necessarily the peak due to the molecular ion. Intensity of a given peak is also
expressed in per cent of total ionisation (Σ) given by peak height and is obtained by dividing the intensity of a
given peak by the sum of the intensities of all peaks, i.e.
Intensity of the given peak × 100
∑ = Peak height = Sum off the iintensities of all peaks
(11.35)
The relative peak heights of the peaks give a measure of the relative numbers of different ions produced
from the sample.
Problem 11.5: Determine the peak heights corresponding to the mass spectral data on methanol given below:-
m/e 15 19 29 31 32 33
Per cent of base peak 35.48 0.29 58.80 100 68.03 0.98
Solution Sum of the intensities of all peaks = 263.58
Substituting the needed data into Eq. (11.35) we get
m/e 15 19 29 31 32 33
Peak height ( or ∑) 13.46 0.11 22.30 37.94 25.81 0.37
11.12 INTENSITY OF THE PARENT PEAK AND FACTORS AFFECTING IT

One finds a number of peaks in the mass spectrum. The parent peak is usually the last distinct peak in the spectrum
and after that one usually observes small peaks due to isotopes at m/e : M + 1, M + 2, etc. For example, the mass
spectrum of NH3 shows peaks at m/e: 14, 15, 16 and 17 and a small peak at 18. These correspond to the molecular
ion and to the fragments.
• • •
−e +• −H + −H +• −H +
NH3 [NH3] [NH2] [NH] [N]
17 16 15 14
The M+• + 1 peak appears in the m/e = spectrum because most elements (e.g. nitrogen and hydrogen) have more
than one naturally occurring isotope. Although most of the NH3 molecules in a sample are present as 14 N1H 3 , a
small but detectable fraction of molecules is also present as 15 N1H 3 and 14 N1 H 2 2 H . These molecules, i.e 15 N1H 3
or 14 N1 H 2 2 H , produce molecular ions at m/e: 18, i.e. at M+• + 1. This is true if the compound is pure and has no
impurities. Thus if the compound contains only C, H, N, O, F and I only, the intensities of the (M + 1) and (M + 2)
peaks can be calculated from the following formulae:
⎡( )⎤
per cent ( ) = 100 ⎢
⎣ ( ) ⎥⎦
= 1.1 × number of carbon atoms + 0.36 × number of nitrogen atoms (11.36)
⎡( ) ⎤ ⎡1. ( number of carbon atoms

t )2 ⎤
per cent ( ) = 100 ⎢ = ⎥ + 0 20 × number of oxygen atoms
⎣ ( ) ⎥⎦ ⎢⎣ 200 ⎦ (11.37)
These formulae are useful provided we have a prior idea of the molecular formula of the compound under con-
sideration. These equations aid in determining the molecular formulae of the compounds.
The parent peak is not equally prominent in all organic compounds. It is big in some cases, small in others
and negligible in some others or totally absent (due to very fast rate of decomposition) in some others. Since
delocalisation of positive charge over a large volume, i.e. polarisability, increases the stability of the compound
the intensity of the molecular ions in such systems will be large. This is possible where ever p-electrons are
present and as such aromatic compounds show the highest intensity for parent ions, i.e. double bonds, cyclic
structures, and especially aromatic (or heteroaromatic ring) stabilise the molecular ion and thus increase the
probability of its appearance. Biemann has classified molecular ions according to their stability (or intensity)
as follows:
Aromatic compounds > Conjugated olefins > Alicyclic compounds > Sulphides > Unbranched hydrocarbons
> Ketones > Amines > Esters > Ethers > Acids > Branched chain hydrocarbons >Alcohols.
The modified version of Biemann series has been expressed as follows:
(a) High Intensity Group Molecular ions of this group follow the order of intensity indicated below:
Aromatic compounds > Conjugated alkenes > Cyclic compounds > Organic sulphides > Short, normal alkanes
> Mercaptans.
(b) Medium Intensity Group The order of intensity of molecular ions in this group are: Ketones > Amines >
Esters > Ethers > Carboxylic acids~Aldehydes~Amides~Halides.
(c) Lowest Intensity Group The molecular ions are not frequently detectable in aliphatic alcohols, nitrites,
nitrates, nitro compounds, nitriles, and highly branched chain compounds.
The nature of the functional groups also affects the stability or intensity of the parent ion like the following
which assist in fragmentation:
(i) side chain of the aromatic system
(ii) alkylic systems; R—CH2—CH = CH2
(iii) hetero atoms like O, S, N, etc., which can stabilise charge.
(iv) branching
Stabilisation of charge is also governed by resonance and inductive effects.
Ionisation voltage, i.e. energy of the bombarding electrons and temperature of the ionising source, also
influence the parent-ion intensity and the fragment pattern. The relative intensity of the molecular ion
increases as the ionisation voltage is decreased from 70–15 V but it decreases for all fragment ions. Thus
one can obtain the molecular mass of the compound. The parent-ion intensity decreases as the temperature
of the ion source, i.e. sample, increases and vice versa. The increase in temperature increases the initial
thermal energy and thus the internal energy of the molecular ion and hence its instability. For instance, the
average internal energy of 1,2-diphenylethane decreases by 0.4 eV on lowering the temperature from 200°
to 75°C.
The effect of temperature on internal energy increases as the molecular size increases but the reduction of
sample temperature for larger molecules is limited by sample vapour-pressure requirements. If the pressure in
the ion source is increased, ion molecular reactions (i.e. collision reactions) take place and ions higher than the
molecular mass may be observed with the reduction in the parent ion intensity.
11.13 RULES FOR THE MODE OF FRAGMENTATION OF MOLECULAR IONS

Before describing the rules for the mode of fragmentation of molecular ions, the most common terms and symbols
used in the discussion need be defined:
M+ – designates an even electron ion (or an ion with paired electrons)
M +−• designates an ion with unpaired or odd electrons or cation radical
M−– designates a negative ion (i.e. an electron is added to the ion)
M −•− represents a negative ion radical or anion radical.
- Full arrow represents the transfer of pair of electrons.
(a) Heterolytic Pair Cleavage A pair of electrons move together towards the charged site.
-fish hook or a single barbed-arrow-represents the transfer of a single electron.
(b) Homolytic Cleavage A single electron moves towards the charged site.
•+
a – cleavage — : R C C C X C C C R Rupture of a bond on an atom adjacent to the radical
γ β α α β γ
bearing atom but not the bond bearing the atom.
(c) Precursor The decomposing ion in any reaction. For example, the parent ion is a precursor to the daughter ion.
(d) Elimination Reaction Reaction in which cyclisation occurs to form a new bond between two parts of an
ion with the loss of the actual group connecting these parts.
(e) Displacement Reaction Reaction in which cyclisation occurs to form a new bond at a carbon atom with
the loss of another group attached to that carbon atom.
(f) Rearrangements Reaction in which the molecular arrangement of the atoms in either the ion or neutral
product is not the same as in the precursor ion. Rearrangements occur via cyclisation, displacement, elimination,
and hydrogen transfer.
Collision induced transformation.
(g) The Even Electron Rule The nitrogen rule is the basis of the even electron rule. According to the former
rule, the molecular masses of compounds containing C, H, N, (O) are odd when the number of nitrogen atoms
is odd. Thus when C, H, O compounds lose neutral molecules, i.e. even electron molecules such as H2O, HCN,
olefins, CH3COOH, etc., even mass fragment ions are formed whereas odd mass fragment ions are formed when
C, H, O molecules lose radicals such as CH3, OCH3, etc. According to the even electron rule, an odd-electron ion
decomposes to an even electron ion or an odd electron ion respectively with the loss of a radical or an even electron
molecule. Whereas an even electron ion always decomposes to an even electron ion with the loss of an even electron
molecule but not to an odd electron ion with the loss of a radical. To summarise, the rule can be written as
M +• → A+ + radical
or M +• → B+• +, even electron molecule
or B+• → C+ + radical
or B+• → D+• + even electron molecule
and E+ → F+ + even electron molecule
but not E+ → G +• + radical
Especially the rule is most common to decompositions which give metastable ions. Exceptions to the rule are
some di-iodides which successively lose iodine radicals from the molecular ion.
(h) Mode of Fragmentation of Positively Charged Molecular Ion The mass spectrum of any compound
shows dozen peaks, some big, some small and some negligible, e.g. hydrocarbons exhibit series of ions repre-
sented by CnH2n+1, CnH2n and CnH2n−1. The mass interval between two consecutive ions in each series is always CH2
( = 14 mass units) units. The largest peak in each ions series represents a CnH2n+1 fragment and this occurs at m/e :
14n + 1, this accompanied by CnH2n by CnH2n−1 fragments.
Ion series m/e

[CnH2n+1]+ 15, 29, 43, 57, …
+• 14, 28, 42, 56, …
[CnH2n]
+ 13, 27, 41, 55, …
[CnH2n−1]
A substance having a strong parent peak often contains a ring, and the more stable the ring, the stronger the
parent peak. Ring compounds usually contain peaks at mass numbers characteristic of the ring. Usually the most
intense ion, called the base ion, is taken as 100 per cent and the intensities (height × sensitivity) of the other ions are
expressed in percentage. Peaks with at least 5–10 per cent of the most intense ion are the only ones of interest. Sec-
ondly one does not usually gather much information from the low mass region of the spectrum as they arise from
any number of routes from the major fragmentation processes. With these restrictions, let us consider the mode of
fragmentation, i.e. the fragmentation pat tern of the molecular ions. As has already been explained earlier, ‘the more
stable the ion the higher is its intensity and if a daughter ion is more stable than the parent ion, its intensity will be
greater. The relative height of the molecular ion peak is greatest for the straight-chain compounds and decreases
as the degree of branching increases and usually decreases with increasing molecular mass in a homologous series
except that of fatty acid esters. Accordingly, the branched chain hydrocarbons do not exhibit molecular ion peak
while, straight chain hydrocarbons show a very weak peak. It follows from this that energetically more favourable
tertiary carbonium ions, (carbocations) are usually produced by fragmentation in preference to primary and sec-
ondary ions, (i.e the stability of carbocations follows the order: tertiary >secondary >primary); hence when one is
considering a branched chain aliphatic system, fission takes place at the branching and the charge goes with the
more substituted carbon atom and also stable allyl and benzyl groupings, ions and radicals; dehydration (water loss
in alchohols) and decarboxylation reactions occur readily, etc. In alkanes, the stability of carbon ion has the order.
> Tertiary > Secondary > Primary
r > Methyl
=CH CH + )
(CH 2 =CH-CH (R 3 C+ ) (R 2 C+ H) (RC+ H 2 ) (C+ H3 )
The stability of free radical lost depends upon (a) nature of the free radical, whether primary, secondary or
tertiary and (b) length of the straight chain since it permits greater dispersal of odd electrons. Greater the disper-
sal of odd electrons, greater the stability of the free radical, e.g. nC 4 H 9• is more stable than C3 H 7•, i.e. generally
the largest substituent at a branch is eliminated most readily as a radical. As stated earlier, the cleavage of the
molecular ion usually involves the rupture of the bonds near the site of localised positive charge. Accordingly, the
modes of fragmentation have been classified as follows. The bond fission at the carbon atom linked to a hetero
atom (i.e. a noncarbon atom) or to a group on which the positive charge resides is called a-. The cleavages of the
successive C⎯C bonds are respectively termed the a-, b-, g-, etc., fissions; a-ruptures are most common. Bond
fissions occurring without the migration of individual atoms or groups are known as ordinary fissions and those
occurring with by such migration are called rearrangement fissions. Products of fission in the former are a part of
the parent molecule while that of the latter are not.
(i) Ordinary Fissions These are primary single-bond cleavage processes associated with saturated and unsatu-
rated aliphatic and aromatic compounds and with some common functional groups such as amine, ketal, ether
(X = O), thioether (X = S), ketone, aldehyde, alcohol (X = O), thiol (X = S), ester, halides (X = F, CI, Br, or I), etc.
(i) Single-bond fissions in alkanes give rise to fragments which are even electron cations:
• +
[R CR3] + • R + CR3
where R represents the alkyl group.

The loss of the largest alkyl group is favoured and the charge resides on it.
The mass spectrum of propane shows two peaks at m/e : 15, 29 and I29+ >I15+. The ions are formed as
+• • +
[CH3 —CH2 —CH3] CH3 + CH2 — CH3
m/e : 29
+• • +
[CH3 —CH2 —CH3] CH3 − CH2 + CH3
m/e : 15
+ + +
Dispersion of electrons is more in C+2H5 as compared to C H 3 . Accordingly, C 2H5 is more stable than C H3.
In case of n-pentane, the mode of fragmentation is
• + m/e 57
CH3 + CH2 CH 2 CH 2CH 3
• +
C 2H 5 + CH2 CH 2 CH 3 m/e :43
•+
[CH 3CH 2CH2 CH 2 CH 3 ]
• + m/e :29
C 3H 7 + CH2 CH 3
+ m/e :15
• CH3
C 4H 9 +
Further fragmentation of the carbocations causes the loss of one or two hydrogen atoms. This gives rise to
additional peaks in the spectrum of the parent compound.
+
+ • H•
CH 3 CH 2 CH 2 ⎯− H →[CH 3 CHCH
H 2 ]+• C H 2 CH = CH2
m/e :43 m /e :42 m/e :41
4
+
The base peak corresponds to 43 .
The mass spectrum of 2-methyl butane is similar to that of n-pentane except that the base peak is observed
at m/e: 57. +•
⎡ CH3 ⎤ + •
⎢CH − CH − CH − CH ⎥ → CH C
|
H − CH 2 - CH 3 + C H 3
⎢ 3 2 3
⎥ 3
m /e :57
⎢⎣ ⎥⎦
57+ being secondary carbocation is more stable than 72+, 43+ and 29+ ions.
Let us now consider the mode of fragmentation of 3-methyl heptane. The formation of carbocations follows
the order:
+•
CH3 CH3 CH3 + CH3
C2H5 CH C4H9
C2H5 CH C4H9 C2H5CH CHC4H9 C2H5CC4H9
+ + +
m/e : 114 m/e : 57 m/e : 85 m/e : 99 m/e : 113
113+ is the most stable ion as compared to 114+, 99+, 85+ and 57+ ions. Thus loss of the largest alkyl radical at the
reactive site, i.e. at the branch is preferred.
(ii) Fragmentation often occurs at the allylic position (i.e. cleavage b to the double bond is favoured and results
in the resonance stabilised allylic carbocation). The rule does not apply to simple alkenes because of ready
migration of the double bond but applies to cyclo-alkenes.
• + + +
R CH2 CH CH2 −R• CH2 CH CH2 CH2 CH CH2
β α
m/e : 41
For example, the cleavage in myrecene occurs at the allylic position.

• + •
+
m/e : 69
+
•
m/e 67
The peak at m/e: 93 is due to increased conjugation followed by allylic cleavage:

• + • +
+
+
•
m/e 93
(iii) The charge is often stabilised by the elimination of a neutral fragment like CO, HCN, olefins, H2O, NH3,
H2S, ketene or alcohols.
+
+
R CH2 CH2 CH2 R CH2 + CH2 CH2
+ + +
O O
+ CO + CO
(iv) When an electron negative atom is present, the rupture occurs adjacent to the hetero atom with the charge
residing on the alkyl fragment. This applies to ethers, esters, halogens, alcohols, amines, etc. For example;
RX + ⎯ → R + + • X where X = Cl, Br, or I

: +• •
C2 H5 -O-C2 H5 C2H+5 + OC2 H5
:
+
O +• O
•
R C OC2H5 R C + OC2H5
+
+• +
R O • O
C R1 + R C ; R1> R.
R1 O O
: +• + •
R − CH 2 − NH 2 R − C H2 + N H2
:
H +• + •
CH3 C •S• CH2 CH3 CH3CH S C2H5 + CH3
CH3
+• •
[R − CH2 − OH] R − C H 2+ + O H
+• •
[R − CH2 − SH] R −C H +2 + SH
Hetero atoms stabilise the charge, i.e. the charge is retained by the hetero atom and the alkyl group appears
as a radical.
• + +
[R − CH2 − Br]+ • R + CH 2 − Br CH2 = B r
• + +
+•
[R − CH2 − Cl] R + C H 2− Cl CH2 = C l
•
[C 2H 5 − O − C2H 5] + •⎯⎯
→ C 2 H 5 + O+ C2H5
+•
O • + +
R +O C NH2 O C NH2
R C NH2
+•
O • +
R + O COC2H5
R C OC2Hs
CH3
+•
+ + +
C3H7 C NH2 C2H5C(CH3) NH2 > C3H7C(CH3) NH2 > C3H7C(C2H5) NH2
C2H5 m/e : 72 m/e : 86 m/e : 100
• +
R − CH2 − S+• H R + CH2 = SH
+• • + +
R CH2 OH R + CH2 OH CH2 OH
m/e : 31
• R R +
R +• − R2
+ ••
R1 C O H C OH C OH
•• ••
•
•
R2 R1 R1
R2 >R1 or R = alkyl, when R2 and/or R1 = H and M − 1 peak can usually be observed. Thus primary alcohols
will give peaks of mass 31, secondary alcohols of mass 30 + R and tertiary alcohols of mass 29 + R, i.e. at m/e:
31, 45 and 59 when R = CH3.
(v) If the hetero atom is connected by a multiple bond (e.g. C=O), the rupture occurs at the carbon-atom
attached to it and the charge is retained by the carbonyl group, i.e. resonance stabilised acylium ion.
+•
O
• + +
−R ••
2
R1 C R2 R1 C O R1 C O
••
•
− R1 + + ••
R1 C R2 R2 C O R2 C O
••
O
+ •
The base peak very often results from the loss of a larger alkyl group. Here, R2 > R1.
(vi) In alkyl substituted aromatic compounds, the cleavage occurs at the benzyl position, i.e. at the bond b to
the ring as benzyl ion will be in resonance with the tropylium ion:
+• +
CH2—R CH2
•
−R + + C2H2
+
m/e 91 m/e 65
Phenyl alanine ester, benzyl acetate, etc., show a strong peak at m/e : 91.
(vii) Alkyl-substituted saturated rings lose side chains at the a-carbon and the positive charge tends to
reside at the ring fragment. The peak corresponding to the loss of two ring atoms is much larger than
for the loss of one ring atom.
+ •
α R
+
•
R +
(viii) Compounds which have a cyclic double bond (i.e. unsaturated rings) undergo fission giving rise to an
olefin and the residual ion following a retro-Diel’s Alder reaction (i.e. double a-cleavage). The positive
charge tends to stay with fragment with low ionisation potential (Stephenson’s rule).
+•
+•
OH
+
OH
m/e : 86 (IP = 8.7 eV) (IP = 9.0 eV)
+•
+•
OH
OH
m/e : 68 (IP = 9.0 eV) (IP = 9.4 eV)
The following relative IP’s and their consequences are also noteworthy.
+•
H +•
O OH
+
m/e : 58 (IP = 9.8 eV)

(IP = 8.8 eV)
+• +•
OH
H
O +
(IP = 8.8 eV)

m/e : 10.4
(IP = 8.5 eV)
+• + + CH2
–6H • +
CH2
Cyclobutane
+•
–H • + + HC
C4 H3 +
HC
+
H H
H
m/e 51
H
+• + H H
•
+ + CH2
CH3—CH2—CH2 +
H H CH2
H H
n-propylketones m/e: 83
m/e: 126 m/e 55
+
CH2 = CH—CH2—CH2
+• +•
CH2
+
CH2
Cyclo-hexene
+•
+•
Limonene
The mass spectra of polycyclic compounds often involve the peaks of fragments that are formed by a fragmentation
which is the converse of diene sythesis (retrodiene-fragmentation). An example is the fragmentation observed in the
spectrum of trans-10 methyl -Δ6 −octan-2-one.
+• CH2 +•
H 1
O CH O
9 2
10 +
6 CH
5 CH3
CH3
CH2
(j) Rearrangement Fissions Uptil now we have discussed the salient features of ordinary fissions in which
the products of fission are a part of the parent molecular ion. Let us now examine the mode of bond fissions that
occur with the migration of individual atoms or groups within the molecule, i.e. in rearrangement fissions, frag-
ments are formed as a result of intramolecular atomic rearrangement. Rearrangements involving the migration of
the groups heavier than hydrogen are termed skeletal rearrangements; but H-atom migrations are most common
especially in molecules that contain hetero atoms. Hydrogen migration occurs not only in a molecular ion but also
in the fragment ion. In hydrogen migration, generally a six-membered transition state is formed although other
transition states are also common. Rearrangement fissions, i.e. molecular rearrangements are often associated
with elimination of small stable neutral molecules such as CO, olefins, water, NH3, ketene, HCN, H2S, or alcohols,
etc., and sometimes radicals also. Some of the typical cases of hydrogen migrations giving rise to transition states
before fragmentation are given here.
Hydrogen Migration via six-member Transition The fragment ions are also formed with the elimination
of a neutral molecule as a result of the rearrangement b fission with the migration of a g-hydrogen atom.
The fission occurs in the six-membered transition state known as the McLafferty transition state. Such a
state is formed when a molecule contains an heteroatom (e.g. O), a p system (usually a double bond and an
abstractable hydrogen atom g to the C = O system). Fragmentation via McLafferty transition state occurs in
aldehydes, ketones, esters and amino acid esters, alcohols, carboxylic acids, amides, unsaturated compounds,
substituted aromatics. For example,
+• +•
H CH3
H 2C CH2
CH2
CH
CH
H2C C CH2
C H2
H2
I-pentene m/e : 42
+• H H
H R
• + • +
N CH N N ••
CH2=CHR +
C CH2 C C
C CH2 CH2
H2
Alkylcyanide m/e 41
+•
H2
C +•
CH2 CH2
CH2
+
CH2
CH2 H
H
H
m/e 92
n-propylli enzeme
+•
CH2
CH +•
CH2 CH=CH2 CH2
+
H C C
O CH2 HO CH3
+•
CH2CH2OH CH2—CH2 +•
CH2
O+ • + H2CO
≡ H H
H
CH2+
•
CH2
+•
O
H + H2CO
CH2 H
H
+•
R1 H +•
OH
γ CH O R1CH
C +
β
CH C R2CH
CH2 X
R2 CH2 X
α
where, R1 or R2= alkyl or H; X = H (aldehyde), CH3 or C2H5 (ketone); OH (acid), OCH3 or OC2H5 (ester), NH2
(amide).
McLafferty rearrangement in alcohols occurs with the elimination of water molecule and alkene.
H
+•
O
•
•
H CH2
[CH2 = CHR] + •+ CH2 = CH2 + H2O
RCH CH2
C
H2
+• +•
H CH3
CH2 O
CH
+ R2 — CH = O
CH
HC CH R2
CH R1
R1
Allylalcohols
Many acetates show a peak corresponding to (M+•—CH3COOH) because of hydride transfer to carboxyl
oxygen via McLafferty transition.
+•
H
+
R1 — CH O R1CH
+ CH3COOH
R2 — CH C — CH3 •
R2CH
O
Higher esters R1C OR2 where R2 = ethyl, or propyl or butyl, etc., in addition to losing alkoxyl group, also
undergo single-and double-hydrogen rearrangements to give ionised carboxylic acid and protonated carboxylic
acid species, respectively. The proportion of double-bond hydrogen arrangement grows with increasing size of R2.
+•
H
O CH — R2 OH CHR2
CH3 C +
C CH2 O CH2
•+
CH3 O m/e : 60
+•
OH CHR2 +
•
OH
CR2
C CH2 CH3 C +
OH CH2
H 3C O
m/e = 61
An alkyl group of more than three carbon atoms in ortho position to the nitrogen atom of pyridines can undergo
migration of hydrogen atom to the ring nitrogen.
Here b—C—C bond fission occurs.
+• +• CH2
N CH2 N
H
H CH2
RCH
C +
R H
CH2
+ •
N CH2
H
Such a fission also occurs in pyrazines since all ring substituents are ortho to one of the nitrogen atoms.
(k) Double McLafferty Rearrangement Ketones in which the groups R1 and R2 contain three or more carbon
atoms exhibit double McLafferty rearrangement. A secondary hydrogen is preferred to a primary hydrogen atom
in this process. The mechanism involves
(i) Ketonisation of the intermediate enol by the hydrogen transfer, and (ii) Transfer of hydrogen to enolic oxygen.
+• +•
H
R1 CH O OH CHR1
+
H2C C CH2 CH2 CH2R2 C CH2
C CH2 CH2 CH2 CH2R2
H2
Enolic form I
Ketonisation
H +•
O CH R2
C CH2
H3C C
H2
+•
OH CHR2
+
C CH2
H3C CH2
Enolic form II
m/e 58
Orthoeffect Hydrogen migration via six-member transition state occurs in ortho-substituted aromatic
compounds or in cis-olefins with the elimination of a neutral molecule or radical. For examples
+•
O
+•
C M ROH.
O R
H
X
X = O,S , NH, CH2, etc.
O +•
C O+ •
C
OCH3 + CH3OH
H
C
H2 CH2
H2 +•
+•
C CH2
OH + H2O
H
C CH2
H2
O +• + O
N N
O •
OH +
H
N NH
H m/e : 121
+•
O +
N • N N
O −OH +
O H + CO
H
C C H
H2 H2
m/e : 92
+•
m/e : 137(M ) m/e : 120
In amines, a hydrogen radical is lost from the parent ion to M+•−1 peak.
CH2 +•
+•
HC H CH2
+ CH3OH
HC OCH3 CH
C
CH O
O C
Such an elimination of neutral molecule enables to distinguish (a) between cis and trans isomers, and (b) ortho-
substituted compound from m- and p-substituted isomers.
(l) Hydrogen Migration via Five-Member Transition

In hydrogen migration rearrangements, generally a six membered transition state is formed although other
transition states are also common. Hydrogen migration via five-member transition occurs as follows in the
following cases.
+• +•
CH2 X CH2
(CH2)n H (CH2)n + HX
CH2 CH2
where, X can be OH, OCOR, NH2, CN, etc.

+• +•
CH2 NH2 CH2
(CH2)n H (CH2)n + NH3
CH2 CH2
(m) Hydrogen Migration via Four-Member Transition

In case of n-alkylamides and o-acetates of phenols, hydrogen arrangements occur via four-member transition state
and ketene (CH2=C=O) is eliminated
+•
NH C O +•
NH2 + CH2 C O
H CH2
+•
O C O +•
OH + CH2 C O
H CH2
Hydrogen migration in molecular and fragment ions of ethers or alcohols or amines also occur via four-member
transition. For example,
CH3 H CH2 +
CH3 CH OH
CH3 CH O C CH3 CH3 CH O C CH3
+ H + H m/e : 45
+
m/e : 87
CH3 CH OH
H CH2
+
CH3 CH N CH2 CH3CH NH2 + C2H2
+
m/e : 44
CH3
+• +
CH3 CH S CH2 CH3 CH3CH S CH2
H CH2
+
CH2 +
•
•
+ CH3 CH SH CH3 CH SH
•
•
CH2
m/e : 61
CH2 H CH2 + •
α + NH2 CH2 + CH3
H2C NH CH2 CH3 CH2
+•
H
OH +
H RCH OH OH
+
RCH CH2 CH R CH C H +
CH2
C CH2 + CH2
OH H2
+ m/e : 31
lon of secondary
alcohol
a-hydrogen abstraction by 1,3-mechanism occurs in long-chain chlorides, e.g. n-pentyl and n-butyl chlorides
with the elimination of HCl. This is also the case of hydrogen migration via four-member transition.
+•
H Cl
| | +•
[R − CH 2 − CH 2 − Cl]+ • ⎯ → R C − CH 2 ⎯⎯
→ R CH CH
C 2 HCl
|
H
For R = C2H5; m/e: 56.

(i) Fission in Ring Compounds via C⎯C or C⎯X (where X = O, S, NH) Bond Followed by Rearrange-
ment Five-membered ring lactones, cyclic ketones (≥ C4) and five-membered ring hetero aromatics, namely
furan, thiophene, and pyrrole, undergo fission via C—O, C—C and C⎯X (X = O, S, NH) bond cleavage followed
by rearrangement respectively with the elimination of neutral molecule or radical. For example, the base peak
(m/e: 56) of g-valerolactone and the same peak of butyrolactone originates as follows:
+• +•
O •• O ••
+
O
C C
•
H2C O H2C O C
+ CH3CHO
H2C CH CH3 H2C CH CH3 H2C
H2C •
m/e : 31
Cyclobutane, cyclopentanone and cyclohexanone undergo fission via C⎯C bond cleavage followed by
rearrangement to give a peak at m/e: 55 as follows:
+ +
+• •
• O •
• O
•
• O
C C
C •
H2C CH2 H2C H C H H3C • CH
H2C CH2 H2C CH2 H2C CH2

+•
M
Cyclopentanone
+
O O
• C C
C2H5 +
CH CH
CH2 CH2
+
m/e : 55
Cyclohexanone also gives characteristic peaks at m/e: 83 and 42 as follows:

+
+ C O
C O
• •
H3C CH H2C CH
CH3 +
H 2C CH2 H2C CH2
C
H2 m/e : 83
C O +
•
H2C CH2 CH2 + •
+ CO + CH2 CH2 CH2 +•
H2C CH2 CH2
C m/e : 42
H2
(ii) Fission in Five-Membered Ring Hetero Aromatics Fission proceeds via C⎯X where, X = O, S, NH,
cleavage and rearrangement to give fragment ions with the elimination of neutral molecule or radical as follows:
+
HC CH HC CH HC CH
HC + CH HC + • CH HC + CH •
X X X
• •• ••
• + +
CH X + HC CH
CH
+
m/e : 39
HC CH HC CH +
X •
CH
HC
+
CH HC + CH
•
X X
•
m/e : 29(X = 0,)
••
45(X = S, : 28(X = NH)
X = O, S, NH
For X = S, NH, we write as follows:

HC CH HC CH CH HC
+ X•
HC CH HC CH CH HC + X•
• +
X• m/e : 58 (X = S),
+ +•
41 (X = NH)
The principal peaks of five-membered hetero aromatics are summarised below:
Heteroaromatic five-membered hetero aromatics m/e peaks

Furan Thiophene Pyrrole
(
39 C3H 3+ ) (
39 C3H 3+ )
( ) ⎛ ⎞ ⎛ ⎞
+ +•
39 C3H 3+ 45 H C ≡ S 28 HC ≡ NH
⎝ ⎠ ⎝ ⎠
⎛ +•
⎞ ⎛ +•
⎞
29 (HC ≡ O+) 58 C 2 H 2 S 41 C 2 H 2 N H
⎝ ⎠ ⎝ ⎠
(iii) Fissions Followed by Cyclisation gd C⎯C fission possibly followed by cyclisation gives rise to peak in
the spectrum of amides, e.g. isobutyramide.
H2 H2
C C
H2C C O • H2C C O
δ
γ R +
R CH2 NH2 H2C NH2
•+ +
m/e : 86
C8 and higher nitriles undergo fission via hydrogen migration followed by cyclisation to give a six-membered
molecular ion
R
CH
H +• H2
CH2
+• NH C
N CH2
C CH2 CH2
+
C CH2
H2C CH2 CHR
H2C CH2 C
H2
C
H2
m/e : 97
These general rules for the fragmentation of molecular ions apply to electron impact mass spectrometry. This
would change depending upon the relative importance of several processes occurring simultaneously. Molecular
ions with much lower energy or quasi-molecular ions with very different fragmentation pattern are generated by
other ionising techniques, i.e. chemical ionisation, desorption ionisation, etc., and the fragmentation caused by
such ionisation follows different rules.
(iv) Fragmentation of Polyfunctional Molecules The presence of more than one functional group in the mol-
ecule makes the interpretation of the mass spectrum complicated. The complexity results due to the competition
between two functionalities to control the fragmentation in the molecule. Based on the relative fractions of ions,
currents iOCH3 and iX arising from decompositions initiated by OCH3, and X-groups in bifunctional decanes (A) at 70
eV, a parameter that measures, the ability of some functional groups relative to OCH3 (which is assumed to be unity)
to control fragmentation in the mass spectra of polyfunctional molecules, has been defined.
C2H5 CH (CH2)4 CH C2H5
X OCH3
(A)
The values of parameter for some functional groups (X) relative to OCH3 are listed in Table 11.5.
Table 11.5 A comparison of the ability of functional groups to control fragmentation in polyfunctional
molecules of general formula (A) with reference to OCH3 taken as unity.
X iX(a) X iX(a)
— COOH 0.02 — SCH3 0.90
— CH2OH 0.06 — OCH3 1.00
— Cl 0.08 —NHCOCH3 1.21

— COOCH3 0.17 —I 1.31
O
CH2
C
CH2
O
— Br 0.23 5.14 (b)
— OH 0.25
— SH 0.36 —NH2 6.16
C O 0.41 (b) —N(CH3)2 12.00
(a) Relative to OCH3 taken as unity, (b) data refer to the 3-methoxy-8-ketodecane and 3-methoxy-8-keto-decane
ethylene ketal derived from (A).
As an example, the characteristic peaks at m/e (mm) : 218(11), 189(7), 186(10), 171(13), 157(3), 141(1), 139(5.2),
138(10), 109(17.5), 89(38.3) and 73(40) in the mass spectrum of 8-methyl mercapto-3-methoxy decane are translated
by the following scheme of fragmentation.
+ +
C2H5 CH OCH3 C2H5 CH SCH3
m/e : 73 m/e : 89
+•
C2H5 CH (CH2)4 CH C2H5
SCH3 OCH3
+•
m/e : 218 SC•
5
2H
• H
CH3OH
C
3
+ +•
m/e : 186 +
CH (CH2)3 CH CH C2H5
CH3SH C2H5 CH (CH2)3 CH CH C2H5
SCH3 H OCH3 +• m/e : 171
m/e : 138
m/e : 189 H OCH3
CH3OH + CH3OH
C2H5 CH CH (CH2)3 CH
+ +
CH (CH2)3 CH CH C2H5 C H CH (CH2)3 CH CH C2H5
SCH3 H OCH3 2 5
m/e : 189 m/e : 139
SCH3 m/e : 157
CH3SH
+
CH3SH C2H5 CH CH (CH2)3 CH
m/e : 109+ H m/e : 141 OCH3

CH 3O
Fig. 11.24 Scheme of fragmentation of 8-methylmercapto-3-methoxydecane.
11.14 INTERPRETATION OF MASS SPECTRA OF UNKNOWN

COMPOUNDS
The validity of structures assigned to the fragments generated by the electron impact and their origin from this
or that part of the molecule under examination is confirmed by comparison with the mass spectra of reference,
i.e. related compounds; particularly, if the structure of one of these is known and by examination of the mass
spectra of deuterated analogues. The derivation of the structure from the fragments requires great skill. When
a molecule contains both aromatic and non-aromatic parts in the structure, the aromatic part shows up a stable
entity and the fragmentation occurs in the non-aromatic part. Fission is often accompanied with elimination of
small stable neutral molecules such as carbon monoxide, water, ammonia, olefins, hydrogen sulphide, hydro-
gen cyanide, mercaptans, ketene or alcohols, often with rearrangements. Thus starting from the M+ ion, one
should follow progressively lower mass numbers; however in many cases it is convenient to start from the lower
mass numbers, i.e. from the peaks at m/e : 28(N2+) and m/e : 32(O2+). M-18 indicates water elimination, M-15
reveals methyl group in the molecule and if in addition there is M-29, it may be an ethyl group. M-28 indicates
the loss of either CO or C2H4. In a group of compounds from the same origin, the difference may be small
like OH being OCH3, vinyl being ethyl, etc., and they can easily be recognised. In some cases, compounds are
labelled with either stable isotopes (18O, 15O) or chemical labelling is resorted to (e.g. incorporation of CH3 or
OCH3 groups). The latter procedure, however, should be adopted only in those cases where there is no risk of
a given chemical label causing the change of the fragmentation routes. Some common losses from molecular
ions and the possible inferences drawn from the respective losses are tabulated in Table 11.6. The other useful
data for interpreting the mass spectra of organic compounds have been summarised in Tables 11.7–11.10.
The mass spectra of samples collected from thin layer plates, columns and greased apparatus contain impurity
peaks at m/e : 149, 167, 279 (phthalic acid derivatives plasticisers); 129, 185, 259, 329 (due to tri-n-butyl acetyl
citrate plasticisers); 133, 207, 281, 355, 429 (silicon grease); 99, 155, 211 (due to tributyl phosphate plasticisers).
So in such cases care must be taken while interpreting the mass spectra of the molecules.
Reproducible breakdown patterns have not only been obtained in small organic molecules with different
functionalities but also in high-molecular mass natural products, namely terpenoids, steroids, polysaccharides,
peptides and alkaloids. For instance in a steroid (A), the parent peak is strong. After ascertaining the formula
from the molecular mass, one should look for a peak corresponding to (M+• − 18), i.e. (M+• − H2O). The peak
at (M+• − 18) indicates the presence of OH group in the molecule with ~ 10 per cent certainty. But for acetates,
one finds a peak corresponding to (M+• − CH3COOH). Side chain at C(l7) can be found out by locating a peak
at 205–245. Then M-peak = 42 + R + 18n where, n is the number of hydroxyls, and R is the mass number of
the side chain. If there is no side chain on C(17), or a C = O then one will not get this peak. Then there will be
fragmentation due to the cyclic double bond and by postulating the double bond at the likely sites, one can derive
some of the fragments in the mass spectrum.
R
18 17
12 16
11 13
1
15
2 9 14
10 8
3 5 7
HO
4 6
(A)
For cholesterol (deposition of cholesterol in the blood vessels causes atherosclerosis) with molecular formula
C27H46O (molecular mass—386.67), there is a methyl group on C—10 also in (A) and
21 22
20 23
R= 26
17 24 25 27
This alcohol is synthesised in the body.

Table 11.6 Some common losses from molecular ions.
Mass loss from the ion (groups

Possible molecular candidate
associated with the mass lost)
1(H) —
2(H2) —
14
15(CH3) —
16(O) Ar-NO2, ≥ N+—O−, sulfoxide
16(NH2) ArSO2NH2, —CONH2
I7(OH) —
17(NH3) —
18(H2O) Alcohol, aldehyde, ketone, etc.
( ) ⎫⎪
⎬ Fluorides
20( )⎪⎭
26(C2H2) Aromatic hydrocarbon

27(HCN) Aromatic nitriles, nitrogen heterocycles
28(CO) Quinones
28(C2H4) Aromatic ethyl ethers, ethyl esters, n-propylketones
29(CHO) —
30(C2H6) —
30(CH2O) Aromatic methyl ether
30(NO) Ar—NO2
3l(OCH3) Methyl ester
32(CH3OH) Methyl ester
32(S) —
33 (H2O + CH3) —
33( ) ⎫⎪
⎬ Thiols
34( )⎭⎪
41 (C3H5) Propyl ester

42 (CH2CO) Methylketone, Aromatic acetate, ArNHCONH2
42 (C3H6) n-or isobutyl ketone, aromatic propylether, Ar—n—C4H9
43 (C3H7) Propyl ketone, Ar—n—C4H9
43 (CH3CO) Methyl ketone
44 (C3H8) —
45 (COOH); 17(OH) Carboxylic acid
45 (OC2H5) Ethyl ester
46 (C2H5OH) Ethyl ester
46 (NO2) Ar—NO2
48 (SO) Aromatic sulfoxide
55 (C4H7) Butyl ester
56 (C4H8) Ar-n-C5H11, ArO-n-C4H9;
Ar-iso-C5H11, ArO-iso-C4H9;
Pentyl ketone
57 (C4H9) Butyl ketone
58 (C4H10) —
60 (CH3COOH) Acetate
Table 11.7 Characteristic decomposition pathways of some saturated carbonium ions.
Precursor ion Decomposition products Transition (m/e) m*

C2H 5
+
C2H3 + H2
+
29 → 27 25.14
C3H 7
+
C3H5 + H2
+
43 − 41 39.09
C4H9+ C4H7+ + H2 57 − 55 53.07
C4H 9
+
C3H5 + CH4
+
57 − 41 29.49
C5H11 +
C3H7 + C2H4
+
71 − 43 26.04
C6H13 +
C4H9 + C2H4 and
+
85 − 57 38.22
C3H7+ + C3H6 85 − 43 21.75
C7H15+ C4H9+ + C3H6 99 → 57 32.82
+ + +•
Alkyl moiety M+, produces the characteristic C n H 2 n 1 and C n H 2 n 1 ion series while M+• produces series C n H 2 n
Table 11.8 m/e values associated with some members of ion types produced by some of the fragmentations.
Functional groups Ion type Ion series m/e
+
*Amine CH 2 = N H 2 30,44, 58,72...
( / e :30 )
≠
Ether ⎫⎪ +
⎬ CH 2 = O H 31,45,59,73
Alcohol⎭⎪ ( / e :31)
+
Ketone CH 3C O 43, 57, 71,85
( / e :43)
Hydrocarbon C 2 H 5+ 29, 43, 57, 71

( / e :29 )
*Molecular mass of saturated monoamine is an odd number. # Peak due to H2O loss suggests alcohol but not ether.
Table 11.9 m/e values of some rearrangement ions formed from various carbonyl compounds.
+• +•
R1 H
OH
CH O −R1CH = CHR2
C
CH C R1 or R2 =
alkyl or H CH2 X
CH2 X
Compound X m/e
Aldehyde H 44
Ketone (methyl) CH3 58
Ketone (ethyl) C2H5 72
Acid OH 60
Ester (methyl) OCH3 74
Ester (ethyl) OC2H5 88
Amide NH2 59
Table 11.10 Order of ease of cleavage of some C6H5X compounds, X. The bond strength of X
increases as we move downwards, e.g. CH3 > OCH3 > COCH3.
X Radicals or neutral fragments lost from M+•

COCH3 CH3
C(CH3)3 CH3
CH(CH3)2 CH3
COOCH3 OCH3
N(CH3)2 H
CHO H
C2H5 CH3
OCH3 CH2O, CH3
I I
OH CO
CH3 H
Br Br
NO2 NO2, NO
NH2 HCN
Cl Cl
CN HCN
F C2H2,F
H C2H2
# The energy required for the decomposition of the functionalities X grows as we move downwards in the table.
Problem 11.6: In the mass spectrum of neopentane there are observed lines corresponding to the following
values of m/e : 57, 41, 29 and 27. What positive ions might these lines correspond to?
Solution The molecular ion [C5H12]+ • for neopentane should appear at the m/e value of 72. The fragmentation
of [C5H12] • ion occurs as follows to yield the positive ions corresponding to the m/e values in questions.
+
+•
CH3 CH3
CH3 C CH3 + e CH3 C CH3 + 2e
CH3 CH3
+• m/e : 72
C5H12
•
CH3
C4H9+
m/e : 57
C2H4
CH4
C3H5+ C2H5+
m/e : 41 m/e : 29
H2
C2H3+
m/e : 27
Problem 11.7: Predict the structural formula of the compound (C9H2O) having a molecular mass of 128 from
the mass spectral data indicated below:
m/e 113 99 71 57 55 43 41 29 27
Intensity (in mm) 1.5 8 45 50 10 50 21 20 15
Metastable peaks : 25.14, 39.09, 53.07.

Solution The metastable peaks are due to the following transitions:
Transition Metastable peak
29+ → 27+ C 2 H 5+ → C 2 H 3+ +H2 25.14
43+ → 41+ C3H 7+ → C3H 5+ + H2 39.09
57+ → 55+ C 4 H 9+ → C 4 H 7+ + H2 53.07
Thus, the peaks at m/e : 29, 43, 57 spaced at an interval of 14 mass units are attributed to the hydrocarbon
ion series of the type C n H 2+n 1 i.e., C+2 H 5 , C3+ H 7 , C+4 H 9 while the peaks at m/e : 27, 41, 55 are attributed to
hydrocarbon ion series of the type C n H 2+n 1 formed from the ion series C n H 2+n 1 with the loss of H2 molecule.
The absence of molecular ion peak in the spectrum rules out the possibility of straight-chain hydrocarbon.
The peaks at 113, 99 and 71 arise due to mass loss of 15(CH3), 29(C2H5) and 57(C4H9) from the molecular ion
peak respectively and indicates the formation of tertiary carbon atoms. Accordingly, the structural formula of
the compound is
CH3
CH3 CH2 C CH2 CH2 CH2 CH3
CH3
The decomposition routes for the formation of expected ions have been summarised below:
CH3
+•
•M−15 = 113+
+•
•+ +•
M − 29 = 99+ CH3 M −57 = 71+
CH3
+•
CH3
CH3
• +
C 2H5 C CH2 + CH2 CH2 CH3
CH3
+
CH3 CH CH3
m/e : 43
CH3
+
C 2H5 C + CH2 − CH2 CH2 CH3
m/e : 57
CH3
CH3
Problem 11.8: The mass spectrum of 3-methyl-3-heptanol, C4H9 C OH, includes lines corresponding to the
C2H5
following mass numbers, m/e : 115, 101, 87, 73, 59, 55, 45, 43, 41, 31, 29, 27. What ions might these lines cor-
respond to?
Solution The m/e values in question results due to the fragmentation modes of molecular ion of 3-methyl-
3-heptanol indicated below:
CH3 +•
C4H9 C OH
C2H5
• +
CH3 CH3 CH3 + C4H9 C OH
• + • C2H5
C4H9 + C OH C2H5 + C4H9 C OH
+ m/e : 101 m/e : 115
C2H5
m/e : 73 +
C4H8 + CH3CH OH
+ m/e : 45
C2H4 + CH3CH OH + +
m/e : 45 C4H8 + C2H5CH OH C4H9CH OH + C2H4
m/e : 59 m/e : 87
+ + +
C2H4 + CH2 OH C4H8 + CH2 OH H C O + H2
m/e : 31 m/e : 31 m/e : 29
The peaks at 57, 43 and 29 are due to the formation of C4 H 9+ ,C3H +7 , and C3H 5+ fragment ions respectively.
+ + +
C4H9 C2H5 C2H3 + H2
m/e : 57 m/e : 29 m/e : 27
+ +
H2 + C4H7 CH4 + C3H5
m/e : 55 m/e : 41
CH3 +• CH3
+ •
C3H7 CH2 C OH C3H7 + CH2 C OH
m/e : 43
C2H5 C2H5
+
C3H5 + H2
m/e : 41
Problem 11.9: In the mass spectrum of 2-chloropropane (CH3 CH Cl) there are observed two doublets
CH3
and a base peak corresponding to the following values of m/e (intensity ratio): 80, 78 (1:3); 65, 63 (1:3) and
43 (Base peak). What ions might the doublets correspond to?
Solution The intensity ratio and a mass difference of two units between the lines corresponding to each of the
doublets suggest that the doublets in the spectrum are due to the two isotopes of chlorine. Accordingly, the m/e
values in question are due to the following modes of fragmentation of the molecular ion:
CH3
•
CH3 CH + CI
+
m/e : 43
CH3 CH3
+• +•
−e
CH3 CH CI (35) + CH3 CH CI (37)
CH3
m/e : 78 m/e : 80
CH3CHCI
CH3 CH3
+• +•
−e CH3 CH CI (35) + CH3 CH CI (37)
m/e : 78 m/e : 80
• + +
CH3 + CH = CI (35) + CH3 CH = CI (37)
m/e : 63 m/e : 65
For 1-bromopropane, the mass spectrum is 122, 124 (1:1) and 43 (base peak).
Problem 11.10: (a) Deduce the structural formula of the compound (molecular formula = C5H12O) from
the mass spectral data indicated below and justify the structures so obtained from the mode of fragmentation
of the compound.
m/e 87( M +• ) 70 57 55 43 42 31 29
A: Peak height 0.5 17 10 28 13 47 33 31

(in mm) (Base peak)
m/e: (
88 M +• ) 87 73 70 55 45 43 42 29
B: Peak height 0.5 0.6 3 0.6 7.5 47 8 1.5 4

(in mm) (Base peak)
m/e: 73 70 59 55 42 29
C: Peak height 25 0.5 47 16 2.3 14
(in mm) (Base peak)
(b) What inference can be drawn regarding the characterization of alcohols?
Solution (a) A:A M +• −1 peak is characteristic of a alcohol. The ( M +• −1) peak at m/e : 87 suggests that the
molecular ion is expected at m/e : 88. The peak at m/e: 70 corresponding to ( M +• − H2O) and the mass loss of 57
giving rise to a peak corresponding to CH2OH at m/e : 31 suggests that the compound in question is a primary
alcohol with formula R—CH2OH, where R ≡ 57. This suggests that the R ≡ 57 ≡ C4 H +9 ≡ CH3 — CH2 — CH2 —
57 57
CH +2 or = 4Cs
C 9Hs or = 4CH 2s + 1H, i.e. R= CH3 — CH2 — CH2 — CH2
12 14
Accordingly, the compound in question is 1-pentanol with the structural formula
CH3— CH2— CH2— CH2— CH2OH
The peaks in the spectrum can be translated by the following mechanisms.
+•
CH3 CH2 CH2 CH2
α
CH2OH ]
+ + •
+ +
29 CH2OH + C4H9 HC O] H2
+ m/e : 31 m/e : 29
43
•
+
CH2OH + C4H9
m/e : 57
+•
H OH
+•
H2C CH2 CH2
(CH2)3 + H2O
H2C CH2 CH2
CH2 m/e : 70
CH3 H +• +
HC OH • CH
CH3 + H2O + (CH2)2
H2C CH2 CH2
CH2 C4H7
m/e : 88 m/e : 55
+•
OH
H
H2C CH2 CH2 + H2O + H2C CH2
+•
H2C CH2 CH2
CH2
CH2
m/e : 42
This lends support to the structure of the compound as 1-pentanol.

+• +•
(B) The peak at m/e : 87 corresponding to ( ) and at m/e : 70 corresponding to ( ) suggest that
+•
the compound in question is an alcohol. The base peak at m/e : 45 and a peak at 73 corresponding to ( )
suggest that the compound is a secondary alcohol with the formula
R CH OH
CH3
Since M +• = 88, R = (88 − 45) = 43 = 3Cs + 7H or = 3CH2's + 1H.

The peaks at m/e : 43 and 29 with mass interval of 14 mass units suggest the presence of the C3H7 group in the
molecule. Thus, the structural formula of the compound is CH3 CH2 CH2 CH OH
CH3
Accordingly, the peaks in the spectrum can be translated as
+•
]
CH3 CH2 CH2 CH OH
• +
CH3 + CH3 (CH2)2 CHOH
CH3 m/e : 73
29 +
CH OH + C3H7
CH3
m/e : 45
• +
CH OH + C3H7
m/e : 43
CH3
CH3
H +•
H2C OH CH
+•
H2C CH CH3 H2O + (CH2)2 CH2
CH2 m/e : 70
• +
CH3 + CH
(H2C)2 CH2
H m/e : 55
+•
H2C OH
+ •
H2C CH CH3 CH + CH3 + H2O
CH2 (H2C)2 CH2
H m/e : 55
+•
H2C OH
H2C CH CH3 H2O + CH2 + •

+ [ CH3 CH CH2 [
H2C
CH2
+•
C3H6
+ •
m/e : 42
This lends support to the structure of the compound as 2-pentanol, a secondary alcohol.
(C) According to the molecular formula of the compound, i.e., C5H12O, the molecular ion peak should appear at
m/e : 88. The absence of a molecular ion peak and the peaks at m/e : 73 corresponding to the M—CH3, and at m/e : 59
corresponding to M +• − 29, indicate that the compound in question is a tertiary alcohol with the structural formula
CH3
R C OH
CH3
where R = 29 ≡ 2Cs + 5Hs or = 2CH2s + lH, i.e R = CH3 — CH2. The peak at m/e : 29 thus corresponds to the C2H5
group. Accordingly, the structural formula of the compound is
CH3
CH3 CH2 C OH
CH3
According to this formula, the peaks in the spectrum can be translated as

+•
CH3
CH3 CH2 C OH
CH3
CH3 •
CH3 + CH3 CH2 C OH
CH3 +
m/e : 73
•
+C OH + C2H5
CH3
m/e : 59
+
C2H5 + CH3
m/e : 29 • C OH
CH3
H
+•
H2C OH
H2O + H2C
H2C C(CH3)2 + • C(CH3)2
H2C
m/e : 70
H
+•
H2C OH
CH2 +•
H2C C(CH3)2 + H2O + (CH3)2C
CH2 m/e : 42
This lends support to the structure of the compound as 2-methyl-2-butanol, a tertiary alcohol.
(b) This is left an exercise for the readers. (See section 11.13)
Problem 11.11: Given below are the characteristic peaks from the mass spectra of (a) benzylalcohol,
(b) phenol, (c) p-methylcresol, and (d) o-ethylphenol.
(a) m/e 108 107 79 77 51
(b) m/e 66 65
(c) m/e 108 107 79 77 51
(d) m/e 122 121 107 104 79 77 51
What ions might these peaks correspond to?
Solution The peaks in the respective spectrum are translated as

+• +
CH2OH CH2O OH
•
H+
+
m/e : 107
m/e : 108
H H
CO +
+
m/e : 79
[C6H7]
+
+
+
[C4H3] + C2H2 H2 + [C6H5]+
m/e : 51 m/e : 77
(a)
+• +•
OH
H H
•
CO + + +H
m/e = 65
Molecular m/e = 66
mass = 94 (M+•)
(b)
OH OH OH
•
H + +
+CH m/e : 107

CH3 2
+•
M = 108
H H
CO +
+
m/e : 79
+ H2
[C6H5]+
m/e : 77
[C4H3]+ +C2H2
m/e : 51
(c)
+• +•
CHCH3 CH2CH3 CH2+
H •
CH3 +
OH OH OH
−H2O m/e : 122
OH
[C6H4CHCH3]+ •
m/e : 104 +
CHCH3 +
•
H+
m/e : 107
OH
H H
CO +
+
m/e : 79
+
H2 +
m/e : 77
+
[C4H3]+ + C2H2 [C6H5]+
m/e : 51
(d)
Problem 11.12: The mass spectrum of a compound exhibits a strong molecular ion peak at m/e : 122 (35 per cent),
the base peak at m/e : 91 and other two peaks at m/e : 92 (65 per cent) and 65 (15 per cent). Metastable peaks appear
at m/e : 46.5 and 69.4 mass units. Deduce the structure of the compound.
Solution The two metastable peaks correspond to the transitions:

+ +
46.5 : 91 → 65 + 26 ( ≡ C2 H 2 )
+ +
69.4 : 122 → 92 + 30 ( ≡ H 2 CO )
The mass loss of 31 from the strong molecular ion peak at m/e : 122 giving rise to a peak at m/e : 91 indicates
that the compound is methyl ether but the strong molecular ion peak suggests that the compound is an aromatic
alcohol and thus the mass loss of 31 is due to — CH2OH but not to because of — OCH3 group. Thus the peak at
CH+2
m/e : 91 is attributed to the benzyl group, i.e. which fragments to yield a peak at m/e : 65 as follows:
CH+2
+ + + CH
m/e : 91 m/e : 65 CH
Ion peak at even mass arising from an even mass molecular ion indicates McLafferty rearrangement reaction.
Thus the peak at 92 arises as
+• +•
CH2 CH2 OH CH2 CH2
O
H
+•
CH2
H
+ H2CO
H
m/e : 92
Thus, keeping in view the mass spectral data in question, the compound is 2-phenyl ethanol with the
structural formula
CH2 CH2 OH
Problem 11.13: Deduce the structural formula of a compound from the mass spectral data indicated below:
m/e 72(M+) 71 57 44 43 29
Peak height (in mm) 25 3 9 34 27 19
Relative abundance 73.5 8.8 26.5 100 79.4 55.9
(as percent of base peak)
Solution Since even molecular mass ion results in even mass base peak, a molecular rearrangement occurs in
the molecule. The strong peak at m/e : 44 is attributed to the aldehyde group (i.e. CHO) in the molecule. This peak
in the aldehyde appears due to an internal molecular rearrangement involving a six-membered cyclic transition
state, i.e. ‘McLafferty transition state’. Thus, there are at least three carbon atoms in the side chain of the aldehydic
group. The peak at m/e : 71 is because of loss of hydrogen from the aldehydic group by a-fission. Accordingly, the
possible structure of the compound in question is
O
R C H
The side chain group must contain at least three carbon atoms, otherwise McLafferty rearrangement will not
be possible.
The mass of side chain group R = 72 − 29 = 43 ≡ 3CH2 s + 1H or 3Cs + 7Hs.

Thus, the side chain group R = C3H7 or the length of the side chain group R can also be deduced from
the peaks at m/e : 15, 29 ad 43. These three peaks with a mass interval of 14 units belong to an ion series of
the type CnH2n+1. Thus the side group R correspond to C3H7=43 mass units. In view of it, the structural
formula of the compound is
O
CH3 CH2 CH2 C H
The mass loss of 15 from the parent molecular ion, i.e. M +• ( CH ) giving rise to a peak at m/e : 57 corre-
O
+
sponding to CH2 CH2 C H lends support to the structure of the compound as butraldehyde.
The modes of fragmentation of the aldehyde in question is summarised below.
+• + +
M − 43= 29 (C—H)
CH3 CH2 CH2 C H
γ β α α O
+• +
O M − H = 71
+• + +
M − 29 = 43 (CH2 C H)
+• + O
M − 15 = 57
+• +•
H H
H2C O CH2 O
+
H2C CH CH2 C
CH2 CH2 H
m/e : 44
Problem 11.14: (a) The characteristic peaks in the mass spectrum of an unknown compound are indicated
below:
m/e 136 135 134 105 91 78 65 51 39 29
(M )
+•
Relative
0.41 5.80 57.4 41 100 50 18 29 23 18
abundance
(Base peak)
What is the likely structural formula of the compound? (b) Based on the following values of m/e : 139, 111, 105
and 77, write the fragmentation pattern of p-chloro benzophenone.
5 80
Solution (a) The number of carbon atoms in the molecule = = 9. So the number of hydrogen
atoms = (134 − 12 × 9) = 26. 0.011 × 57.4
Accordingly, the molecular formula of the compound if it contains C and hydrogen only should be C9H26; but
this is impossible. For a compound containing C, H and O, the molecular formula is C9H10O. The percent abun-
dance of M + 2 peak, i.e. 0.41 reveals that the compound does not contain a sulphur atom.
The strong molecular ion peak at m/e : 134 and a base peak at m/e : 91 show that the compound is aromatic in
nature and has a phenyl ring also. So the formula of the compound can be expressed as
R where R (C9H10O C6H5) C3H5O
The mass loss of 29 giving rise to a peak at m/e : 105 and a peak at m/e : 29 (due to a-cleavage)
g ) suggest
gg that the com-
pound possesses an aldehydic group, i.e. — CHO and the peak at m/e : 78 corresponds to M +• CH 2 CHO . ( )
Accordingly, the structural formula of the compound is
CH2 CH2 CHO
The peaks at m/e : 65, 39 and 51 appears as

CH+2
+•
•
CH2 CH2 CHO CH2CHO +
+ CH
C5H5 + + +
CH
m/e : 65 m/e : 91
+
C3H3 + C2H2
m/e : 39
+• +•
CH2 CH2 CHO CH2 CH3 + CO
+
+ •
C4H3 + C2H2 C6H+5 + C2H5
m/e : 51 m/e : 77
This lends support to the structure of the compound in question.

(b) The peaks characteristic of p-chlorobenzoquinone are translated by the following mechanism.
O +•
C Cl
+
C O + • Cl
m/e : 105
CO + +
m/e : 77
•
+ Cl + CO
m/e : 111
+
O C Cl + •
m/e : 139
+
CO + Cl
m/e : 111
•
+ + O C Cl
m/e : 77
Similarly, we can write the characteristic peaks of other aromatic ketones such as
O O
C C
OCH3
;
Problem 11.15 Deduce the structure of a compound (C6H14O) from the mass spectral data indicated below:
m/e 102 (M+•) 87 59 45 43 27
Peak height (in mm) < 0.5 9 5 50 25 8
Solution: The peaks at m/e : 45, 59 and 87 suggest that the compound in question is either an ether or an alcohol.
Since there is no peak corresponding to loss of a water molecule from the molecular ion, i.e. (M+•—H2O) = 84, the
possibility of alcohol is ruled out. The peaks at m/e : 43 and 59 indicate that the compound contains propyl ester
group (i.e. OC3H7). Accordingly, the formula of the compound is written as
R—OC3H7
R = (102 − 59) = 43 ≡ 3Cs + 7Hs or ≡3CH2s + 1H

So, R = C3H7 and the general structural formula of the compound can be written as
C3H7—O—C3H7
The two possible structural formula corresponding to C3H7—O—C3H7 are
CH3 CH3
CH3 CH2 CH2 O C3H7 HC O CH
CH3 CH3
I II
The absence of peak at m/e : 71 corresponding to M+•— C2H5 rules out the possibility of structure I.
The structural formula II seems to be correct since according to this structure, the peak at m/e : 45 can be trans-
lated by fragmentation followed by the molecular rearrangement indicated below:
H CH2
H 3C CH3
+• • +
HC O CH CH3 + CH3 C O C CH3
H H
H3C CH3
+
CH3 CH OH + CH3 CH CH2
m/e : 45
The fragmentation pattern of the ether in question is summarised below:
H3C CH3
+
HC O CH
CH3 CH3
87 +
89+
+
43+(C3H7) C2H+ +
3 (27 ) + CH4
Problem 11.16: Deduce the structural formula of the compound from the following mass spectral data.
m/e 196/198 (M+•) 167/169 151/153 123/125 103 75 47
Peak height (in mm) 1/1.1 1/1.2 19/18 28.5/26.5 5.0 13 27
Metastable peaks 102.2 100.2 54.7
Solution The metastable peaks in question are attributed to the following transitions:
+ +
⎤
.2 : 153 → 125 28(≡ C2H4 ) ⎥
⎥ (i)
+ +
100.2 : 151 → 123 + 28(≡ C2H4 ) ⎥⎦
+ +
54.7 : 103 → 75 + 28(≡ C2H4) (ii)
The M and M + 2 peaks at 196 and 198 respectively with intensity ratio of 1:1 suggest that the compound
contains one bromine atom. The mass loss of 45 giving rise to a 151/153 doublet with an intensity ratio of 1:1
indicates that the compound contains either a carboxylic group (⎯ COOH) or an ethyl ester group (⎯ OC2H5).
However, the 167/169 doublet, still with an intensity ratio of 1 : 1 that arises because of mass loss of 29 from the
molecular ion, rules out the possibility of the presence of ⎯COOH group in the molecule.
The peak at m/e : 103 in the form of a singlet corresponds to (M+• — CH2Br) ≡ (M+• — 93). This indicates that
the compound contains a CH2Br group. Thus, the structure of the particle corresponding to the peak at m/e: 103
can be written as
R1
+
C OC2H5
R2
m/e : 103
Let us now find the nature of R1 and R2 groups. The mass due to both the groups should be (103 − 57) ≡ 46
46 46
mass unites. So ≡ 3CH2s + 4Hs or ≡ 3Cs + 10Hs, i.e. R1 + R2 ≡ C3H10, but this is not possible. If either
14 12
of the groups contains an oxygen atom then R1 + R2 ≡ C2H6O (since CH4 ≡ O). Since the compound contains an
ester group so most probably R2 = OC2H5 and R1 = H. Thus the compound in question is most likely a double ether
with the structural formula
H5C2O CH OC2H5
CH2
Br
The 123/125 doublet with intensity ratio of 1:1 and a singlet at 75 arise due to transition of (i) and (ii)
respectively, i.e.
+ +
H2C O CH CH2 HOCH
(i) +
H2C H CH2 CH2 CH2
Br Br
m/e : 123/125
+
H5C2O CH O CH2 +
H5C2O CH OH + CH2
H CH2 m/e : 75
(ii) CH2
+
C2H4 + OH CH OH
m/e : 47
This lends support to the structure of the compound as a double ester.

The major routes of fragmentation of the compound have been summarised below.
CH3 CH2 O CH O CH2 CH3
+•
CH2 M C2H5 = 167/169
+• +•
M CH2Br = 103 M OC2H5 = 151/153
Br
m/e : 196/198
Problem 11.17: In the mass spectrum of a compound with molecular formula C4H8O2 there are observed
peaks corresponding to the values of m/e indicated below:
m/e 27 29 39 41 42 43 45 60 73 88(M+•)
Relative content, 39.3 19.8 14.8 23.7 24.7 22.3 19.1 100 27.1 1.6
as per cent of base peak (Base peak)
Solution The mass loss of 45 from the molecular ion peak, i.e. M+• — 45 giving rise to peak at 43+• and a peak
at m/e : 45 indicates that the compound in question is carboxylic acid.
The empirical formula of the acid will be R—COOH, where R = C4H8O2—RCO2H=C3H7 or the mass of the
alkyl part of the acid = 88 — 45 = 43 ≡ 3CH2s + 1 H or ≡ 3Cs + 7Hs. Thus the empirical formula of the com-
pound is C3H7COOH and the peak at 45 corresponding to (M+• —C3H7), i.e. +COOH. The ion at 73 corresponds
• +
to M +• CH 3 i.e. C3H7COOH→ C H 3 + C2 H 4 —COOH. Thus a residual mass due to M+• — 15 − 45 = 28 ≡ 2Cs
+ 4Hs ≡ 2CH2s.
Accordingly, the structural formula of the compound is CH3— CH2— CH2— COOH.
Since an even molecular ion gives rise to an even mass base peak, a molecular rearrangement occurs in the mol-
ecule. The McLafferty rearrangement in this compound yields a peak at m/e : 60. This substantiates the structure
of the compound as butyric acid.
H +• +•
O H
H2C O CH2
+ C OH
H2C C OH CH2
H2C
C
H2 m/e : 60
+•
CH2 C O + H2O
m/e : 42
The other peaks result due to the decomposition of the ion series of the types Cn H +2 n 1 to Cn H +2 n 1
+ + + +
C3H7 C3H5 + H2 C2H5 C2H3 + H2
m/e : 43 m/e : 41 m/e : 29 m/e : 27
+
C3H3 + H2
m/e : 39
The peak at m/e : 39 being intense cannot be the metastable peak for the transition 43+ → 41+ which is expected
41 × 41
at m/e : = 39.09.
43
Problem 11.18: The characteristic peaks in the mass spectrum of l-chlorohexane are indicated below:
m/e 122/120 107/105 93/91 85 79/77 65/63 43
Intensity ratio of peaks in the

1:3 1:3 1:3 — 1:3 1:3 —
doublet
What might be the ions corresponding to these peaks?
Solution The peaks in each of the doublets in the spectrum are two units apart and the intensity ratio of
peaks in each doublet is 1 : 3. This suggests that the doublets in the spectrum results because of the two iso-
topes, i.e. 35Cl and 37Cl of chlorine. The 122/120 doublet corresponds to the molecular ion of the compound
in question.
The following modes of fragmentation give rise to ions corresponding to the other peaks in question:
+•
CH3 CH2 CH2 CH2 CH2 CH2 Cl +
CH3(CH2)3CH2CH2 + Cl•
+
CH3(CH2)3CHCH3
m/e : 85
+
•
C4H9 + CH2 CH2 Cl
+
CH2 CH2 Cl
m/e : 63
• +
C3H7 + CH2 CH2 CH2 Cl
CH2 CH2 CH2 Cl+

m/e : 77
+ •
C3H7 + C3H6Cl
m/e : 43 +
+• Cl
Cl
• H2C CH2
C2H5 CH2 CH2 C2H5 +
H2C CH2
C C
H2 H2 m/e : 91
+
•
CH3 + CH2 (CH2)3 CH2 Cl
+
CH2 (CH2)3 CH2 Cl
m/e : 105
Note that the stability of C4H8Cl+ ion (m/e : 91) is because of a five-membered cyclic structure.
+
C3H7+ CH3 CH2 CH2
+
CH3 CH2 CH3
Problem 11.19: The mass spectra of 3-heptanone (a) and methyl phenyl ketone (b) show peaks at m/e : (peak
height): 114 (M+•; 4 mm), 85 (8 mm), 72 (7 mm), 57 (44 mm) (base peak), 41 (9 mm), 29 (31 mm), 27 (12 mm),
and at m/e : 120 (M+•) 105, 77, 51 respectively. What ions might these peaks correspond to?
Solution (a) The peaks in the mass spectrum of 3-heptanone originate from the following modes of frag-
mentations:
+
O
α α
CH3 CH2 C CH2 CH2 CH2 CH3 m/e: 114
+ +
C2H5C O C2H5 + CO
m/e: 57 m/e: 29
indicates 3-heptanone
+
C2H3 + H2
m/e: 27
+
C4H9 C3H5+ + CH4
m/e: 57 m/e: 41
+ + +
C4H9 C O C4H9 + CO C3H5 + CH4
m/e: 57
m/e: 41
+
+
H CH3 OH
CH3 CH
O C H
C2H5 C +
CH2 CH2 CH2
C2H5 C
C
H2 m/e: 72
m/e: 114
(b) By Table 11.10, we write:

+
O O
C + e C + 2e
CH3 CH3
m/e: 120
+
C O + CH3
m/e: 120
+ +
+
C4H3 + C2H2 + CO
m/e: 51
m/e: 77
Problem 11.20: The mass spectrum of a compound contains a molecular ion peak at m/e : 130(M+•) plus
peaks at 71, 70, 59, 57.5, 44, 42, 28 and 14. Deduce the structure of the compound.
Solution The ion at m/e : 59 and another ion corresponding to (M+• − 59), i.e. m/e : 71 suggest that the compound
in question is an ester and contains an acetate group. The ions at m/e : 70, 42, 28, 14, belong to the ion series of the
70
type Cn H +2n• . This suggests that the hydrocarbon component of the compound contains at least = 5 carbon atoms
12
O
70
or = 5 CH2s. Accordingly, the formula of the compound can be written as CH3 C OR where, R contains at
14
least five carbon atoms.
Now R = 130 − 59 = 71 = 5Cs + 11Hs or ≡ 5CH2s + 1H, i.e R = C5H11.
Since, there is no ion corresponding to (M+•—CH3), i.e. m/e : 115, the presence of methyl group in the side
chain of the hydrocarbon part of the molecule is ruled out.
Thus, the compound in question is amyl acetate with structural formula

O
CH3 C OC5H11
Ion peak at m/e : 70, an even mass arising from an even mass molecular ion suggests a McLafferty rear-
rangement, i.e.
+•
(CH2)2 CH3
H
O C H +•
CH3COOH + C5H10
CH3 C CH2
O
This further lends support to the structure of the compound in question. The other ions are formed by the
following routes of fragmentations:
+•
C3H6 + C2H4
+• m/e : 42
C5H10
m/e : 70 +•
C3H6 + C2H4
m/e : 28
+ •
CH2 + CH2
m/e : 14
+• •
1/3(CH3COOC5H7)3+ + 4H• [CH3COOC5H11] CH3 + 1 (COOC5H11)++
2
m/e : 42 m/e : 57.5
The peak at m/e : 44 is attributed to +•CO2 and CO+ ion also contributes to peak at m/e : 28
Problem 11.21: The mass spectrum of the compound with structural formula
CH3
CH3 (CH2)17 C CH2 COOCH3
CH3
exhibits peaks at m/e : 74, 115, 295 and 368. What ions might these peaks correspond to?
Solution The molecular ion corresponding to the compound in question will appear at m/e : 368 while the
peaks at m/e : 295 and 115 are due to cleavage at the quaternary carbon atom on either side.
+•
CH3
CH3 (CH2)17 C CH2 COOCH3
C
H3
m/e : 368
CH3
•
CH3 (CH2)17 C+ + CH2 COOCH3
CH3
m/e : 295
• +
CH3 (CH2)17 + (CH3)2 C CH2COOCH3
m/e : 115
+•
OH
CH2 C
OCH3
The peak at m/e : 74 is attributed to which is translated by McLafferty rearrangement as
follows:
+•
(CH2)16 CH3 +•
H OH (CH2)16CH3
O C H
C + CH
CH3O C C(CH3)2 H3CO CH2
C(CH3)2
C
H2 m /e : 74
Problem 11.22: The mass spectral data of a compound with molecular formula C9H12S is indicated below:
m/e 152 (M+•) 137 110 77 66 65 43

(Base peak)
Abundance,
45 7 100 7 11 8 12
as per cent
Metastable peaks are observed at m/e : 123, 79.6 and 54.1. Deduce the structure of the compound.
Solution The metastable peaks are due to the following transitions.
123 152+ → 137+ + 15 (CH3)

79.6 152+ → 110+ + 42 (CH3⎯CH=CH2)
54.1 110+ → 77+ + 33 (HS)
The strong molecular ion peak suggests the presence of an aromatic group. The transition states suggest that
the mass loss is due to elimination of CH3, CH3⎯CH=CH2 and HS groups respectively. The loss of HS from base
peak indicates that ⎯ SH is bounded to phenyl ring. In view of it, the structural formula of the compound may
be written as
S R
where, R = C9H12S ⎯ C6H5S = C3H7 ≡ CH3⎯CH⎯CH3.

The initial loss of CH3 group indicates R to be isopropyl group. Accordingly, the structure of the compound is
S CH CH3
CH3
Even mass molecular ion resulting in even mass base ion suggests molecular rearrangement in the molecular
ion. Thus, the base peak at m/e : 110 appears due to such a process as follows:
+•
H CH2
S CH CH3 +• SH + CH3 CH CH2
m/e : 110
•
+ SH
m/e : 77
+• SH
+•
H H
− H•
CS + +
m/e : 65
m/e : 66
This further lends support to the structure of the compound in question which is isopropyl phenyl thioether.
The peaks in the mass spectrum appear as follows:
• +
+• S + CH3CHCH3
S CH CH3 m/e : 43
CH3 +
•
CH3 + SCH CH3
m/e : 137
Problem 11.23: The mass spectrum of nitrobenzene shows the presence of particles with the following
mass numbers: 123 (M+•, prominent), 93, 77, 65, 51, 46 and 30. Based on this data, what would be the modes
of fragmentation of nitrobenzene? (Use Table 11.10.)
Solution The modes of fragmentation of C6H5NO2 giving rise to the peaks in question are
+• +•
NO2 O N O O+
•
+ Rearrangement
+ NO2 + NO
•
m/e : 46
m/e : 123 m/e : 93
O•
+
• + + CO + +
NO2 + (C6H5) NO +
m/e : 30 m/e : 65
m/e : 77
+
C4H3 + C2H2
m/e : 51
Problem 11.24: The mass spectrum of a compound exhibits peaks at m/e : 129, 114, 86, 58 and 44.
Deduce the structure of the compound if it shows metastable peaks at m/e : (a) 29.5 and 16.9 and (b) 16.9 mass units.
Solution (a) The metastable ions at 29.5 and 16.9 are translated by the following transitions:
29.5 : 114+ → 58+ + 56 (≡ C4H8)

16.9 : 114+ → 44+ + 70 (≡ C5H10)
The transition states suggest that the mass loss is due to elimination of C4H8 and C5H10 groups respectively.
The odd mass molecular ion peak at m/e : 129 and peaks at m/e : 58 and 44 suggest that the compound in question
CH3
+ + +
is an aliphatic mono-amine. The peaks 44+ and 58 correspond to CH2 = N H and CH NH respectively. Conse-
|
CH3 CH3
quently, the general structure of the compound may be written as
CH3
R CH N R⬘
CH3
A mass loss of 43 from the molecular ion, i.e. (M+• —C3H7) suggests that the side chain group R is C3H7.
Accordingly, the structure of the compound is written as
CH3
C3H7 CH N R⬘
CH3
So, R′ = 129 − 100 ≡ 29 = 2Cs + 5Hs or ≡ 2CH2s + 1H, i.e. R′ = C2H5.

Thus, the structural formula of the compound is

CH3
CH3 CH2 CH2 CH N CH2 CH3
CH3
A mass loss of 15 from the molecular ion, i.e. (M+• — CH3) giving rise to a peak at m/e : 114 lends support to
the structure of the compound.
The fragmentation of the molecular ion followed by molecular rearrangement of the products of fragmentation
results in ions at m/e : 58 and 44, i.e.
CH3
+•
CH3 CH2 CH2 CH N CH2 CH3
CH3
CH3 H CH2
•
C3H7 + CH N+ CH2
CH3
m/e : 86
CH3
+ CH2
CH NH +
CH2
CH3
m/e : 58
CH2 H
+ •
C3H7 CH N CH2 + C H3
CH3
m/e : 114
+
C5H10 + HN CH2
CH3
m/e : 44
CH3 H CH2 CH3

+ + CH2
•
CH3 + CH2 CH2 CH N CH2 CH NH + 2
CH2
CH3 CH3
m/e : 58
(b) In this case, the transition 29.5: 114+ → 58+ + 56 is not allowed, so the peaks at m/e : 58 and 86 will not
appear in the spectrum. Since the transition 16.9: 114+ → 44+ + 70 is allowed, the peaks at m/e : 44 and 114 will
appear in the spectrum. Accordingly, the structural formula of the compound should be
R CH2 N CH2 CH3
CH3
Thus R = 129 − 72 = 57 ≡ 4CH2s + 1H or ≡ 4Cs + 9Hs = C4H9.

So the structural formula of the compound is
C4H9 CH2 N CH2 CH3
CH3
According to this formula, the peak at m/e : 44, will be translated by the following routes of fragmentations
followed by molecular rearrangement.
+•
C4H9 CH2 N CH2 CH3
CH3
H CH2 m/e : 129
+
•
C4H9 + CH2 N CH2
CH3
m/e : 72
+
CH2 N H + C2H4
CH3
m/e : 44
• +
CH3 + C5H11 N CH2
CH3 m/e : 114
H2C H
+
(CH2)4 N CH2
CH3
+
C5H10 + HN CH2
CH3
m/e : 44
Problem 11.25: (a) In the mass spectrum of benzamide C6H5CONH2, there are observed peaks corresponding
to the following values of m/e : 121, 105, 77, 44. What ions might these peaks correspond to? (b) The mass spec-
trum of an amide with molecular formula R—CH2CH2CH2CONH2 includes lines corresponding to the following
mass numbers (m/e): 86, 59, 44. What particles might these peaks correspond to?
Solution (a) Loss of NH2 from molecular ion gives rise to a resonance stabilised benzoyl cation that in turn
fragments to a phenyl cation.
• +
•O •
• + •
C6H5 C NH2 NH2 + C6H5C Q•
m/e : 121 m/e : 105
+ +
C6H5 + CO
m/e : 77
The loss of phenyl radical yields a peak at m/e : 44.

O
+• • +
C6H5 C NH2 C6H5 + CONH2
m/e : 44
(b) The peak at 59 in primary amides is because of the molecular rearrangement.

+•
H +•
H H
O CHR CH2 •O • O+
• • •
+
H2N C CH2 CHR C C
NH2 NH2
C
CH2 C
H2
H2
m/e : 59
The peak at m/e : 44 is due to the loss of alkyl part of the amide.
+•
RCH2—CH2—CH2— CO N H 2 → RCH 2 CH 2 CH •2 + CONH +2
m/e :44
The peak at m/e : 86 arises from g d C—C cleavage accompanied by cyclisation.

H2 H2
C C
•
H2C C O R + H2C C O
R CH2 NH2 H2C NH2

+• +
m/e : 86
Thus, the peaks at m/e : 44, 59, 86 are characteristic of the primary amides.
Mass spectral data of an

11.15 COMPUTERS IN MASS SPECTROMETRY unknown compound
Modern spectrometers are fitted with computers and they play an important role in assist-
ing the mass spectroscopists in interpreting the mass spectral data. The spectra of a large Preliminary inference
marker
number of compounds are known to us and to match the unknown with the known would
be a tedious problem. Computers can be used for this purpose. Even then if all the peaks
are to be matched then it will consume a lot of computer time and the process becomes
expensive. It is advisable to select five or six largest peaks and try to match them with the Functional Functional
mass data of the compound stored in the memory of the computer. Very often the large groups groups
present absent
peaks are to be found at the low mass region, which is not the ideal region to compare
for identification. It is better to choose these peaks at different mass regions. These days
softwares are available that can recognise patterns of structure like alcohol, ketone, ester,
Structure
etc., by their mode of fragmentation. Softwares are also available to interpret the spectra generator
of specific type of compounds like the sequence of amino acids in polypeptides and
proteins as well as the sequence in oligonucleotides. A schematic diagram for structure
determination of an unknown compound from its spectrum is shown in Fig. 11.26. List of possible
candidates
11.16 APPLICATIONS OF MASS SPECTROMETRY

Predictor applied
to each candidate
MS is a versatile tool which has contributed to the development of many sciences and
found applications in all branches of sciences and industries. It was originally used for
accurate measurements of atomic masses and isotopes of elements and by the early List predicated mass
1960s the ever-increasing precision in measurement of isotopic masses and abun- spectra
dances gave rise to new international standards for atomic masses of elements. Since
1960, the applications of this technique blossomed like a mushroom cloud. The period Consistency
of determination of purity, accurate molecular mass and molecular formula of com- check
pounds have been reduced from years to days. Sample size, as small as, sub-ngm, can
be handled by this technique. The technique has played a significant role in material List of candidates whose
analysis and process monitoring in the petroleum, chemical and pharmaceutical indus- predicted spectra are
tries and is being used in food processing and electronic industries, e.g. the petroleum consistent with the original
spectrum
chemists use this technique for grading a complex mixture of hydrocarbons. The tech-
nique is widely used by forensic scientists to know about the nature of poison given
to the victim; in forensic science, the size of the sample under examination is usually Score function
small and here the sensitivity and specificity of mass spectrometry are very important.
Spark source mass spectrometry has been utilised for the detection of trace elements
List of candidate structures
in liver tissue (especially arsenic, mercury, lead, etc.), hair, glass, metal fragments and ranked from most to
synthetic fibres. Organic mass spectrometry is used for the identification of drugs, least preferred
e.g. strychnine, drug metabolites and other materials including pesticides, namely Fig. 11.26 Flowsheet diagram
dichlorodiphenyltri-chloroethane (DDT), 1-naphthyl-methylcarbamate (Serin), etc., for structure determination of
and the pyrolysis products of polymers, by sport scientists to know about the banned an unknown spectrum.
performance-enhancing drugs used by the athletes from their urine samples, by medical scientists for analysis of
drugs like lysergsaüre diethylamid (LSD, C20H25N3O, 323) in illicit preparation and morphine (C17H19NO3, 285.35) in
the urine of heroin addicts, by environmental scientists to check organic and inorganic polluting agents, etc., in water
and in the atmosphere. Mass spectrometry is also used in geology to have a better estimate of ages of rocks on earth
and other planets. MS, being non-invasive (thus politically viable), is used for international monitoring of nuclear
facilities. The technique has proved to be very useful in the identification and analysis of biological macromolecules
including oligonucleotides and proteins; the main actors in the make-up of life.
The applications have been discussed here under two subtitles, namely nonbiological and biological applications.
However, there may be some overlap between these two classes of applications.
11.16.1 Nonbiological Applications of Mass Spectrometry

Some of the typical nonbiological applications of this technique are the following:
(a) Leak Detection Inexpansive and portable helium mass spectrometers have been widely used as leak
detectors and as the most sensitive gauges for the most extreme vacuum we can produce. The sensitivity of the
helium mass spectrometer is of the order of 10−11 atm ml/s of helium and the exact leak locations within 1.0 mm
can be detected. The instrument is coupled to a vacuum system under examination and a fine jet of helium is
sprayed around any joints or other locations suspected of leaking. Helium entering the system through the leak-
ing location, if any, is drawn into the instrument for detection and the exact location of the leak is determined
for repair. The major merit of this method is that the entire system is scanned first for locating the exact leaking
points and the repair work is done after the scanning.
(b) Identification of Molecular Species from Fragmentation Pattern Each compound has a characteristic
mass spectrum which can be used as a ‘molecular fingerprint’ alike infrared spectrum. In other words, we can say that
like infrared spectrum, mass spectrum also enables to discern which peaks are of importance in identifying the dif-
ferent molecular systems. Even positional isomers can be distinguished from their mass spectra. This is demonstrated
here on the basis of mass spectra of the two isomeric butanes, viz., n-butane and isobutane shown in Figs 11.27(a)
and (b). Both the spectra have the same number and positions of peaks, except that they differ in their relative intensi-
ties. The molecular ion occurs at mass number of 58 for each isomer, but in each case it is not the most abundant ion.
We know that molecular ions are odd electron species, and because of energetic considerations they have the general
tendency to become even electron ions. The most abundant ion for the butanes occurs at a mass number of 43; and is
attributed to C3H7+ ion, i.e. (even number of electrons):
•
C3H7 CH3 + e C3H7+ + CH3 + 2e
m/e : 43
or alternatively, we can write
+• •
C4 H10 → C3 H +7 +CH 3
(m/e :58) (m /e :43)
The low-intensity peak at a mass number of 15 in each spectrum corresponds to +CH3 which also has an even
number (6) of outer shell electrons. This reveals that the fission of a methyl-carbon bond occurs less frequently.
• +
C3 H 7 − CH 3 +e → C3 H 7 + CH3 + 2e
+• • +
or C3 H 7 − C H 3 ⎯ ⎯
→ C3 H 7 + C H 3
m / e = 58 m / e = 15
The fairly strong peak at a mass number of 29 in the straight-chain isomer is due to the fission of an ethyl–ethyl
bond:
•
→ C H 5+ + C2 H 5
C2 H 5 − C2 H +e ⎯⎯ 2e
m / e: 29
+
n-butane C3H 7
CH3CH2CH2CH3
Relative number if ions C2H+5
+•
M
+
CH 3
0 10 20 30 40 50 60
Mass number
(a)
Isobutane C3H+7
CH3 CH CH3
Relative number if ions
CH3
C2H+3
CH+3 M+ •
0 10 20 30 40 50 60
Mass number
(b)
Fig. 11.27 Mass spectrum of (a) n-butane, and (b) isobutane.
A peak at a mass number of 29 is hardly evident in the spectrum of a branched-chain isomer. This shows the
absence of an ethyl group in the branched chain isomer; only rearrangement of the structure of isobutane can
yield at peak at a mass number of 29. The strong peak at a mass number of 27 in the mass spectrum of isobutane
is attributed to C2H3+ which is formed according to the following bond fissions:
+ •
CH3 CH CH3 + e CH3 CH + CH3 + 2e
CH3 t CH3
rec
Di ) (m/e : 43)
H4
(−C
+ • + •
−H
C2H3 CH3 CH (+CH3)
(m/e : 27) (m/e : 28)
Thus, the comparison of mass spectra of n- and iso-butanes shows that the peaks at m/e : 29 (C2H5+)
and m/e : 27 (C2H3+) respectively can be fixed a criterion to distinguish between n-butane and isobutane.
Similarly, a comparison of the mass spectra of the four isomeric butyl alcohols indicate that the base peaks
(except that of the iso-isomer) at m/e : 3 1 (CH3O+), 45 (C2H5O+), 59 (C3H7O+), and 74(C4H10O+•, molecular
ion) respectively can serve to identify the normal, i.e. primary, secondary, tertiary and isobutyl alcohols.
The characteristic peaks in the low-pressure mass spectrum of methane are observed at m/e : 16, 15, 14, 13,
12 and are respectively attributed to CH4+•, CH3+, CH2+•, CH+ and C+• ions. While the peaks at m/e : 17, 19, 41 in
the high-pressure mass spectrum of CH4 are due to the following ion–molecule reactions occurring by collision
of methane molecule with ions.
CH+•
4 → CH 3+ + H • ⎤
⎥
• ⎥
CH +•
4 + CH 4 → CH 5+ + C H 3 ⎥
m/e:17(Base peak,
7.0mm
m ) ⎥
⎥Under high pressure
+
CH 4 + CH 3 → C2 H5 + H 2
+ ⎥
m/e:29( 4.5 mm ) ⎥
⎥
CH 4 + C2 H 5+ → C3 H 5+ + 2H 2 ⎥
m/e:41(0.70
1 mm ) ⎥⎦
The mass spectra of aldehydes, ketones and esters have common features. A comparison of the intensity of the peak
due to a-cleavage with that of the a ′- cleavage reveals that the latter is preferred with the loss of larger alkyl radical
than the former. Thus, the peaks due to a and a ′ fragmentations in simple aldehydes, ketones and esters can be fixed a
criterion for their identification. For example, the peaks at m/e : 58, 57 and 29 are characteristics of propionaldehyde.
+•
O+ O O+
α⬘ α •
•
C2H5 + C H C2H5 C H C2H5 C+H
m/e : 29 m/e : 58 m/e : 57
+
O O+• O+
•
•
C2H5 + C CH3 C2H5 C CH3 C2H5 C + CH3
m/e : 43 m/e : 72 m/e : 57
+ +•
O O O+
•
•
C2H5 + C OCH3 C2H5 C OCH3 C2H5 C + OCH3
m/e : 59 m/e : 88 m/e : 57
Identification of Optical Isomers Optical isomers can be identified by GC-MS. For example, the GC-QPMS
of a sample of mixture of the two optical isomers of the N-acetyl derivatives of the amino-acid alanine with the
L-form labelled with an atom of deuterium is shown in Fig. 11.28. The upper trace containing two peaks is the
normal signal from the GC detector. From this trace alone it is not possible to make out which peak is due to which
compound. The lower trace is the mass scan of the sample at a 10 second interval. The presence of two peaks in the
spectrum at m/e : 159 (intense) followed by a satellite at 158 and 158 (intense) followed by a satellite at 159, clearly
shows the presence of two optical isomers in the sample; the former is attributed to the L-form while the latter to
the D-form of the acid. Based on this, one can assign the signals in the GLC trace to the two optical isomers. The
first signal corresponds to L-form while the second to the D-form.
GLC output signal
M = 159 + Trace 158
M = 159 + Trace 158
O CH3 +
+
O CH3
CH2 CH C NH CH COOCH3
CH2 CH C NH CH COOCH3
H
D
Quadrupole
158
159
output signal
159
158
300 sec 250 200 150 100 50

Fig. 11.28 GC-Quadrupole mass spectrum of N-acetylated derivatives of amino acid alanine with the L-form labelled
with an atom of deuterium. No helium separator is used between GLC and quadrupole mass spectrometer. The satellite
peaks being of very low intensity are not visible here in the spectrum.
(c) Determination of Atomic Masses of Elements Most samples of elements in nature are a mixture of
isotopes. MS has proved to be an ideal method for separation of isotopes. The isotopes of many elements have
been discovered by this means and their isotopic masses measured. The existence of isotopes of elements explains
the markedly non-integral nature of some atomic masses.
The average atomic masses of polyisotopic elements are determined from the peak heights corresponding to the
respective isotopic mass numbers in the mass spectrum. From the peak heights in the mass spectrum, the percentage of
relative abundance of isotopes and the number of atoms in 100 atoms of the isotopic mixture are calculated. The relative
isotopic mass contributions to a sample of 100 atoms of the isotope mixture are then calculated from the formula.
Mass unit of an isotope in 100 atoms of the isotopic mixture = Isotopic mass number × number of atoms of the
isotope in 100 atoms of the isotopic mixture.
The addition of mass units of constituent isotopes in 100 atoms of isotopic mixture will yield the total mass
units for 100 atoms which on division by 100 will give the average atomic mass of the element.
Atomic masses of some polyisotopic elements are listed here in Table 11.11 for ready reference.
Table 11.11 Exact atomic masses of principal polyisotopic elements.
Isotope Mass Isotope Mass
1
H 1.00783 31
P 30.9738
2
H 2.01410 32
S 31.9721
l2
C 12.00000 (std) 33
S 32.9715
13
C 13.00336 34
S 33.9679
14
N 14.0031 35
C1 34.9689
15
N 15.0001 37
Cl 36.9659
16
O 15.9949 56
Fe 55.9349
17
O 16.9991 79
Br 78.9183
18
O 17.9992 81
Br 80.9163
19
F 18.9984 127
I 126.9045
28
Si 27.9769
Problem 11.26: The mass spectrum of naturally occurring lead shown in Fig. 11.29 yields the following data.
Detector current
203 204 205 206 207 208

Mass number
Fig. 11.29 Mass spectrum of naturally occurring lead.
Isotopic mass number Peak height × 10 (in mm)
204.0 2
206.0 24
207.0 22
208.0 52
Calculate the average atomic mass of lead.
Solution Following the general procedure for the determination of average isotopic atomic masses of polyiso-
topic elements, we write
Isotopic mass Per cent relative Number of atoms Mass units of isotopes
number abundance in 100 atoms of the in 100 atoms of the mixture
mixture
204.0 2 2 408
206.0 24 24 4944
207.0 22 22 4554
208.0 52 52 10816
Thus, the total mass units of isotopes in 100 atoms of the isotopic mixture
= 408 + 4944 + 4554 + 10816 = 20722
20722
Accordingly, the average isotopic atomic mass of lead = = 207.22
100
Problem 11.27: From the mass spectrum of nickel in Fig. 11.30, determine the atomic mass of nickel and
compare it with the periodic table value (i.e. 58.933). 11.2
Solution From Eq. (11.35); we write

Relative intensity
Relative intensity of an isotopic peak × 100

Peak height =
Sum off the isotoopic relative intensities
4.4
The relative intensity data at various mass numbers shown in
Fig. 11.30 is converted into peak-height data by making use of
the formula (11.35). The resultant peak height data at different 0.6
0.2 0.1
isotopic mass numbers so determined are
58 60 62 64
Isotopic mass number Relative intensity Peak height Mass number
58 11.2 67.87 Fig. 11.30 Mass spectrum of naturally occurring nickel.
60 4.4 26.66
61 0.2 1.21
62 0.6 3.63
64 0.1 0.6
Sum = 16.5
Proceeding as in Example (11.26), the average isotopic atomic mass of Ni comes out to be 58.733 amu.
The table value of atomic mass of Ni is 58.933 amu. Thus the difference between the present and the
table value
0 2 × 100
= 0.2 amu or = = 0.3 per cent
58.933
(d) Determination of Percentage Content of Xm* in a Diatomic Gas (say X2) Sample Containing Isoto-
pic Binary Mixture of the type XmXm and XmXm*
Let relative abundance of XmXm in sample of air (X2) = h1
and relative abundance of XmXm* in the air sample = h2
h
Thus, the relative abundance of Xm* in the air sample = 2
2
Accordingly, the per cent content of Xm* in the binary mixture
h2
× 100
100 h2 100 100
x= 2 = = = (11.38)
h2 2h1 h2 2( h1 / h2 ) 1 2R 1
+ h1
2
h1
where, R = is the relative abundance of XmXm and XmXm* in the air sample.
h2
Further, if a diatomic gas X2 contains three isotopic species of the type XmXm, XmXm* and Xm*Xm* and their
*
abundances
are respectively h1, h2 and h3; then it can easily be proved that the percentage content of the element X m is given by
( h2 + 2h3 ) × 100
2( h1 + h2 + h3 ) (11.39)
2
Problem 11.28: The mass spectrum of oxygen produced from heavy water is characterised by the presence
of peaks of the isotopic forms of oxygen 16O16O and 16O 18O with wave heights of 101 and 25 mm, respectively.
Determine the percentage content of 18O in the mixture.
Solution The content of nitrogen 18O is determined from the formula
100
x= (11.38)
2R + 1
where R is the ratio of the peak heights of the isotopes of the element being determined in the mass spectrum.
h16O 101
R= = = 4 04
h18O 25
100 100
Hence, x= = = 11.06 per cent
2 × 4.0
04 + 1 9.08
Problem 11.29: The mass spectrum of nitrogen produced from the air is characterised by the presence of
peaks at m/e = 28, 29 and 30 having peak heights (i.e. height × 10) 980, 360 and 53 mm respectively. Identify the
isotopic species and calculate the percentage content of the element 15N in the gas.
Solution The mass numbers at m/e = 28, 29 and 30 correspond to the isotopic species N14N14; N15N14 and N15N15
respectively. Substituting the data given in the problem into the general formula (11.39), we obtain percentage
content of element 15N
( ) × 100 466 × 100

= = 19.21
2( ) 2 × 1213
(e) Determination of Molecular Mass Molecular mass is a useful quantity for the identification of compounds,
and it can easily be determined by mass spectrometry by measuring the intensity of the molecular ion peak as a func-
tion of the energy of the impact electrons. Intensity of molecular ion peak goes on increasing as the energy of the
bombardment electron beam is increased. It is assumed that no ions heavier than the molecular ion will be produced
by the ionising voltages ranging from 9 to 14 V. Thus the mass of the heaviest ion except that of the isotopic molecular
ion gives the molecular mass of the molecule under study. The technique is especially useful where conventional
methods fail to give reliable results as in the case of C6F8Fe(CO)3 and Fe(CO)4(CF2)4. These compounds have been
shown to be monomers from their mass spectral data. The molecular ion peak of the latter has been observed at m/e:
368. Furthermore, the molecular mass of a compound in a mixture can also be determined provided the components
of a mixture have different ionisation potentials, e.g. in a mixture of hydrocarbons, the proper selection of the ionisation
voltage will provide the mass spectrum of the desired component. n-parafins, olefins and aromatics have ionisation
potentials respectively over a voltage range of 10.2−11.2 eV, 9.4−9.7 eV and ≥9.7 eV.
The isobaric compounds, i.e. compounds having the same mass numbers, can easily be identified from their
exact molecular masses as determined from their high-resolution mass spectra. For example, the exact masses
of some molecular species with mass number 28, viz., H2CN, CH2==CH2, N2 and CO respectively are 28.0187,
28.0313, 28.0061 and 27.9949.
(f) Determination of Molecular Formula The mass defect is the difference between the accurate mass and
the mass number. The mass defect of a molecular ion can be determined accurately by high resolution (i.e. double
focussing) spectrometers. This defect enables one to determine the molecular formula directly. Let us suppose
that the measured molecular mass of the compound is 60.0324 and the molecule does not contain sulphur and
halogens which can be detected by their isotope peaks. The following compounds have mass number 60.
CH2NO2 60.0085 C2H6NO 60.0449

CH4N2O 60.0324 C2H8N2 60.0688
CH6N3 60.0563 C3H8O 60.0575
C2H4O2 60.0211
A simple examination of the table shows that the assumed compound has a molecular formula CH4N2O. Tables
due to Beynon are available for each mass number, the possible formula corresponding to the general formula
and their mass defects. They can be referred or they can be computed. This type of analysis can be applied to the
fragment ions and also their compositions can be determined.
Beynon Method for Computation of Abundance Parameters The number of possible formulas can be
restricted on the basis of relative abundance of natural isotopes of different elements (Table 11.12) and at masses
1 and 2 or more units larger than the parent ion. The calculations are made for all possible combinations of natu-
rally occurring heavy isotopes of the elements and the values so obtained are compared with the observed values
in order to choose the molecular formula of the compound.
The per cent of heavy isotopic contributions from a peak at mass M to the peak at mass M + 1 is calculated
from the formula
( )100
= 0.015x + 1.1w + 0.37 y + 0.037z (11.40)
M
and to the peak at mass M + 2 by
( )100
= 0.20 x + 0.006w ( ) + 0.004w y + 0.0002wx (11.41)
M
Table 11.12 Natural abundance of principal polyisotopic elements.
Element Natural abundance (as per cent) Element Natural abundance (as per cent)
1
H 99.985 28
Si 92.18
2
H 0.015 29
Si 4.71
12
C 98.892 30
Si 3.12
32
S 95.02
13
C 1.108 33
S 0.76
14
N 99.63 34
S 4.22
l5
N 0.37 35
Cl 75.53
16
O 99.76 37
Cl 24.47
l7
O 0.037 Br
79
50.52
18
O 0.204 Br
81
49.48
Natural abundance based on 100 atoms of most common isotope can also be calculated and reported, e.g.
1.108 × 100 4.22 × 100
for 12C = 100; 13C= = 1.120; for 32S =100, 34S = = 4.441
98.892 95.02
where w, x, z, and y are the numbers respectively of C, H, O and N in the compound CwHxOzNy. Based on these for-
mulas, Beynon calculated the abundance parameters for all combinations of C, H, N and O up to mass 500. For ready
reference, contributions of C, H, O and N to parent peaks of masses 72 and 73 are reported here in Table 11.13. Thus
the most probable formula of such compounds can be obtained from the relative heights of isotopic peaks M + 1 and
M + 2. For example, suppose we have 100 molecules of methane (CH4). There will be 1.11 molecules that contain 13C
and 4 × 0.015 molecules that contain 2H. These heavier isotopes should contribute an M+• + 1 peak whose intensity
is about 1.11 + 4(0.015) = 1.17 of the intensity of the molecular ion peak. This conforms to the observed intensity of
the M+• +1 peak in the observed spectrum of methane, i.e. P(12C1H4) : P(13C2H4) : : 100 : 1.15.
Table 11.13 Heavy isotope contributions to parent peaks of masses 72 and 73.
M+• Empirical Per cent of M +• M+• Empirical Per cent of M +•
formula intensity formula intensity
M +• + 1(M +• + 2) M + 1(M +• + 2)
+•
72 CH2N3O 2.30 (0.22) 73 CHN2O2 1.94 (0.41)
CH4N4 2.67 (0.03) CH3N3O 2.31 (0.22)
C2H2NO2 2.65 (0.42) CH5N4 2.69 (0.03)
C2H4N2O 3.03 (0.23) C2HO3 2.30 (0.62)
C2H6N3 3.40 (0.04) C2H3NO2 2.67 (0.42)
C3H4O2 3.38 (0.44) C2H5N2O 3.04 (0.23)
C3H6NO 3.76 (0.25) C2H7N3 3.42 (0.04)
C3H8N2 4.13 (0.07) C3H5O2 3.40 (0.44)
C4H8O 4.49 (0.28) C3H7NO 3.77 (0.25)
C4H10N 4.86 (0.09) C3H9N2 4.15 (0.07)
C5H12 5.60 (0.13) C4H9O 4.51 (0.28)
C4H11N 4.88 (0.10)
C6H 6.50 (0.18)
This means that the number of carbon atoms in a compound can be calculated if the relative intensities of both
the M and M + 1 peaks are known.
Relative intensity of M + 1 Relative intensity of M + 1
peak as per cent of base peak peak as percent of M
umber of carbon atoms = or = (11.42)
0 01 1× Relative intensity of 1.1
M peak as pper cent of base peak
This formula works because 13C is the most important contributor to M+• + 1 peak and the approximate natural
abundance of 13C is 1.1 per cent.
In general, the number of isotopic peaks is one more than the number of carbon atoms of the particular ele-
ment spaced at an interval of one mass unit and the intensities of the peaks correspond to the coefficients of the
binomial expansion (a + b)n, i.e.
n(( )an 2b 2 n(n−1)( )an −3b 3 ... n (11.43)

( ) n = an + nan 1b + + + +b
2 3
where, a is the relative abudance of light isotopes, b is the relative abundance of heavy isotopes and n is the
number of atoms of the particular element present in the molecule.
The number of peaks expected = n + 1. In order to obtain the intensities of the peaks, the terms of the binomial
expansion in which the sum of the superscripts is the same are collected and arranged in the ascending order of
the masses. Accordingly, for n = 1, number of peaks will be 2 with intensity ratio : 98.89 : 1.11 or 100 : 1.12. For
n = 2 carbon atoms.
Number of peaks expected = 2 + 1 = 3
The intensities of the peaks can be worked out by applying the binomial expansion (11.43), i.e.
(98.89 12
C + 1.11 13
C)2 = 9.77 × 103 (C12)2 + 2.19 × 102 (12C13C) + 12.32 (13C)2
= 9.77 × 103 24C + 2.19 × 102 25C + 12.32 C26
If M is the mass of the molecular ion, i.e. (C12)2 = M then the above expression can be written as
(98.99 12C + 1.11 13

C)2 = 9.77 × 103 M + 2.19 × 102 (M + 1) + 12.32 (M + 2)
The relative intensity of the peaks will be,
Peak at m/e: M M+1 M+2
Relative intensity 9.77 × 103 2.19 × 102 12.32
Relative intensity 100 2.24 0.126
as per cent of M
2.24
According to formula (11.42), the number of carbon atoms = = 2.
1.11
The presence of Cl, Br, S and Si can be deduced from the unusual isotopic abundance patterns of these
molecules. These and other elements such as P, F and I are also detectable from the unusual mass differences that
they produce between some fragment ions in the spectrum.
The presence or (absence) of chlorine (33 per cent), S (4.4 per cent), Br (98 per cent) or Si (3.35 percent) is
indicated by the sharp peaks at mass M + 2 and M + 4, etc. Just like carbon isotopes—the number of isotopic peaks
is one more than the number of atoms of the particular element, i.e. Cl, Br, Si or S-but spaced at intervals of two
mass units and the intensities of the peaks correspond to the coefficients of the binomial expansion. When two
elements are present in the molecule, the binomial expansion used in (a + b)n (c + d)m. In such a case, the number
of peaks = n + m + 1; and the intensities of the peaks can be worked out by the rules applied to the binomial expan-
sion (a + b)n, the peak of the highest mass may sometimes be too weak to be detected. Let us now calculate the
intensity ratio of the molecular ion clusters owing to halogen isotopes.
For n = 3 chlorine atoms in the molecule, the number of peaks expected = 4, and the binomial expansion is
(3 35Cl + 37Cl)3 = (3 35Cl)3 + 3(3 35Cl)2 (37Cl) + 3 (3 35Cl) (37Cl)2 + (37Cl)3
Now combining the terms in which the sum of the superscripts is the same and arranging the resultant terms in
the ascending order of masses, we obtain
(3 35Cl + 37Cl)3 = 27(35Cl)3 + 27(35Cl)2(37Cl) + 9(35C1)(37Cl)2 + (37Cl)3

If M is mass of the molecular ion containing three 35Cl, i.e. (35Cl)3 then the above equation reduces to
(3 35Cl + 37Cl)3 = 27 M + 27 (M + 2) + 9(M + 4) + (M + 6)

Accordingly, the intensity ratio of the peaks at M, M + 2, M + 4 and M + 6 will be 27 : 27 : 9 : 1.
For n = 3 chlorine atoms and m = 2 bromine atoms. Number of peaks expected = 3 + 2 + 1= 6
In order to determine the intensity ratio of the peaks we apply the formula (a + b)n (c + d)m as follows:
(3 35Cl + 37Cl)3(79Br + 8lBr)2 = [(3 35Cl)3 + 3(3 35Cl)2 (37Cl) + 3(3 35Cl)(37Cl)2 + (37Cl)3][(79Br)2 + 2(79Br)(81Br)
+ (81Br)2]
= 27 (35Cl)3(79Br)2 + 54(35Cl)3(79Br)(81Br) + 27(35Cl)3(81Br)2 + 27(35Cl)2(37Cl)(79Br)2
+ 54(35Cl)2(37Cl)(79Br)(81Br) + 27(35Cl)2(37Cl)(81Br)2 + 9(35Cl)(37Cl)2(79Br)2
+ 18(35Cl)(37Cl)2(79Br)(81Br) + 9(35Cl)(37C1)2(81 Br)2 + (37Cl)3(79Br)2 + 2(37Cl)3(79Br)(81Br) + (37Cl)3(81Br)2
Combining the terms having the same sum of the superscripts and arranging the resultant terms in the ascending
order of the masses, we get:
(3 35Cl + 37Cl)3(79Br + 81Br)2 = 27(35Cl)3(79Br)2 + [54(35Cl)3(79Br)(81Br)

+ 27(35Cl)2(37Cl)(79Br)2] + [27(35Cl)3(81Br)2 + 54(35Cl)2(37Cl)(79Br)(8lBr)
+ 9(35Cl)(37Cl)2(79Br)2] + [27(35Cl)2(37Cl)(8lBr)2 + 18(35Cl)(37Cl)2(79Br) (81Br)
+ (37Cl)3(79Br)2] + [9(35Cl)(37Cl)2(81Br)2 + 2(37Cl)3(79Br)(8lBr)] + [(37Cl)3(81Br)2]
If M is the mass of the molecular ion containing three (35Cl) and two (79Br), i.e. (35Cl)3(79Br)2, then the above
expression reduces to
(3 35Cl + 37Cl)3(79Br + 81Br)2 = 27M + [54(M + 2) + 27(M + 2)] + [27(M + 4) + 54(M + 4) + 9(M + 4)]
+ [27(M + 6) + 18(M + 6) + (M + 6)] + [9(M + 8) + 2(M + 8)] + (M + 10)
= 27M + 81(M + 2) + 90(M + 4) + 46(M + 6) + 11(M + 8) + (M + 10)
Accordingly, the intensity ratio of the peaks at M, M + 2, M + 4, M + 6, M + 8 and M + 10 will be 27 : 81 : 90:

4 6: 11 : 1.
Such type of data has been generated for different numbers of particular halogen atoms in the mol-
ecule as well as for various combinations of different halogen atoms and the resultant data are presented
graphically in Fig. 11.31. Contributions due to C, H, N and O are usually small as compared to those of
Br and Cl.
To summarise, the following tools may be applied for determining the molecular formula by mass spectrometry.
(i) Convert the intensity (as per cent of base peak) to intensity as per cent of M+•.
(ii) According to nitrogen rule, if M+• is even, the compound must contain even number including zero of
nitrogen atoms.
(iii) Number of carbon atoms = Relative abundance of (M+• + 1)/1.1
(iv) The relative abundance of the M+• + 2 peak indicates the presence (or absence) of S(4.4 percent), Cl(33 per
cent) or Br(98 per cent).
(v) The molecular formula of the compound can now be accomplished by determining the number of hydro-
gen atoms and adding the appropriate number of oxygen atoms, if necessary.
Let us now apply these rules to the mass spectral data in order to accomplish the molecular formula.
m/e 86(M+•) 87 88
Intensity 10 0.56 0.04

(Per cent of base peak)
Intensity as per cent of M+• 100 5.6 0.4
According to nitrogen rule, there is no nitrogen present in the molecule and the number of carbon atoms
56
= = 5. Since the per cent of M + 2 ions is 0.4, the presence of S, Cl or Br is ruled out. Thus, the number
1.1
of hydrogen atoms = 86 − 60 = 26.
Accordingly, the molecular formula of the compound is C5H26. But C5H26 is impossible. For a molecule con-
taining C, H and one oxygen atom,
H = 86 − (5 × 12) − 16 = 10
Accordingly, the molecular formula of the unknown compound is C5H10O.

The number of possible molecular formulas can be further reduced provided the spectrum is recorded on a
high-resolution instrument. In such a case, the molecular formula of the ion for combinations of C, H, N and O
can be obtained from the exact mass defect alone by using the expression 11.44.
Cl Cl2 Cl3
M M M M+2
M+2
M+4
M+2
M+4 M+6
M+2
M M+2 M+2 M+2
M M+4
M
Br M+4
M+4 M
M+6
M+6 M+8
M+2 M+2 M+2 M+2

M+4 M+4
M+4
M
Br2 M+4
M M M+6
M
M+6
M+6 M+8 M+8
M+2 M+4 M+4

M+4 M+4
M+2
M+2 M+6
M+6
M+2
Br3
M M M+6
M M+8
M+6 M+8 M
m/e
Fig. 11.31 Theoretically expected mass spectra of typical Cl and Br systems.
Exact mass defect + 0.005

005
511x − 0.0031y
Number of hydrogen atoms = (11.44)
0.0078
where, x is for oxygen and y is for nitrogen.
In order to determine the number of hydrogen atoms, keep in mind the nitrogen rule and choose such integral
values of x and y so that the denominator becomes an integral multiple of the numerator within 0.0002 units.
Computer programs are available or may be developed to determine the molecular formula of a compound
from its exact molecular mass.
For example, if the exact molecular mass of a compound containing hydrogen, nitrogen and carbon is 28.0187,
0.0187 − 0.0031 × 1 0.0156
the number of hydrogen atoms = = = 2.
0.0078 0.0078
Accordingly, the mass due to the carbon atoms present in the molecule (28 − (14 + 2)) = 12 (≡ 1 C).
Thus the molecular formula of the compound is H2CN.
Problem 11.30: Determine the molecular formula of a compound containing C, H, O or N whose exact
molecular mass is 59.0370.
Solution Exact mass defect = 0.0370.

Substituting the value of mass defect into Eq. (11.44), we get
0.0370 + 0.00 x 0.0031y
Number of hydrogen atoms =
0.0078
For x = 1 and y = 1, we obtain
0.0370 + 0.00 0.0031 0.0390
Number of hydrogen atoms = = = 5, i.e. an integral number.
0.0078 0.0078
The mass due to carbon atoms in the molecule = [59 − (5 + 14 + 16)] = 24 (= 2C)
Thus, the molecular formula of the compound is C2H5NO.
Problem 11.31: The characteristic peaks in the mass spectrum of an unknown compound are indicated below:
m/e 51 77 105 181 182 183 184
Relative 19 62 100 8.2 60.0 8.5 0.61

abundance (Base peak) (M+•)
(a) Determine the number of carbon atoms in the compound, and (b) what is the likely molecular formula for the
compound?
85
Solution (a) According to Eq. (11.42), the number of carbon atoms = = 13.
60 × 0.011
(b) Now number of carbon atoms = 13
Accordingly, the number of hydrogen atoms = 182 − 13 × 12 = 26
For a molecule composed of carbon and hydrogen, the molecular formula of the compound should be C13H26.
But C13H26 is impossible since it does not correspond to either CnH2n + 2 or CnH2n − 2. The possibility correspond-
ing to C2H2n is also ruled out because the mass spectral data cannot be explained on the basis of this molecular
formula.
For a molecule composed of C, H and one O,
Number of hydrogen atoms = 182 − (13 × 12) − 16 = 10
Thus, the possible molecular formula of the compound is C13H10O, i.e. C6H5COC6H5.
Problem 11.32: (a) Predict the Intensity ratio of the M+•, M+• + 2 and M+• + 4 peaks of CH2Cl2, and (b) for the
M+•, M+• + 2, M+• + 4 and M+• + 6 peaks of CHBr3. (c) Determine the possible molecular formula of the unknown
compound from the mass spectral data indicated below:
m/e 64 66
Intensity 3 1
Solution (a) Here n = 2, a = 3 and b = 1.

Substituting these values into Eq. (11.43), we obtain
(3 35Cl + 37Cl)2 = 9(Cl35)2 + 635Cl37Cl + (Cl37)2
Since (Cl35)2 = M+•, the above equation can be expressed as
(3 35Cl + 37Cl)2 = 9M+• + 6(M+• + 2) + 1(M+• + 4)
Thus the intensity ratio of the peaks in question is 9 : 6 : 1.
(b) In this case, n = 3, a = 1, and b = 1.
Proceeding as in the preceding problem, we obtain
(Br79 + Br81)3 = (Br79)3 + 3(Br79)2Br81+ 3Br79(Br81)2 + (Br81)3
Since (Br79)3 = M+•, the above equation in term of M+• becomes
(Br79 + Br81)3 = M+• + 3(M+• + 2) + 3(M+• + 4) + (M+• + 6)
Thus the intensity ratio of the peaks, i.e. M+•, M+• + 2, M+• + 4 and M+• + 6 is 1:3:3:1.
(c) The intensity ratio of the two peaks and the mass interval of two units between the peaks suggest that the
compound contains one chlorine atom. For a compound containing C, H and Cl, the mass due to carbon and
hydrogen atoms is 64 − 35 = 29. Thus the possible molecular formula of the compound is C2H5Cl.
Problem 11.33: Determine the empirical formula of the compounds containing C, H, O or N unless otherwise
indicated corresponding to the exact molecular masses indicated here:
180.0939, 190.9540, (Cl), 334.0873 (S), 109.0528.
Solution From Eq. (11.31), we arrive at the following results:
Molecular mass Molecular formula

180.0939 C4H12
190.9540 (mass defect 0.0460) C6H3Cl2NO2
334.0873 C17H18O5S
109.0528 C6H7NO
Problem 11.34: Deduce the number and type of halogen atoms present in a molecule from the mass spectral
data indicated below:
m/e M M+2 M+4 M+6 M+8
Per cent abundance 35 45 11 … …
Solution The number of peaks is one more than the number of halogen atoms contained in the molecule.
Accordingly, the number of halogen atoms = 3 − 1 = 2.
For n = 2, the possible combinations of chlorine and bromine atoms are 2 + 0, 0 + 2, 1 + 1.
For Cl = 2, we write from the binominal expansion (a + b)n
a = (3 35Cl + 37Cl)2 = 9(35Cl)2 + 6 35Cl37Cl + (37Cl)2
If M is the mass of the molecular ion, M+ •, i.e. M = (35Cl)2 then the above expression in terms of M becomes:
a = 9M + 6(M + 2) + l (M + 4)
The intensity ratio of the peaks 9:6:1 does not conform to the expected ratio 35:45:11 or 3:4:1.
Thus Cl = 2 is not possible.
for Br = 2, the binomial expansion (a + b)n yields.
b = (79Br + 81B)2 = (79Br)2 + 2(79Br)(81Br) + (81Br)2
In terms of mass of the molecular ion, we write
b = M + 2(M + 2) + 1(M + 4)
The intensity ratio of peaks, i.e. 1 : 2 : 1 in this case also does not conform to the expected ratio.
Thus Br = 2 is also ruled out.
For Cl = 1 and Br = 1, we write from the binomial expansion (a + b)n (c + d)m:
g = (3 35Cl + 37Cl)(79Br + 81Br) = 3(C135 Br79) + 3(35Cl81Br) + 37Cl79Br + 37Cl81Br
Collecting the terms where the sum of the superscripts is the same and arranging the resultant terms in the
ascending order of the masses, we write
g = 3(35Cl 79Br) + 3(35Cl79Br 2Br) + 1(35Cl79Br 2Cl) + 1(35Cl79Br 2Cl2Br)
In terms of mass of the molecular ion, i.e. M = 35Cl79Br, the above expression becomes
g = 3M + 4(M + 2) + l(M + 4). [4(M + 2) = 3 (M + 2) + 1 (M + 2)]
The intensity ratio of peaks, i.e. 3 : 4 : 1 for Cl = 1 and Br = 1 conforms to the expected ratio. Thus the
compound contains one chlorine atom and one bromine atom.
Problem 11.35: Deduce the molecular formulae of the compounds corresponding to the mass spectral data
indicated below:
(a) m/e 90(M+•) 91 92

per cent of base peak 100 5.61 4.69
(b) m/e 89(M •)
+ 90 91 92
per cent of base peak 17.12 0.58 5.36 0.17
(Continued )
(c) m/e 230(M+•) 232 234

per cent of base peak 1.10 2.12 1.06
(d) m/e 140(M •)+ 141 142
(e) m/e 151(M •)+ 152 153 154
per cent of base peak 100 10.4 32.1 2.9
Solution (a) The per cent abundance of M+• + 2 peak (i.e. 4.69) indicates the presence of sulphur atom in the
molecule. Consequently, the contribution of C and H atoms to per cent abundance of M+• +1 peak
= (5.61 − per cent abundance due to sulphur atom) = 5.61 − 0.78 = 4.83
4 83
Thus, the number of carbon atoms in the molecule = =4
1.11
or
Mass of C and H atoms in the molecule = 90 − 32 = 58.
58 58
Accordingly, = 4CH2 groups + 2H atoms or = 4Cs + 10Hs
14 12
Thus, the molecular formula of the compound is C4H10S. The compound is dialkyl sulphide C2H5SC2H5.
An alternative procedure: For a compound containing C and H atoms, the number of hydrogen atoms in the
molecule = (90 − 12 × 4) = 42. But it is not possible since the compound contains sulphur atom also. So for a
compound composed of C, H and S atoms, the possible number of hydrogen atoms = (90 − 12 × 4 − 32) = 10.
So, the formula of the compound is C4H10S.
(b) Expressing the data as per cent of molecular ion, we obtain
m/e 89 90 91 92
per cent of base peak: 17.12 0.58 5.36 0.17
per cent abundance as 100 3.38 31.3 0.99
molecular ion M+
The odd mass number and per cent of M + 2 peak relative to that of the percent of M+ peak indicates the presence
3.38 0.38
of odd number of nitrogen atoms and Cl atom. The number of carbon atoms = = 2.7, i.e. the number
1.11
should not exceed 2. Accordingly, the residual mass = 89 − 2 × 12 − 14 − 35 = 16 (= O).
Thus, the molecular mass of the compound is C2ClNO. Note that if 16 ≡ 16H the formula of the compound
should be C2ClNH16 which is not possible.
(c) The even mass number and intensity ratio of M, M + 2 and M + 4 peaks, i.e. 1 : 2 : 1 indicate that the
compound contains even number of nitrogen atoms and two bromine atoms. Accordingly, if number of nitrogen
72
atoms is taken as zero, the residual mass = 230 − 2 × 79 = 72. = 5CH2 groups + 2 hydrogen atoms. Thus, the
molecular formula of the compound is C H Br . 14
5 12 2
(d) m/e 140 (M+•) 141 142
per cent of M+ • peak 100 8.9 5.74
The even mass number and per cent of M + 2 peak relative to that of the per cent of M+• peak indicate the
presence of even number of nitrogen atoms and a sulphur atom. If number of nitrogen atoms is zero then
108
the residual mass is 140 − 32 = 108. Accordingly, = 7CH2 groups + 10 hydrogen atoms or the contribution
14
of C and H ions to percent abundance of M + 2 peak = 8.9 − 0.78 = 8.12. Thus, the number of carbon atoms in
8 12
the molecule = = 7. Accordingly, the molecular formula of the compound is C7H24S. Or for a compound
1.11
containing S and seven carbon atoms, the number of H atoms = (140 − 12 × 7 − 32) = 24. So the formula of
the compound is C7H24S.
(e) Odd mass number of molecular ion and the per cent abundance of M + 2 peak, i.e. 32.1 indicate that the
compound contains nitrogen and chlorine atoms.
Accordingly, the residual mass = 151 − 35 − 14 = 102
102
So, = 8Cs + 6Hs
12
Thus, the most probable formula of the compound is C8H6Cl N.
Since, the base peak is the molecular ion peak, the compound seems to be an aromatic compound with struc-
tural formula CN(C6H4)CH2Cl.
10.4 0.38
Alternative method: Number of carbon atoms = = 9.
1.11
For a compound containing C and H; the number of hydrogen atoms = 151 − 12 × 9 = 43
So, the molecular formula of the compound is C9H43.
But it is not possible since the compound contains Cl and N also. So for a compound containing C, Cl and H,
the molecular formula can be written as C9H8Cl.
The formula contains nitrogen atoms also. Since the atomic mass of nitrogen is equivalent to mass of CH2
102
group, the formula of a compound is C8H6ClN. Or we can start from CH2 group, i.e. = 7CH2 + 4Hs = C7H18
(≡ C H ). 14
8 6
(g) General Rules for Writing Molecular Formula There are two general rules that help in writing the
molecular formulas of organic compounds, viz., hydrogen rule and nitrogen rule.
(i) Hydrogen Rule According to this rule, organic compounds (containing C, H, O and N) with even
molecular mass will have an even number (including zero) of hydrogen atoms. If the molecular mass is
divisible by some even numbers, the number of hydrogen atoms must also be divisible by at least one of those
numbers. But if the number of hydrogen atoms are divisible by two or more such numbers, the highest divi-
sor will correspond to the minimum number of hydrogen atoms present in the molecule. As examples, CH4,
m/e : 16, H2NNH2, m/e : 32, C17H35COOH, m/e : 284; C6H5OH, m/e : 94; C7H12N2O4, m/e : 188; C17H14N2O3,
m/e : 294, etc.
(ii) Nitrogen Rule The rule has already been stated under the title The even electron rule. According to this
rule, an organic compound with an even (including zero) number of nitrogen atoms will have an even mass
number and if the compound contains an odd number of nitrogen atoms, the mass number will also be odd.
This rule applies to all organic compounds containing C, H, N, Si, O, S, As, P, halogens and B. As examples,
molecules giving even mass molecular ions are H2O, m/e : 18, CH4, m/e : 16; H2NNH2, m/e : 32; aminopyridine-
C5H6N2, m/e : 94, etc., and molecules giving odd mass molecular ions are NH3, m/e : 17; C2H5NH2, m/e : 45,
quinoline, C9H7N, m/e : 129, etc.
The nitrogen rule not only applies to molecular ions (i.e. M+•) but to all ions, i.e. fragment ions. However,
the fragment ions must contain all the nitrogen atoms of the molecular ion. Thus, the nitrogen rule can also
be stated as follows: An odd-electron ion will have an even/odd mass number provided it has an even/odd
number of nitrogen atoms and an even-electron ion will have an odd/even mass number if it contains an
even/odd number of nitrogen atoms. As an example, mass spectrum nitrobezene (C6H5NO2) exhibits a molec-
ular ion peak at m/e : 123 and the two ion fragments, namely NO2+ and NO+, are observed respectively at
m/e : 46 and 30, i.e. both these fragment ions appear at even mass numbers. On the other hand, the molecular
ion peak in the mass spectrum of 2, 4-dinitrophenol is observed at an even mass number of m/e : 184 while
the fragment ions (M+—H) and (M+—H—CO) are observed respectively at m/e : 183 and 185, i.e. both the
fragment ions are observed at odd mass numbers as expected according to nitrogen rule. For ready reference,
the rule is summarised in Tables 11.14 (a) and 11.14 (b).
Table 11.14(a) Relative parity of A and N for even- or- odd-numbered electron neutral species.
Neutral species Mass number Number of N atoms Number of electrons
Molecule ]
0
even ⎫⎪ even ⎫⎪ even
⎬ ⎬
odd ⎪⎭ odd ⎪⎭
Radical]
•
odd ⎫⎪ even ⎫⎪ odd
⎬ ⎬
even ⎭⎪ odd ⎭⎪
Table 11.14(b) Relative parity of m/e and N for even- or– odd-numbered electron positively charged species:
The nitrogen rule.
Charged species Mass number (m/e) Number of N atoms Number of electrons
Ion radical ]
+•
even ⎪⎫ even ⎪⎫ odd
⎬ ⎬
odd ⎪⎭ odd ⎪⎭
Ion ]
+
odd ⎫⎪ even ⎫⎪ even
⎬ ⎬
even ⎪⎭ odd ⎪⎭
Once the molecular formula of the compound is accomplished, the number of unsaturated sites R can be
determined by the ring rule.
(iII) Ring Rule According to this rule, the sum total of the number of rings and double bonds in an organic
molecule of general formula CwHxNyOz is given by
y −x
R = w +1+ (11.45)
2
where terms and symbols have their usual meanings. For an even electron ion (M+), the true value will be fol-
lowed by 0.5. So in order to obtain the true value, this factor of 0.5 should be omitted.
As examples, in case of a compound with molecular formula C2H2; w = 2 and x = 2
0 2
Therefore, R = 2 +1 + = 2. Accordingly, there is a possibility of two double bonds in the molecule.
2
Thus the structural formula of the compound is HC ≡ CH. For a molecule of molecular formula C6H6; R = 6 + 1
0 6
+ = 4. Accordingly, there is a possibility of one ring and three double bonds. Thus, the molecule is benzene.
2
For a compound with formula C7H5O; R = 5.5. This implies that the molecule gives an even electron ion and the
true value of R should be 5. Accordingly, it contains one ring and four double bonds, i.e. the structural formula
C O+
of the molecular ion is . For a compound with molecular formula C12H10CO, R = 9, i.e. two rings
+•
O
C
and seven double bonds. Thus, the structural formula of the molecular ion is .
For a compound with molecular formula C4H4O, R = 3, i.e. one ring and two double bonds. This data fits into
CH CH
the structural formula O.
CH CH
Similarly, we can arrive at the following results:
Molecular formula Structural formula
••
N
H2C CH2 N •• H2C N
C2H4O C4H4N2 or CH2N2
O N N N
• • • •
H2C CH2
H8C4O H2C CH2
C5H5N
N
O • •
Furthermore, the ring rule can also be extended to compounds that contain other elements in addition to C,
H, O or N. The number of additional elements is also counted respectively towards C, H, O or N on the basis of
principle of corresponding valency. Thus, the number of Si atoms will be counted towards carbon atoms, halogen
atoms will be taken as hydrogen atoms, phosphorous atoms will be considered as nitrogen atoms, S atoms will
be counted towards oxygen atoms, etc. Thus, the more general formula corresponding to the formula CwHxNyOz
is written as:
Iw, IIx, IIIy, IVz, where, I = C, Si; II = H, halogens; III = N, P; and IV = O , S, etc.
It is to be kept in mind that the above deduction is made on the lowest valence states of the elements and the
double bonds formed to elements in higher valence states are not counted. Otherwise we may arrive at negative
values and unexpected positive values of double bonds, e.g. R = 1 for nitro methane, R is negative for H3O and
equals zero for C2H5SO2C2H5.
Problem 11.36: Sort out the odd and even electron ions corresponding to the following molecular formulas:
CH4S, CH4, C4H8O, CH3Cl, H3O, C6H4ClBr, C7H5ClBr, C3H9SiO, C2H5NO, N2H4, C29F59, C3H8N.
Solution Applying the nitrogen rule [Table 11.14 (b)], we arrive at the following results:
Molecule Molecular mass Number of Nature of Ion

nitrogen atoms
CH4S Even, 48 Even, 0 Odd, CH4S+•
CH4 Even, 16 Even, 0 Odd, CH4+•
C4H8O Even, 72 Even, 0 Odd, C4H8O+•
CH3Cl Even, 50 Even, 0 Odd, CH3Cl+•
H3O Odd, 19 Even, 0 Even, H3O+
C6H4ClBr Even, 190 Even, 0 Odd, C6H4ClBr+•
C7H5ClBr Odd, 203 Even, 0 Even, C7H5ClBr+
C3H9SiO Odd, 89 Even, 0 Even, C3H9SiO+
C2H5NO Odd, 59 Odd, 1 Odd, C2H5N+•
N2H4 Even, 32 Even, 2 Odd, H4N2+•
C29F59 Odd, 1469 Even, 0 Even, C29F59+
C3H8N Even, 58 Odd, 1 Even, C3H8N+
Note: Even ions are mass spectrometrically active.
(h) Determination of Ionisation Potential of a Molecule and Appearance Potential of an Ion Molecular
ion is formed when the energy of the bombarding electron beam equals the ionisation potential of the molecule.
The ionisation potential of the molecule consists of energy required to break the chemical bond, the ionisation
potential of formed ionised fragment of a molecule, the kinetic and excitation energies of all the fragments
formed. With a further increase in the energy of the electrons, excitation of an ionised molecule may lead to dis-
sociation. The latter results in the appearance of ions with smaller masses. Neutral fragments of a molecule may
also appear.
The smallest energy of an electron at which dissociation of a molecule occurs with the formation of new ions
is called the appearance potential of the ion. These potentials, i.e. ionisation and appearance potentials, can con-
veniently be measured with the aid of a mass spectrometer.
For the reaction, R – X + e →R• + X + 2e; R• + e → R+ + 2e the appearance potential AR+ for fragment R+
obtained from the molecule RX (i.e. R – X + e → R+ + X• + 2e) is related to the dissociation energy DR−X of the
R—X bond and ionisation potential IPR• of the radical R• as
AR+ = IPR• + DR–X (11.46)
The intensity of the molecular ion peak (RX + e → RX + 2e) grows with the increase in the energy of the bom-
− +•
barding electrons. However, in case of fragment ions, the intensity decreases as the ionising voltage is increased.
Thus, it is possible to determine the ionisation potential from the ionisation efficiency energy curve obtained by plot-
ting the intensity of the molecular ion peak as a function of the energy of the impact electrons. When the energy of
the impact electrons exceeds the ionisation energy, the fission of the molecular ion occurs leading to fragment ions
whose intensities decrease with the increase in the energy of the bombarding electrons. Thus, by extrapolation of
the ionisation efficiency energy curve for the molecular ion, the ionisation potential of the molecule can be obtained
(Fig. 11.32). When the peak corresponds to the fragment, extrapolation of the ionisation efficiency curve produces
the appearance potential of that fragment. It is also possible to determine the bond-dissociation energies from the
knowledge of the appearance potentials of fragment ions, which can be determined from the ion-efficiency curve as
described above. For example, in order to determine the ionisation potential of N2, argon is used as an internal stan-
dard. First of all, blank mass spectrum is recorded in order to check the absence of molecular ion peaks at m/e : 28 and
m/e: 40 respectively, for N2 and Ar in the background material. Now the mass spectrum of a mixture of Ar and N2 is
recorded. The intensity of the molecular ion peak of Ar+ (40) is measured as a function of energy of impact electrons.
The electron beam energy is then reduced in steps till the intensity becomes zero. The experiment is repeated a
number of times and then from the plot between electron beam energy for m/e = 40 (Ar+) and intensity, the ionisation
potential is determined. The standard value of ionisation potential is 15.76 eV. The difference between the standard
and observed experimental values of ionisation potential is used to correct the ionisation potential of nitrogen.
Absolute intensity
10 20 30 40 50 60
Electron voltage
Fig. 11.32 An ionisation efficiency energy curve of a system.
In Eq. (11.46), Ek, the kinetic energies of the particles produced; and Eex the excitation energy of the fragments
(i.e. electronic, vibrational, and rotational energy of the fragments if they are formed in excited states) being small
have been neglected.
The appearance potentials of selected ions in methane measured with the aid of mass spectrometer are given in
Table 11.l5 and the energies of conversion R• + e → R+ + 2e and heats of formation of carbonium ions have been
listed in Table 11.16. The energy of decomposition of some organic molecules into two univalent radicals at 298 K
is listed in Table 11.17 for ready reference.
Table 11.15 Energies of conversion R• + e → R+ + 2e and heat of formation of carbonium ions.

Ionisation Potential Relative energy of radical Heat of formation of ion
R+
of radical (eV) conversion (kcal mole−1) (kcal mole−1)
CH 3+ - - 262
CH 3CH 2+ - - 224
(Continued)
Table 11.15 Energies of conversion R• + e → R+ + 2e and heat of formation of carbonium ions—Cont'd.
CH 3CH 2CH +2 8.69 29.2 216
CH 3C+HCH 3 7.90 11.1 190
CH 3CH 2 CH 2CH +2 8.64 28.1 207
(CH ) 2CHCH +2 8.35 21.4 211
CH3CH2+CH+CH3
7.93 11.7 181
(CH3)2C+CH3 7.42 0 166
Table 11.16 Appearance potentials of ions when methane is bombarded by electrons.
Reaction Appearance potential, eV

CH4 → CH 4
+
13.1
CH4 → CH + H 3
+
14.4
CH4 → CH2+ + H2 15.7
CH4 → H +:CH2 + H
+
22.7
CH4 → :CH + 3H+
23.3
CH4 → H + C + 3H
+
29.4
Table 11.17 Energy of dissociation of molecules into two univalent

radicals at 298 K (in kcal mole−1).
H CH3 C2H5 C6 H5
H 104.2 103 98 102

CH3 103 86 84 90
C2H5 98 84 80 88
C6H5 102 90 88 98
The C—H bond dissociation energy for the aliphatic radical Rn = Cn H2n+1 in the series Rn—X (X = H, OH, halogen,
C6H5, etc.) can be obtained from the expression given below:
DC—H = D(CnH2n + 1— H) = 77.6 + 8.0 ∑ 0.4 k i (11.47)
where, ki is the number of C atoms is each of the three radical chains R1, R2 and R3
R3
R1 C R2
For the example, in the case of

CH3
H3C C CH2 CH3
where, k1 = k3 = 1 and k2 = 2, the energy values, obtained from Eq. (11.47), for the detachment of an H atom from
various hydrocarbons, are in good agreement with experiment.
Such studies can also be extended to inorganic systems. As an example, the mass spectrum of solid Cr2O3
shows peaks corresponding to Cr, CrO, CrO2, O and O2 species. The appearance potentials and bond dissociation
energies of these species have been determined.
Problem 11.37: The appearance potential of CH3+ ion is 14.4 eV. Calculate the dissociation energy of the
•
C⎯H bond if the energy of ionisation of the C H 3 radical is 9.83 eV.
Solution According to Eq. (11.46), the appearance potential of CH3+ ion equals the sum of the bond breaking
•
energy DCH3 — H and the energy of ionisation of the radical C H 3 . Accordingly, we write
A CH3+ = IP
PCH•3+ DCH3 H or DCH
CH 3 H = A CH3+ IPC•H
3
Substituting the data into the above expression, we get
DCH3 — H (14.4 83) eV = 4.57 eV = 440.86 kJ mole –1

– 9.83
The value of the C—H bond energy obtained by kinetic and photochemical methods respectively is 451.9 and
396.6 kJ mole−1.
The C—H bond energy DCH3 — H can also be obtained from the appearance potentials of C2H5+ ion in propane
and ethane.
•
In propane, A Pr + = 14.5 eV, and in ethane A Et + = 15.2 eV, we write: C3 H8 → C2 H 5 + C H 3 ; 14.5 eV
+
C2 H5 C2 H5
(i)
C2 H 6 → C2 H 5+ + H; 15.2 eV (ii)
It is known from thermochemical data that

CH4 + C2H6 → C3H8 + 2H; 5.08 eV (iii)
The algebraic operation (i) + (iii) − (ii) yields
•
CH 4 → C H 3 + H; 4.38 eV
Thus the value of C—H bond energy, DCH3 — H = 4.38 eV = 422.6 kJ mole−1.
(i) Latent Heat of Vaporisation The latent heat of vaporisation of liquids can be determined by making use
of Clausius–Clapeyron equation:
−ΔH vap
a
In p = + constant (11.48)
RT
where, p = vapour pressure of the liquid

ΔHvap = latent heat of vaporisation of the liquid
R = gas constant
T = temperature (K)
The value of p can be obtained from the peak height in mass spectrum. Thus, by studying the change in peak
intensities as a function of sample temperatures, one can determine the value of ΔHvap. The plot of ln p or intensity
1
of positive peak XT against will be a straight line with slope
T
−ΔH vap
a
( )= . From the slope, we get
R
ΔHvap = m × R (11.49)
Similarly, one can also determine the heat of sublimation (ΔHsub) of solids by studying the change in peak
intensities as a function of sample temperatures since intensity of peaks in a spectrum is directly proportional to
the pressure of the sample in the ion source. The sample placed in a reservoir is made to enter the source through a
small hole by diffusion only. A sufficient amount of sample is kept in the thermostated cell so that the solid phase
is always present in the cell. Such studies also show the formation of n-mers in the vapour over some high melt-
ing solids. For examples, monomers and dimers are produced in the vapour over sodium, potassium and cesium
chlorides while monomers, dimers and trimers are observed in the vapour over lithium chloride. MoO3 has been
found to consist of trimer, tetramer, and pentamers. The individual heats of sublimation of the dominant gaseous
species have also been obtained by utilsing the Clausius–Clapeyron equation.
(j) Reaction Mechanism Mass spectrometric determinations of molecular mass to less than one mass unit
have been applied to the investigation of reaction mechanisms. Isotopic labelling of a reacting molecule (most
frequently deuterium replacing hydrogen, or 18O replacing 16O) enables the change in position of individual
atoms during reactions to be determined from an examination of mass spectra of the various products, e.g. two
mechanisms are possible of the esterification of benzoic acid by methanol.
18
C6H5 C O CH3 + H2O
I
I II II 18 I O
C6H5 C O H + CH3 O H
II 18
O
C6H5 C O CH3 + H2O
According to Mechanism I, the bond between the acyl and hydroxyl is severed in the acid molecule, and the
bond between hydrogen and oxygen in the alcohol molecule is broken; according to Mechanism II, the bond
between the hydrogen and the oxygen in the acid is ruptured, while in the alcohol molecule, the break occurs
between the alkyl and the hydroxyl.
It is very difficult to make a choice between these two mechanisms on the basis of theoretical hypotheses or analogies,
because instances are known where the alcohol or acid behaves according to each of the above-mentioned mechanisms,
e.g. alcohols react with hydrogen halides according to the only one possible mechanism that is depicted by II.
Br – H + HO – R → R—Br + H2O
On the other hand, in a number of esterification reactions, the RO—H bond is broken according to Mechanism I.
This problem has been solved by carrying out the reaction in CH3OH enriched with heavy isotope 18O and
studying the reaction by mass spectrometry or by radiometric method. The mass spectrum of the reaction
products reveals that the highest-mass molecular ion from the benzoic ester produced occurs at 156, indicating
that some 18O has been incorporated into this molecule. The molecular ion for H2O occurs at a mass number
18, which indicates no incorporation of 18O. As neither the ester nor the water can exchange 18O directly with
CH3 18OH at any significant rate, we can conclude that the esterification reaction takes place by Route I. Path II
would require 18O incorporation into water rather than the ester.
Thus we can establish the mechanisms of chemical reactions simply by labelling a particular atom (a slight
enrichment would suffice) and then by locating it by using mass spectrometry alone.
(k) Structural Elucidation of Natural Products The improvement in the mass spectrometric techniques such as
direct introduction system, resolution, sensitivity, etc., has made this technique very useful for the structural elucidation
not only of small molecules but also of high molecular mass natural products such as proteins and peptides, sugars,
lipids, alkaloids, terpenes, nucleic acids, etc. In this section we have introduced the structural elucidation of some
natural products, namely peptides, sugars, nucleosides, glycerides and carotenes.
(i) Structural Elucidation of Peptides Before describing the methodology for determining the sequence of amino
acid units in peptides, it is nice to comment on the mass spectra of amino acids, the basic units of peptides and proteins.
(ii) Mass Spectra of Amino Acids For an organic compound to be amenable to mass spectrometric analysis, it
must be sufficiently volatile. It is for this reason that amino acids are converted into their methyl or ethyl esters; the
amino group can be reacted to give trifluroacetyl derivatives, trimethylsilyl derivatives, etc. The structural formulas and
molecular masses of the important amino acids of the proteins are recorded here in Table 11.18 for ready reference.
Table 11.18 The most important amino acids of the protein.
Trivial name Abbreviation Formula (Molecular mass)

NH2CH2COOH
Glycine Gly
(75.07)
NH2CH(CH3) COOH
Alanine Ala
(89.10)
NH2CHCOOH
Valine Val CH(CH3)2
(117.15)
Table 11.18 The most important amino acids of the protein—Cont'd.

NH2CHCOOH
Leucine Leu CH2CH(CH3)2
(131.18)
NH2CHCOOH
Isoleucine Ileu H3C—CH—C2H5
(131.18)
NH2CHCOOH
Aspartic acid Asp CH2COOH
(133.11)
NH2CHCOOH
Asparagine Asn CH2CONH2
(132.13)
NH2CHCOOH
Glutamic acid Glu CH2CH2COOH
(147.13)
NH2CHCOOH
Glutamine Gln CH2CH2CONH2
(146.15)
NH2CHCOOH
Ornithine (+) Orn CH2CH2CH2NH2
(132.16)
NH2CHCOOH
Lysine Lys CH2CH2CH2CH2NH2
(146.19)
NH2CHCOOH
CH2
Arginine Arg
CH2CH2NH—C—NH2
(174.20) NH
NH2CHCOOH
Serine Ser CH2OH
(105.10)
NH2CHCOOH
Threonine Thr CH(OH)CH3
(119.12)
NH2CHCOOH
Cysteine CySH CH2SH
(121.16)
NH2—CHCOOH
Cystine CySSCy CH2S—

2
(240.30)
NH2—CHCOOH
Methionine Met CH2CH2SCH3
(149.21)
NH2—CHCOOH
Phenylalanine Phe CH2C6H5
(165.19)
(Continued)
Table 11.18 The most important amino acids of the protein—Cont'd.

NH2—CHCOOH
Tyrosine Tyr CH2C6H4OH—n
(181.19)
NH2—CHCOOH
Tryptophan Try H2C

NH
(204.23)
H2C—CH2
H2C CHCOOH
Proline Pro
NH
(115.13)
HOH C—CH2
H2C CHCOOH
Hydroxyproline Hypro
NH
(131.13)
NH2 CH COOH
H2C—C CH
Histidine His
N NH
CH
(155.16)
86 (a)
75 CH3 (CH2)3 CH COOC2H5

NH2
50
30
25
44 102 130
M+1
120 (b)
102
75
CH2 CH COOC2H5
NH2
50
74
25 29 91
M+1
M
20 40 60 80 100 120 140 160 180 200 220
Fig. 11.33 The spectra of ethylesters of (a) norleucine, and (b) phenylalanine.
Let us now examine the mass spectra of amino acids as examples, the acids chosen are norleucine and
phenylalanine. The molecular mass of the former is 131 and that of the latter is 165. The mass spectra of ethyl
esters of the two acids have been shown in Fig. 11.33(a) and (b). The major peaks in the mass spectra of both
the acid esters are listed below.
M+• M+• + 1
m/e 30 44 86 102 130 159 160
Norleucine:
Peak height 16 4 35 2.5 3.5 0.5 1.5
(in mm)
m/e 29 74 91 102 120 193 194

Phenylalanine:
Peak height 7 18.5 7.5 28 33 1 3
(in mm)
In phenylalanine, the peaks at m/e: 120, 102 and 91 are translated from the molecular ion peak as follows:
+•
O
C6H5 CH2 CH C O C2H5
NH2
• +
COOC2H5 + C6H5 CH2 CH NH2
m/e : 120
• +
C6H5CH2 + NH2 CH COOC2H5
m/e : 102
+ •
C6H5CH2 + NH2CHCOOC2H5
+
( ≡ C7H+7)
m/e :91
In phenylalanine, the peak at 102 is more +•

O
abundant than in isoleucine. This is because
the fragmentation at the position gives two CH3(CH2)3 CH C O C2H5
stable positive ions at m/e: 91 and 102. The • +
NH2 C2H5 + (CH3)(CH2)3
fragmentation pattern reveals that in all amino CH COO
acids, one should look for cleavage on either NH2
side of the a-carbon atom as expected. m/e : 130
In isoleucine, the peaks at m/e : 130, 102 and
86 are translated from the molecular ion peak at •
COOC2H5 + CH3(CH2)3
+
CH
m/e : 159 by the following mechanism.
Another derivative of amino acid, the trifluo- NH2
roacetylester shows similar behaviour. m/e : 86
O
(iii) Determination of the Sequence of Ami- • +
O C2H5
CH3(CH2)3 + NH2CHC
no Acid Units in Peptides The composition m/e : 102
of amino acids (i.e. various types of amino acids
and numbers of each kind) in peptides can easily +•
O
be determined by the use of automatic amino acid
analysers even without the use of mass spectrome- CF3CO NH CH C O C4H9
ters. But the determination of the sequence of amino
CnH2n+1 • +
acids in the peptide chain is important as it helps in C4H9 + CF3CONHCHOO
the elucidation of primary structure of peptides and CnH2n+1
proteins. The chemical methods are time consuming + •
CF3CONHCH + COOC4H9
and complicated. This problem has been solved by
mass spectrometry and is demonstrated here with CnH2n+1
the following examples. Just like amino acids, pep- +
CF3CONH CH2 + CnH2n
tides are also nonvolatile; the nonvolatility grows as +• m/e : 126
CnH2n+1 O
the number of amino acids in the peptide increases
and as such is not amenable to mass spectrometric CF3CO NH CH C OC4H9
analysis. But esters of N-acylated peptides being
+•
volatile can be introduced into the mass spectrome- OH
ter for analysis. Fragmentation of the molecular ion
CF3CO NH CH COC4H9 + CnH2n
initially proceeds with the elimination of the alkoxy m/e : 227
group which results in the formation of a linear ion, whose positive charge is localised at the C-terminal ester bond, and
the N-terminal bond is protected by the acyl residue. Further fragmentation of this ion proceeds with the successive
elemination of amino acid residues with transfer of the positive charge along the chain. The mass spectrum of the methyl
ester of N-decanoylprolyl-alanyl-valine (Fig. 11.34) exhibits major peaks at m/e : 524, 493, 394, 323, 252 and 155 amu.
The fragmentation of the ester of N-acetylated oligopeptide may be depicted by the following scheme.
H2C CH2
C9H19CO H2C CH CO NH CH CO NH CH CO NH CH CO OCH3
N
CH3 CH3 CH
100 H3C CH3
224
Relative intensity per cent
155
394
50
252 323 366

465
509 524
493
100 150 200 250 300 350 400 450 500

m/e
Fig. 11.34 Mass spectrum of methyl ester of N-decanoylpropyI alanyl-alanyl-valine.
O O O O O +
C9H19C N CH C NH CH C NH CH C NH CH C OCH3 m/e 524
H2C CH2 CH3 CH3 CH

CH2 CH3 CH3
O O O O
C9H19C N CH C NH CH C NH CH C NH CH C O+ ...m/e 493
H2C CH2 CH3 CH3 CH
CH2 CH3 CH3

O O O
C9H19C N CH C NH CH C NH CH C O+...............................m/e 394
H2C CH2 CH3 CH3
CH2
O O
C9H19C N CH C NH CH C O+............................................................m/e 323
H2C CH2 CH3
CH2
O
C9H19C N CH C O+..........................................................................................m/e 252
H2C CH2
CH2
C9H19 C O+............................................................................................................m/e 155
Thus, if the amino-acid composition of the oligopeptide under consideration is known, the amino-acid
sequence in the peptide chain can be established from its mass spectrum. Let us now examine a peptide derived
from the silk fibroin in relation to the amino-acid sequence in it. It is found to be an octapeptide and contains
four different types of amino acids, namely four molecules of glycine, two of alanine, and one each of valine and
tryptophan. The mass spectrum of N-acyl peptide methyl ester of this octapeptide shows the main characteristic
peaks at m/e : 832, 610, 539, 454, 383, 270, 199 and 144 amu. The examination of the spectrum reveals that the
fragmentation of the peptide fits into the following scheme.
1 2 3 4 5 6 7 8
Acetyl-MeG l y-MeA l a-MeG l y-MeV a l-MeG l y-MeA l a-MeG l y-MeT r yOMe
The amino acids are represented by numbers.
Mass Amino acid sequence
832 1–8
610 1–7
539 1–6
454 1–5
383 1–4
270 1–3
199 1–2
114 1
Peptides having 8 to 9 amino-acid units have successfully been studied by this technique. 10 mg quantity of the
sample is sufficient for such studies.
Problem 11.38: The parent peak mass of an peptide has been measured relative to the parent peak in the
spectrum of dibromobenzene (236.8638). The measured ratio of unknown mass/reference mass is found to be
1.001197. Calculate the exact mass of the peptide and deduce the molecular formula of the peptide.
⎡ Unknown mass ⎤
Solution Exact molecular mass of the peptide = ⎢ ⎥ × Exact massof reference
⎣ Reference mass ⎦
= 1.001197 × 236.8638 = 237.1474.
For a compound containing C, H, N and O; from Eq. (11.44), we get
0 + 0.0051 × 2 − 0.0031 × 3
Number of Hs =
0 0078
0 + 0.0102 − 0.0093
= = 19
0 0078
Accordingly, the compound contains 2Os, 3Ns and 19Hs. So the molecular formula of the compound is written
as CXH19N3O2.
Now, the mass due to x carbon atoms in the peptide = 237.1474 − (19 × 1 + 3 × 14 + 2 × 16) = 237.1474 − 93
144
= 144 ≡ =12Cs.
12
Hence, the molecular formula of the peptide is C12H19N3O2.
In cyclic peptides, a ring opening is followed by fragmentation. As an example, at a phenylalanine residue,
transfer of hydrogen leads to an open-chain ion as indicated below:
H C6H5
+• +•
O CH OH CHC6H5
C CH C NH CHCO
NH
R R
Here, R = remainder of the chain.

The ring opening can also occur at a position adjacent to a bulky amino-acid side chain giving rise to a radical
ion. This is attributed to the release of excess vibrational energy as rotational energy.
+•
O CH2R′ O O CH2R′ O+
•
NH C CH NH C NH C CH NH C
R R
The ions obtained after ring opening can undergo further fragmentation, as described for linear peptide
derivatives. The structure of the cyclopeptide alkaloid pandamine has been confirmed by mass spectrometry.
(iv) Structural Elucidation of Derivations of Sugars Just like amino acids and peptides, sugars are nonvola-
tile and hence as such can’t be studied by mass spectrometry. Due to this reason, sugars are converted into methyl
esters or acetates. The configurations of a- and b-glucoses are given here in the box for ready reference.
Configuration of a - and b-glucoses
CH2OH CH2OH
O O
H H
H H H OH
OH H OH H
OH H
OH OH
H OH H OH
a-D-Glucose b-D-Glucose
These two streoisomers of glucose, which differ only in the configuration of carbon-1 (for aldoses, of carbon-2 for
ketoses), are known as anomers (the a- or b-anomers). The above formulas represent the anomers of D-glucose.
This method of assignment of amonosaccharide to the a- or b-series is however applicable only in those cases
if there is no other pair of hydroxyl groups on one side of the plane of the ring (in the cis position). The NMR
method is not only used for the assignment of a monosacchride to the a- or b-anomers (the D-series) but also for
the determination of the equilibrium between a- and b-anomers in solution. The chemical shift for the protons
of the glycoside hydroxyl of a- and b-monosacchrides varies markedly: for a-anomers (the D-series), the shift is
5.2d, while for the b-anomers it varies with in the range of 4.55–4.8d. The significant peaks in the mass spectrum
of methyl ester of glucose are observed at m/e : 219, 205, 187, 176 and 149. Fragmentation of molecular ion of
monosaccharide occurs by several paths simultaneously, giving rise to several series of characteristic ions; the
most important fragmentation routes for the methyl ester of glucose are presented by the following scheme:
CH3OCH2 CH3OCH2
O+ O+
H H
H H H
CH3O OCH H −CH3OH
3 OCH3 H
H OCH3 H OCH3
m/e : 219 m/e : 187
•
−OCH3
H CH3OCH2 +•
O+ O
H H H H H H H
CH3O OCH H OCH3 CH3O OCH H OCH3

•
CH3O C +OCH
OCH3H 3
3 3
H OCH3 H OCH3 H OCH3

m/e : 205 m/e : 250 m/e : 176
−CH4
CH3OCH2 +
O
H H CH3OCH2
OCH3
CH3O H C O+ H
O H C OCH3
•
CHOCH3 CH3O
m/e : 149
The scheme shows that all the fragmentations proceed by a single route involving b-fissions, and that the variety
of fragmentation routes is because of the large number of functional groups in the sugar molecule.
The mass spectrum of a methylated monosaccharide in the first place gives information about the number of
carbon atoms in the carbon chain of the sugar since an ion of maximum mass number (M − 31) is formed with
•
the expulsion of the glycosidic CH2OH(CHOH)3 C =O methoxy radical (— O CH3) from the ring of the molecular
H
ion (m/e : 250) and has a mass equal to 219 in the case of a hexose. In case of pentose the molecular ion peak is
observed at m/e : 206 while the (M – 31) peak is observed at m/e : 175. Thus in monosaccharides, the (M – 31)
rule can be fixed a criterion to identify the hexoses, pentoses, tetroses CH2OH(CHOH)2 C =O, etc. Secondly, the mass
H
spectrum makes it possible to measure the size of the oxide ring in sugar derivatives since various characteristic peaks
are obtained in pyranose (A) and furanose (B), for example, the rupture of side chain from the ring (b-fission):
CH3O CH2 CH3OCH2
(m/e : 45) O O
H (m/e : 89) CH3O CH
H H H
(m/e : 205) CH3O OCH3 (m/e : 161) H OCH3

OCH3 H OCH3 H
H OCH3 H OCH3
(A) (B)
System m/e M+•− mass

loss
(A) 205 250 – 45
45 250 – 205
(B) 161 250 – 89
89 250 – 161
It is also possible to determine the structure of a large number of sugar derivatives by mass spectrometric analysis. The
elucidation of the structure of partly methylated sugars is of great importance since they are obtained in the determination
of the structure of polysaccharides by the methylation method. In order to know the site of free hydroxyl groups in
partially methylated monosaccharide, they are treated with deuterated methyl iodide to replace the free hydroxyl groups
by the deuteromethoxy groups. The incorporation of three deuterium atoms per hydroxyl group will manifest itself in
the mass spectrum, the m/e values in the deuterated compound will increase by three units compared with the starting
partially methylated monosaccharide. Thus, from a simple comparison of the mass spectrum of undeuterated monosac-
charide with that of the deuterated monosaccharide,
one can draw a conclusion as to the locations of free CH2OH CH2OCD3
hydroxyl groups in the starting partially methylated O O
derivatives. For example, the methylglycosides of 2, H H
H H H H
3, 4-trimethyl glucose and 2, 3, 6-trimethylglucose CD3I
give deutero methyl derivatives A and B. CH3O OCH3 CH3O OCH3

OCH3 OCH3
The mass spectra of the deuterated compounds H H
are, compared with that of 2, 3, 4, 6-tetramethylg- H OCH3 H OCH3
lucose C, the parent compound; the fragmentation A
scheme of which has already been discussed earlier CH2OCH3 CH2OCH3
for ready reference. O O
H H
⎧A … 222 205 190 176 173 162 152 H H H H
⎪ CD3I
⎪
m/e ⎨B … 222 208 187 179 173 162 149 HO OCH3 CD3O OCH3
OCH3 OCH3
⎪ H H
⎪⎩C … 219 205
5 187 176 173 159 149 H OCH3 OCH3
H
173 = (205 – CH3OH), 159 = (187 – CO) B
A simple examination of the spectral data indicates that the spectra of compounds A and B not only differ from
one another but also from the parent compound C; the m/e values for a number of ions in A or B have grown
by three units as compared to their counterparts in C as expected. The peak at m/e equal to 205 in the parent
compound C appears at m/e : 208 in the deuterated derivative B and remain unchanged from A which shows that
fragmentation in the deuterated derivative A occurs at carbon atom 6 according to the following scheme:
•
A: m/e (253) ⎯− C H2OCD3 → m/e (205)
C: m/e (250) ⎯− C H2OCD3 → m/e (205)
Hence, the site of deuteromethoxy group in the deuterated derivative A is at carbon atom 6. Contrary to this, a
simple comparison of the mass peak at m/e: 187 in the parent compound with its counterpart in A and B reveals
that the fragmentation in the deuterated derivative B occurs at carbon atom 4 according to the following scheme.
•
B: m/e (253) ⎯− O CD3 → m/e (219) ⎯− CH3OH → m/e (187)
C: m/e (250) ⎯− O CH3 → m/e (219) ⎯− CH3OH → m/e (187)
Hence, the deuteromethoxy group is situated at carbon atom 4 in the deuterated derivative B.
The complicated problems encountered in the investigation of structure of carbohydrates have also been
solved by mass spectrometric analysis. For example, the structure of disaccharides can now be determined from
mass spectral data alone since a number of peaks in the mass spectra of disaccharides (listed here in the box)
with different types of bonds between monosaccharides can be differentiated.
CH2OH CH2OH
O O
H HO H
O H H
H OH H OH H H H OH
HO H HO H
H OH H OH O
H
H O H O OH H OH H
CH2OH CH2OH OH OH
Maltose H H
4-(a-glucosido)-glucose
4-(a-glucopyranosyl)-D-glucose
(b-anomer of maltose)
CH2OH CH2OH
O O
H HO H
H H
H OH O H OH O H H OH
HO H HO H O
H OH H O OH H
OH H OH
H H H H
CH2OH CH2OH H OH H OH
Cellobiose
4-(b-glucosido)-glucose
4-(b-D-glucopyranosyl)-D-glucose
(b-anomer of cellobiose)
CH2OH H OH
O
O
H H
H OH H
H OH H OH H H H H
HO H O HO H O
H OH H OH HO O
H OH
H H OH H HOH2C
O
CH2OH CH2OH H OH H
Trehalose
α, α- glucosidoglucoside
1-(α-D-glucopyranosyl)-α-D-
glucopyranoside
O
CH2OH CH2OH
HO H O O
H
OH OH H H
H H H H OH
O O HO
HO H HO H
O
HO H H
H H H
H H OH H OH H
CH2OH CH2OH
H OH H OH
Laclose (milk sugar)
4- (β-galactoside)-glucose
4-(β-D-galactopyranosyl)-D-glucose
(β- anomes)
CH2OH
CH2OH CH2OH H
O
H O H O
H
H OH
HO H O HO H O H
H OH H OH CH2OH
H H OH H O H HO
HO
CH2OH CH2OH OH H
H OH
Sucrose (saccharose)
α-glucosido-β-fructoside
(α-D-glucopyanosyl)-β-D-fructofuranoside
The electron impact mass spectrum of the anilide formed by the treatment of the acetate of an oligosaccharide
shows a stable molecular ion at m/e equal to 999. The first act of fragmentation of the molecular ion proceeds with
the subsequent elimination of the monosaccharide units from the reducing end (series R) or from the non reducing
end of the chain (series N) giving rise to primary fragments given here.
m/e (non reducing end): 331+ 288+ 380+•

m/e (reducing end): 364+ 288+ 347+•
Thus, the mass spectrum of anilide of acetate of oligosaccharide will exhibit peaks at m/e values indicated;
m/e : 999 (M+•), 380+•, 364+, 347+•, 331+, 288+

+
R2 R1
CH2OAc CH2 CH2
O O O
O O
NHC6H5
AcO AcO AcO

OAc OAc OAc
OAc OAc OAc
N1 N2
Further, one can differentiate between isomeric glucosylglucoses (molecular mass = 454) from mass spectra.
For example, some characteristic peaks of methylated isomeric glucosyl-glucoses containing 1 → 6 and 1 → 4
bonds are indicated below:
Methylgentiobiose m/e : ⎯ 353 ⎯ ⎯

(1 → 6 bond)
Methylcellobiose m/e 380 ⎯ 305 161
(1 → 4 bond)
Data reveal that we can differentiate between gentiobiose (1 → 6 bond) and cellobiose (1 → 4 bond), etc.
Differences in the mass spectra arise from the different routes of fragmentation of disaccharides caused by
breakdown of the reducing monosaccharide unit since this unit is replaced by the non reducing monosaccharide
in different positions at carbon atoms 6, 4, etc.
In conclusion, we can state that by identifying each fragment mass and suitably following them, one can
establish the sequence of monosaccharide units in the oligosaccharide chain. The primary fragments split into
smaller fragments, the structure of which depends on the type of glycosidic linkage. Thus mass spectral data
enables not only to determine the sequence of monosaccharide units in the chain but also the nature of bond
between them by differentiating 1 → 6 bonding units from 1 → 4 bonds, etc. Thus one can accomplish the entire
structure of oligosaccharides from mass spectral data alone.
(v) Structural Elucidation of Nucleosides Mass spectra of deoxyribosides are recorded by the direct introduction
of the sample into the ion source and the fragmentation pattern is shown in scheme 1. The pyrimidine or purine resi-
due (B) yields a prominent peak at m/e : M − 1. Sometimes rearrangement processes involving one or two hydrogen
atoms give rise to peaks at m/e : B + 1 and B + 2 respectively. These are removed preferentially from hydroxyl group as
indicated by the spectra of the O, N-per deuterio derivatives. Thus, the relative intensities of B + 1 and B + 2 peaks can
be fixed a criterion to differentiate between ribosides and 2′-deoxypentosides, while their mass indicates the identity
of the base.
The mass of the sugar fragment gives rise to a peak at m/e : 117 in deoxypentosides and at m/e: 133 in
pentosides. The ribose fragment generally is less abundant than the 2′-deoxyfragment.
Scheme 1: Mass spectra and fragmentation pattern of some nucleosides

MM = 244, Uridine MM = 228, deoxyuridine MM = 267, adenosine M = 251, deoxyadenosine
OH OH NH2 NH2
N N N N
N N
O O
N 111(B) N 111(B) N N N N
HOCH2 HOCH2 134(B) 134(B)
O 133(S) O 117(S) HOCH2 HOCH2
O 133(S) O 117(S)
HO OH HO
HO OH HO
m/e (peak height) m/e (peak height) m/e (peak height) m/e (peak height)
244 (1 mm) 228 (1 mm) 267 (3 mm) 251 (3 mm)
M − 89 (3 mm) M − 89 (2 mm) M − 89 (16.5 mm) M − 89 (11 mm)
B + 30 (9 mm) — B + 30 (26 mm) B + 30 (3.9 mm)
133 (S) (24 mm) 117 (S) (34.5 mm) 133 (S) (1 mm) 1 17 (S) (2 mm)
113 (B + 2H) (34.5 mm) 113 (B + 2H) (5 mm) 136 (B + 2H) (26 mm) 136 (B + 2H) (11 mm)
112 (B + H) (7 mm) 112 (B + H) (16 mm) 135 (B + H) (34.5 mm) 135 (B + H) (34.5 mm)
Peaks at lower mass (< m/e : 100) are mainly due to fragments of sugar moiety. The higher mass peaks at m/e :
M, M − 18 (loss of water), M − 89 and B + 30 are of interest. The loss of 89 mass units is indicated by the follow-
ing mechanism.
O H O H
B HO B
H2C B
H +
H H H
H H
H H
H
HO OH B + 30
OH OH
M 89
Thus, the 5′-deoxynucleoside fails to exhibit a peak at M − 89 due to the absence of 2′-hydroxy group. The
mass spectrum of bacterial RNA, in which the methyl group of methionine-methyl-d3 is incorporated into
thymine riboside shows a fragment peak at 129 instead of 126 (B + H) indicating thereby the incorporation
of the intact CD3 group.
The trideuterated riboside of 6-methyl-aminopurine shows a peak at m/e: 152 instead of 149 (B + H) for the
normal compound.
(vi) Structural Elucidation of Triglycerides Triglycerides are esters of glycerol with structural formula:
CH2OCOR1
CHOCOR2
CH2OCOR3
Glyceride of stearic acid, called tristearin is found in Ox tallow. For this glyceride, R1 = R2 = R3
= CnH2n+1, and n = 17. For glycerides of palmitic acid called tripalmitin; R1 = R2 = R3 and n = 15. Sources
of these glycerides are mutton, coconut oil, etc. Note that cow butter in addition to the glycerides indicated
above contains glycerides of butyric acid and other glycerides of even number of carbon atoms between
n = 4 and n = 15. Cephalin an phosphatide is a diglyceride of fatty acid is found in brain. It is contained in many
O
animals and plants (egg, liver tissue, soya bean, etc.) In this compound, R3 P O CH2 CH2 N+H3 the
O−
mass spectra of pure glycerides help to elucidate the structure of unknown glycerides. The principal fragmentation
pattern of a glyceride with general formula (A) is shown in Fig. 11.35.
R1,2 or 3 CO M – R1, 2 or 3 COO CH2 – OCOR1
CHOCOR2
R1,2 or 3 CO + 74
CH2OCOR3
M – R1, 2 or 3 COOH
R1,2 or 3 CO + 128
M – 18 M
M – R1 or 3 COOCH2
m/e
Fig. 11.35 Cracking pattern of triglycerides.
The base peak (M—R1, 2, 3,COO), in most cases, is due to the loss of acyloxy group from the complete molecule.
In mixed triglycerides, i.e. R1 ≠ R2 ≠ R3, a peak is observed by the loss of each acyloxy group present in the mol-
ecule. Peaks corresponding to M—R1, 2, 3,COOH ions are also distinguishable in the mass spectra. The second-most
intense peak is usually that of the positively charged acyl ion. A peak corresponding to each fatty acid present in the
molecule is observed. The peaks of smaller intensity 14 mass units lower are associated with M—R1, 2, 3,COO peak.
One of these peaks, M—(R2COO + 14) is much smaller than the M—R1COOCH2 and the M—R3COOCH2 peaks.
These compounds also show a set of peaks at m/e : 74 and 128 higher than each acyl ion peak.
All the triglycerides always show peak at M—18. The presence of a metastable peak corresponding to this
transition suggests that this is mainly due to ionic fragmentation and not to thermal cracking. These observations
led to infer that a complete structural analysis of triglycerides containing straight-chain fatty acids can be carried
out from their mass spectral data.
As an example, let us now deduce the structural formula of a triglyceride from its mass spectral data. The exact molec-
ular mass of hydrogenated optically active allene-containing tetraester triglyceride isolated from the seed oil of sapium
sebiferum is 920.801. Deduce the structural formula of the glyceride from its mass spectral data indicated below.
m/e Peak height (in mm)
920 <l
748 <l
637 12
(Continued)
(Continued)
607 7
593 1
425 3
395 9
371 3
341 5
297 3
267 13
171 3.4
155 18.5
0.801 + 0.0051 × 8
The number of hydrogen atoms in the molecule = = 108
0.0078
Thus, the empirical formula of the compound can be written as CX H108O8.
920 − 108 − 16 × 8 684

Since molecular mass = 920, the number of Cs = = = 57,
12 12
Accordingly, the formula of the compound is C57H108O8.
The general formula of the triglyceride can be written as
CH2OCOR1
CHOCOR2
CH2OCOR3
where, R′s = CnH2n+1.

Now the problem is to find the nature of R1, R2, and R3 from the mass spectral data. The parent peak of the
hydrogenated triglyceride is observed at m/e : 920. The peak at m/e : 155 is due to the acyl group (C9H19CO)
derived from decanoic acid while at m/e : 297 is due to the intact (C8 + C10)-acyl fragment, i.e. C17H33O2CO.
The peak at m/e : 267 is attributed to the acyl group derived from stearic acid is C17H35CO. The peaks at m/e :
341 ad 395 are respectively due to the fragments 74 and 128 mass numbers higher than the acyl group origi-
nating from stearic acid, i.e. C17H35O2CO + 74 = 341 and C17H35CO + 128 = 395. The peaks at m/e : 371 and
425 correspond to the fragments 74 and 128 mass numbers higher than the intact (C8 + C10) acyl group origi-
nating from decanoic acid, i.e. C17H33O2CO + 74 = 371 and C17H33O2CO + 128 = 425. The absence of peaks
at m/e : C9H19CO + 74 and C9Hl9CO + 128 suggests that decanoic acid is not esterified to the glycerol moiety.
The two peaks at m/e : 637 and 607 are respectively attributed to the fragments remaining after acyloxy loss
of stearic acid and the 18-carbon diester moiety, i.e.
+
C17H35COOCH2 CH2
C17H35COOCH C17H35COO + C17H35COOCH
C17H33O2COOCH2 C17H33CO2COOCH2
m/e : 637
C17H35COOCH2 C17H35COOCH2
C17H35COOCH C17H35COOCH + C17H33CO2COO
C17H33CO2COOCH2 +CH
2
m/e : 607
The peak at m/e : 593 is attributed to 920—C17H33O2COOCH2. This suggests that the C8 + C10 group is esterified
to a primary hydroxy group of glycerol, otherwise the peak at 593 should appear at 593 + 14 = 607 mass number.
C17H35COOCH2
C17H35COOCH C17H33O2COOCH2 + C17H35COOCH2
C17H33CO2COOCH2 C17H35COOCH2
+
m/e : 593
The peak at m/e : 748 corresponds to M—(C9H19COOH). Elimination of acid occurs via McLafferty rear-
rangement.
Accordingly, the final structural formula of the hydrogenated triglyceride is
CH3(CH2)16OCOCH2
CH3(CH2)16OCOCH
CH3(CH2)8CO2(CH2)7COOCH2
+
(vii) Structural Elucidation of Carotenes The mass
spectra of six isomeric carotenes and of lycopene show C20H28
an intense molecular ion peak at m/e : 536 (C40H56). The
peak at m/e : 480 is shown by all isomers containing the
m/e : 536
a-ionone ring. The peak appears due to ‘retro-Diels’
+
reaction.
The open-chain isomers (g- and d-carotene and lycopene) + C20H28 +
exhibit a peak at m/e : 467 due to the rupture of the doubly
allylic bond; the charge is again kept by the polyene frag- m/e : 480
ment. Strong peaks (m/e : 430 or 444) are observed due to +
elimination of toluene or xylene from the middle of the poly-
ene chain. The thermal production of toluene and xylene from C20H28
carotene is well known.
Thus the isomers of carotenes containing the a-ionone ring m/e : 536
and the open-chain isomers can be differentiated on the basis of
their mode of fragmentation. The former will exhibit a peak at
+ C20H28
m/e : 480 while the latter will show a peak at m/e : 467.
The xylene or toluene is produced as follows: CH2 H2C
+
m/e : 467
C9H15
H15C9
+ +
R1 R1
R3 R4
+ +
R1 R1
+ (R1+R2 = C30H44) + (R3+R4 = C30H44)
R2 R4
m /e : 430 m /e : 444
(l) Age of Rocks and Fossils The decay of radioactive elements is independent of the physical and chemical con-
ditions imposed on them. Although the decay of an individual particle from a given nucleus is a random process, the
gross decay of many nuclei sample provides a very convenient way of measuring times. If an igneous rock, formed
as a result of volcanic eruption, contained a small amount of uranium-238 with half-life of 4.5 × 109 years, it would
steadily decay leaving less uranium-238 and depositing more stable Pb-206 according to the scheme:
238
92 U 206
82 Pb
P b 8 42 He + 6 −01e
By measuring the ratio of uranium to lead in rock samples, an exact time can be determined for the origin of
geological deposits on earth and other planets.
Another similar method of dating is the decay of the isotope K-40 into Ca-40 or Ar-40 by the positron
emission, i.e.
+
18 Ar + 1 β + v
→ 40
40
19 K
with the half-life of 1.3 × 109 years. Because it is more readily measured, Ar-40 is most often used to date the
geological samples.
Thus, if the decay rate of 238U to 206Pb, 40K to 40Ar or 87Rb to 87Sr is known, and the ratio of the isotopes of one
of these pairs (depending upon the nature of the geological deposit) in the reference sample and in the unknown
sample understudy is determined by mass spectrometry then the age of the minerals and rocks can be estimated.
The other most important method of absolute dating, sometimes called chronometric dating, is that of radiocar-
bon dating. It was developed by the American chemist Willard Libby and his co-workers in 1949. Radiation from
space produces neutrons that enter the earth’s atmosphere and react with nitrogen to produce the carbon isotope
C-14. Since all plants use CO2 from the atmosphere for growth, a portion of the carbon in plants is radioactive C-14
and the plants are slightly radioactive. When a plant dies, no additional C-14 is taken in, and that within the plant
body begins to decay without being replaced. Measurements of the relative amounts of C-14 and C-12 in an organic
archeological sample by mass spectrometry provides a sensitive method of dating. Recently, a new approach to
radio carbon dating using an Accelerator Mass Spectrometer (AMS) has been developed. AMS directly counts C-14
atoms, rather than counting rates of disintegration. AMS can date a sample as small as a single kernel of grain or a
fleck of wood preserved inside a bronze axe socket. This method can date items that are up to 90,000 years old.
Another element that is continuously produced in the atmosphere by cosmic impact is tritium, 3 H. It is formed
by neutron collision with 147 N according to
14
7 N 1
0 n ⎯⎯
→ 126 C 3
1 H
Tritium, the heavy isotope of hydrogen, decays with a half-life of 12.4 years by b decay into the stable helium
isotope 23 He :
3
1 H → 23 He + −01 β + v
A small percentage of water contains 31 H instead of 11 H and thus by measuring the relative amounts of 31 H and
1
1H in samples of water by mass spectrometry at various locations, some indication may be obtained as to how
long this water was in the form of rain.
We know that the rate of radioactive decay is directly proportional to the number of atoms present at time t,
dN dN
i.e. − ∝ N or − = λN
dt dt
which on integration gives
1 N0
t= ln (11.50)
λ N
0.693
where, λ = , is the decay or disintegration constant, N0 is the number of radioactive atoms at time t = 0, N
t1/ 2
is the number of radioactive atoms at time t and t1/2 is the half-life of the radioactive isotope.
The expression (11.50) when used in mass spectrometry modifies to
⎡ ⎤
⎢ Ratio of relative abundance of parent and ⎥
1 ⎢ daughter isotopes
s in the standard sample ⎥
t = ln ⎢ ⎥ (11.51)
λ ⎢ Ratio of relative abundance of parent and ⎥
⎢ daughter isotopes in the sample under study
dy ⎥
⎣ ⎦
To summarise, precise age dating is based on the rate of decay of radioactive nucleids. If the decay rate of 238U
to 206Pb; 40K to 40Ar, and 87Rb to 87Sr is known, and the ratio of one of these pairs is measured, the age of minerals
and rocks can be determined. Similarly, the age of the plant fossil is determined from the 14C/12C ratio in the plant
fossil and in the living organisms.
The age of the minerals can also be determined by another method, also based on radioactive decay, known as
the isotope dilution method.
Isotope Dilution Method In this method, an element like Li, C or N which has isotopes and whose natural abundance
is small cannot be estimated by spiking it with that isotope. This is similar to the radioactive tracer technique.
Let us say, we have to estimate oxygen in a substance A which is in a tracer.
Let the mass of the substance (A) taken be ‘a’; oxygen content of it be x per cent.
Let us add enriched oxygen ‘b’ or 18O content of m per cent. After equilibration, let 18O content becomes n per cent.
Thus, axn = b(m − n)
b ( m − n )100
or x= (11.52)
na
Li in granite has been determined by using Li tracer. In determining the geological ages of minerals, potassium
minerals contain argon due to
40
K Positron
ddecay
→ 40A
Ar
Since, 40Ar content is very small, it is mixed with 36Ar and by the 36Ar content of the mixture, one can determine
the age of the mineral. By the isotope dilution method, one can determine quantities as small as 1011 atoms.
Problem 11.39: The mass spectrum of an aged sample of rain water collected from a pond is characterised by the
presence of peaks of the isotopic forms of hydrogen 1HOH and 3HOH, which are equal to 60 and 0.4 mm respectively,
while the mass spectrum of fresh sample of rain water exhibits peaks corresponding to 1HOH and 3HOH equal to 50
and 0.6 mm respectively. Calculate for how long this water was in the rain form? The half-life of 3H is 12.4 years.
Solution Since, t1/2 = 12.4 years (for 31 H )
0.693 −1
Therefore, λ = yr = 5 87 × 10 −2 yr −1
12.4
Substituting the value of l and the data given in the problem into Eq. (11.51), we obtain
1
t= ln[(0.60 / 50) /(0.40 / 60)]
5.58 × 10 −2 ( yr −1 )
ln1.8 5.87 × 10 −1
= = = 10.5 yrs.
5.. × −2 ( yr −1 ) 5.. × −2 ( yr −1 )
Problem 11.40: The 14C/l2C ratio in a plant fossil as well as in a living organism has been examined on an
isotope ratio mass spectrometer. The 14C/12C ratio in the former has been found to be 0.617 times than that found
in the latter. How old is the plant fossil? t1/2 for 14C is 5770 years.
0.693
Solution The decay constant l for 14
C= = 1.20 10 −44 ( yr 1 )
5770( yr )
Substituting the value of l and the given data into Eq. (11.51), we obtain,
1 1
t, the age of the fossil = 4 1
ln
1.20 10 ( yr ) 0.617
= 8.33 × 103 (yrs) × ln 1.620

= 8.33 × 103 (yrs) × 4.82 × 10−1 = 4015 years
(m) Quantitative Analysis of a Mixture of Gases The procedure employed in quantitative analysis by mass
spectrometry is basically the same as used in infrared spectrometry and in ultraviolet spectrometry. Let us suppose
we want to analyse a mixture of four gases, namely N2, O2, CH4 and CO2. About one cc of the gas sample at NTP is
introduced into the mass spectrometer and let its pressure inside the ion chamber be 100 microns (1 m = 10−3 torr). The
mass spectrum will exhibit molecular ion peaks at m/e : 16, 28, 32 and 44. In addition, there will be peaks due to the
fragmentation of molecular ions into ions like 12−15 for CH4, and 28 for CO2. For analysis, peaks are selected on the
basis of both intensity and non-interference by peaks of another component. If possible, monocomponent peaks (per-
haps of parent ion peaks) are chosen. It is assumed that• at the pressure inside the ion chamber (i.e. = 100 m) there are
no ion-molecule reactions (e.g. CH4 + CH4+ → CH5+ + C H 3 , etc.) and peak heights represent the original composition.
But the direct peak height ratios of monocomponents cannot be fixed as the criterion to determine the composition as
these ratios will actually be that of ions and not that of parent molecules. Therefore, peak heights need be standardised
with the parent gases. Different molecules have different ionisation energies and different rates of diffusion through
molecular leaks. Further peak at m/e : 28 is not due to N2 alone but has contribution from CO2 also. So the latter has to
be subtracted out in order that the total sum of the contributions of all components of the mixture comes out to nearly
100 per cent. For standardisation, 100 m of each gas is taken and their peak heights are measured at the respective
chosen values of their masses. The peak heights per micron known as the sensitivity factors or parameters (S) are then
determined. Thus for CO2 (m/e = 44), if the measured peak height is h1, then h1 = S1pl where p1 is the partial pressure of
CO2 in microns of Hg, for O2; h2 = S2p2 and for nitrogen, h3 = S3p3 = S1p1 where S1p1 is the contribution from CO2. Once
the sensitivity values of the pure components are known, all the partial pressures due to the components of a mixture
of gases can be determined from peak heights. Calculations of sample components become simple if the components
of the mixture give spectra with at least one peak whose height is due entirely to the monocomponent.
In general, we can write
Height of the analysis peak in the pure compound
o
Sensitivity ( ) = (11.53)
Pressure of pure compound in the sample chamber
Partial pressure due Height of the monocomponent (i.e. analysis ) peak in tthe mixture
to each component = (11.54)
of a mixture Sensitivity factor
Mole per cent of Partial prressure of each component in the mixture

each component = (11.55)
in the mixture Sum off the partial pressures due to all the components of a mixture
If the mixture has no monocomponent peak, then simultaneous equations are set up from the coefficients (per
cent of base peak) at each analysis peak. If there are n-components in a mixture, there will be one equation for
each compound in the mixture with n-terms.
Let x1, x2, x3..., xn be the relative contributions respectively of the components A1, A2, A3 ..., An of a gas mixture.
If h1, h2, h3 ..., hn are the peak heights of the monocomponent peaks at m/e : m1/e, m2/e, m3/e,..., mn/e of the pure
compounds A1, A2, A3 ..., An, constituting the mixture then we can write
mix
⎛ m1 ⎞
⎜⎝ ⎟⎠ : h mix
n x 1h1′ x 2 h2′ x 3 h3′ + ... + x n hn′
e
mix
⎛ m2 ⎞
⎜⎝ ⎟⎠ : h2mix x 1h1″ x 2 h2″ x 3 h3″ ... x n hn″
e
mix
⎛ m3 ⎞
⎜⎝ ⎟⎠ : h3mix x 1h1″′ x 2 h2″′ x 3 h3″′ ... x n hn″′
e
mix
⎛ mn ⎞
⎜⎝ ⎟ : hnmix x 1h1n x 2 h2n x 3h3n ... x n hnn (11.56)
e ⎠
where (m/e’s)mix are the respective peaks of the mixture.
The solution of these equations will give the values of x1, x2, x3..., xn. If S1, S2, S3 ..., Sn are the sensitivity parameters
of the respective components of the mixture then the partial pressures due to the constituent components in microns
can be obtained by dividing the values of x’s by the sensitivity parameter values S’s.
For fast calculations, a set of equations in terms of each unknown and the analytic masses is obtained by
inversion of the coefficient matrix, i.e.
x1 = y 1⬘ ( h1 ) mix ⬘
h ) i ⬘
3 h3 ) i
+ + ⬘
n hn ) mix
x 2 = y ⬘⬘( h1 ) mix + y ⬘⬘( h2 ) mix + y ⬘⬘3 ( h3 ) mi + + ⬘⬘

n ( hn )mix
x 3 = y ( h1 ) mix + y ⬘⬘⬘( h2 )mix + y ⬘⬘⬘

3 ( h3 )
mix
+ ... + y ⬘⬘⬘
n ( hn )
mix
x = y 1n ( h ) mix y 2n (h
( h ) mix y 3n ( h ) mix y nn ( hn ) mix
Substituting the values of peak heights from the spectrum of the mixture into the inverse matrix equations, we
can obtain the relative contributions of each component of the mixture. The sum of the partial pressures determined
should be equal to the total sample pressure, the discrepancy if any would indicate an unexpected component or a
change in operating sensitivity.
Many solids like alloy steels, rocks, semiconductor, reactor materials, etc., have been analysed for trace metals
and quantities as small as one in 109, i.e. nanogram quantities have been estimated.
A computer program in C++ entitled ‘Mole-PER.CPP’, based on Gause Jorden method to determine the
constituent components of a mixture has also been given and the results of the problems 11.41−11.43 determined
manually by simple elimination method are compared with those determined using the program. The values
within braces in the tables are from the program.
Problem 11.41: Determine the contents of the components in an alcohol mixture, using the mass spectral data
of pure components and the intensities of the peaks in the mass spectrum of the mixture given below. The sensitivity
parameters (in divisions/10−3 torr) for methyl and ethyl alcohols respectively are 8.76 and 17.98.
m/e Percentage of component Peak intensity
CH3OH C2H5OH
15 35.48 9.44
19 0.29 3.13
27 ⎯ 21.62
29 58.80 21.24
31 100 100
32 68.03 1.14 400
33 0.98 —
43 7.45
45 37.33
46 16.23 800
Solution According to the set of Eqs (11.56), we write
68.03x1 + 1.14x2 = 400
16.23x2 = 800
where x1 and x2 are the relative contributions of methyl and ethyl alcohols respectively.
Solving for x1 and x2, we obtain
x1 = 5.05; x2 = 49.29
In order to determine the mole per cent of the components, we proceed as follows:
Component Values Sensitivity, Partial pressure Mole
of x (in divisions/micron) (in microns) p = x/S per cent
CH3OH 5.05 8.76 0.576 17.37
(5.05)
(5. (0.576) (17.354)
C2H5OH 49.29 17.98 2.74 82.62
(49.29) (2.746) (82.645)
∑3.316 (3.323)
Problem 11.42: Determine the mole per cent of each component in a mixture, using the mass spectral data of pure
components and the intensities of the peaks in the mass spectrum of the unknown mixture given below. The sensitivity
parameters (in divisions/micron) respectively for methyl, ethyl, n-propyl alcohols are 8.76, 17.98, 26.51.
m/e Percentage of component Line intensity in the

CH3OH C2H5OH n-C3H7OH spectrum of the mixture
15 35.48 9.44 3.77

19 0.29 3.13 0.90
27 ⎯ 21.62 15.20
29 58.80 21.24 14.14
31 100 100 100
32 68.03 1.14 2.25 2600
33 0.98 ⎯ ⎯
39 ⎯ 4.00 3000
43 7.45 3.18
45 37.33 4.39
46 16.23 ⎯ 800
59 9.61
60 6.63
Solution Let x1, x2 and x3 be the contributions of the methyl, ethyl and n-propyl alcohols respectively.
Proceeding as in the above problem, we write
2600 = 68.03x1 + 1.14x2 + 2.25x3
3000 = 4x3
800 = 16.23x2
Solving for x1, x2 and x3, we get
x2 = 49.29; x3 = 750, x1 = 12.58
Component Values of x Sensitivity Partial pressure Mole

(in divisions/m) (in m); p = x/S Per cent
CH3OH 12.58 8.76 1.43 4.40
(12.58) (1.43) (4.42)
C2H5OH 49.29 17.98 2.74 8.44
(Continued)
Component Values of x Sensitivity Partial pressure Mole

(in divisions/m) (in m); p = x/S Per cent
(49.29) (2.74) (8.44)
n-C3H7OH 750 26.51 28.29 87.15
(750) (28.30) (87.13)
∑32.46 (32.480)
Problem 11.43: Determine the mole per cent of each component of a gas mixture, using the mass spectra of
pure components and the intensities of the lines in the mass spectrum of the mixture given below in the tabular
form. The sensitivity coefficients S for n-butane, isobutane, propane, ethane and methane are, respectively, 0.396,
0.464, 0.335, 0.392, and 0.246.
Line intensiy in the

m/e Percentage of component
spectrum of the mixture
n-butane isobutane Propane Ethane Methane

15 5.30 6.41 6.19 4.57 85.6 167.3
16 0.12 0.18 0.15 0.08 100 134.2
28 32.6 2.62 59.1 100 ⎯ 757.2
30 0.98 0.13 2.20 26.2 ⎯ 130.1
31 ⎯ ⎯ ⎯ 0.55 ⎯ 2.4
43 100 100 22.8 ⎯ ⎯ 207.4
44 3.33 3.33 29.0 ⎯ ⎯ 145.4
50 1.29 0.89 ⎯ ⎯ ⎯ 0.6
55 0.93 0.42 ⎯ ⎯ ⎯ 0.7
57 2.42 3.0 ⎯ ⎯ ⎯ 2.5
58 12.3 2.73 ⎯ ⎯ ⎯ 5.0
Solution Let x1, x2, x3, x4 and x5 are the contributions of n-butane, isobutane, propane, ethane and methane
respectively.
Proceeding as in the above problems, the base peaks chosen are m/e : 16, 30, 44, 57 and 58.
Accordingly, we write
m/e : 16,134.2 = 0.12x1 + 0.18x2 + 0.15x3 + 0.08x4 + 100x5
m/e : 30,130.1 = 0.98x1 + 0.13x2 + 2.20x3 + 26.2x4
m/e : 44,145.4 = 3.33x1 + 3.33x2 + 29.0x3
m/e : 57,2.5 = 2.42x1 + 3.0x2
m/e : 58,5.0 = 12.3x1 + 2.73x2

Solving this set of equations. for x1, x2, x3, x4, and x5, we obtain
x1 = 0.270; x2 = 0.615; x3 = 4.961; x4 = 4.535 and x5 = 1.3295.
Component x S p = x/S Mole per cent

n-butane 0.270 (0.271) 0.396 0.681 (0.686) 2.01 (2.03)
isobutane 0.615 (0.614) 0.464 1.325 (1.323) 3.92 (3.93)
propane 4.961 (4.912) 0.335 14.808 (14.662) 43.82 (43.56)
ethane 4.535 (4.539) 0.392 11.568 (11.581) 34.24 (34.41)
methane 1.3295 (1.329) 0.246 5.404 (5.404) 15.99 (16.05)
Σ33.786 (33.656)
Program Name: MOLE_PER.CPP

Purpose: This program in C++ is for quantitative analysis of mixtures by mass spectrometry.
Compiled by: Hardeep Kaur Randhawa, Assistant Professor, Department of Computer Science and
Applications, Kanya Maha Vidyalaya, Jalandhar.
#include <iostream.h>
#include <conio.h>
#include <stdlib.h>
#include <iomanip.h>
const int SIZE = 35;
class mol_per
{
private:
int n;
float h[SIZE][SIZE], s[SIZE], x[SIZE], mol_pt[SIZE], par_pr[SlZE];
public:
void input_data( );
void cal_rel_cont( );
void cal_mol_p( );
void output( );
};
class out1
int xl, y1;
public:
out1 (int i1, int j1) {x1=i1; y1=j1;}
~ out1( ){ }
friend ostream & operator<<(ostream &stream, out1 o1);
};
class out2
{
int x2, y2;

public:
out2(int i2, int j2) {x2=i2; y2=j2;}
~ out 2( ) { }
friend ostream &operator<<(ostream &stream, out2 o2);
};
//Function to enter data
void mol_per:: input_data( )
{
int i, j;
cout<<′′\n INPUT : \n”;
cout<<′′\nEnter the number of constituent components of the mixture (n) : ′′;
cin>>n;
cout<<′′\nEnter the coefficients, constant term and the sensitivity for′′;
cout<<′′each constituent of the mixture : \n\n";
for (i=0; i<n; i++)
{
for (j=0; j<=n; j++)
cin>>h[i][j];
cin>>s[i];
}
}
//Function to calculate relative contributions of the components of a mixture
void mol_per :: cal_rel_cont( )
{
int k, m, p, q, i, j;
float temp, temp1, max;
for (k=0; k<n; k++)
{
max=abs(*(*(h+k)+k));
p=k;
for(m=k+1; m<n; m++)
{
if (abs(*(*(h+m)+k))>max)
{
max=abs(*(*(h+m) + k));
p = m;
}
}
if(p!=k)
{
for (q=k; q<=n; q++)
{
Temp1 = *(*(h+k)+q);
*(*(h+k)+q) = *(*(h+p)+q);
*(*(h+p)+q) = temp1;
}
}
temp = *(*(h+k)+k);
if (temp= =0)
{
cout<<′′h[′′<<k+1<′′][<<k+1<<′′] = ′′<<temp<<endl;
cout<<′′Division by zero is not permissible′′<<endl;
exit(0);
}
for (j=k; j<=n; j++)
*(*(h+k)+j) = *(*(h+k)+j)/temp;
for (i=0; i<n; i++)
{
if(i!=k)
{
temp = *(*(h+i)+k);
for(j=k; j<=n; j++)
*(*(h+i)+j) = *(*(h+i)+j) − temp *(*(*(h+k) + j));
}
}
}
for (i=0; i<n; i++)
{
*(x + i) = *(*(h+i)+n);
}
}//End of member function cal_rel_cont( )
//Function to calculate partial pressure and molepercent
void mol_per :: cal_mol_p( )
{
int i;
float sum;
sum=0.0;
for(i=0; i<n; i++)
{
*(par_pr+i) = *(x+i)/(*(s+i));
sum = sum + *(par_pr+i);
}
for(i=0; i<n; i++)
{
*(mol_pt+i) = (*(par_pr+i)*100)/sum;
}
}//End of member function cal_mvl_p( )
ostream &operator« (ostream &stream, out1 o1)
{
register i, j;
stream <<”\n”;
for(j=l; j<=o1.y1; j++)
{
stream<<′′*′′;
for(i=0; i<o1.x1; i++)
stream<<′′−′′;
}
stream<<′′*′′<<′′\n′′;
return stream;
}
ostream &operator<<(ostream &stream, out2 o2)
{
register i2, j2;
for(i2=1; i2<=o2.x2; i2++)
stream<<′′ ′′;
stream<<′′|′′;
for(j2=1; j2<o2.y2; j2++)
stream<< ′′ ′′;
return stream;
}
//Manipulator to print ′′Press any key to continue′′
istream &cont(istream &stream)
{
cout<<′′\nPress any key to continue :\n′′;
getch( );
return (stream);
}
//Manipulator to clear the screen
ostream &clr_scr(ostream &stream)
{
clrscr( );
return(stream);
}
//Function to pause the output
void stop(int i)
{
out1 ob(25, 3);
if (i==7 || i==15 || i==23 || i==31)
{
cin>>cont;
cout<<clr_scr;
cout<<ob;
}
}
//Manipulator to set the format
ostream &fmt(ostream &stream)
{
stream.setf (ios::fixed);
stream.width (10);
stream.precision (5);
stream.setf(ios::showpoint);
return stream;
}
//Function to display output
void mol_per::output( )
{
Out1 a(25, 3), *pa;
out2 b03(0, 3), b25(2, 5), b0_10(0, 10), b69(6, 9), b0_13(0, 13), b99(9, 5);
out2 b57(5, 7), b70(7, 0), b80(8, 0), b78(7, 8), b89(4, 9);
out2 *pb03, *pb25, *pb0_10, *pb69, *pb0_13, *pb99, *pb57, *pb70;
out2 *pb80, *pb78, *pb89;
int i;
pa = &a;
pb03 = &b03; pb25 = &b25; pb69 = &b69; pb0_10 = &b0_10;
pb80 = &b80; pb99 = &b99; pb57 = &b57; pb0_13 = &b0_13;
pb70 = &b70; pb78 = &b78; pb89 = &b89;
cin>>cont;
cout<<clr_scr;
cout<<′′\n OUTPUT : \n′′;
cout<<*pa;
cout<<*pb03<<′′Relative Contribution′′<<*pb25<<′′Partial Pressure′′<<*pb57<<′′Mole Per cent′′
<< *Pb70;
cout<<endl<<*pb0_13<<′′x[i]′′<<*pb99<<′′par_pr[i] (micron)′′<<*pb89<<′′,mol_pt[i]′′<<*Pb80;
cout<<*pa;
for(i=0; i<n; i++)
{
cout<<*pb0_10<<fmt<<x[i]<<*pb69<<fmt<<par_pr[i]<<*pb78<<fmt<<mol_pt[i]<<*pb80;
cout<<*pa;
stop(i);
}
cin>>cont;
}//End of function output( );
void main( )
{
mol_per obj;
char ch;
cout<<clr_scr;
do
{
obj.input_data( );
obj.cal_rel_cont( );
obj.cal_mot_p( );
obj.output( );
cout<<′′\nDo you want to solve any other equation?′′;
cout<<′′Enter y or Y for yes and any other key for no : ′′;
ch = getche( );
}while (ch = = ‘y’ || ch == ‘Y’);
}
/*End of the program*/
Let us solve problem 11.46 by using this program.
This program accepts a number of components, coefficients, i.e. peak heights of the pure components at
a particular selected value of the mass number, the constant term, i.e. the peak height of the respective component
in the spectrum of the mixture at the selected mass number, the sensitivity (in divisions/m) corresponding to the
respective constant. The program displays relative contributions, partial pressures (m) and composition of the
mixture in mole per cent.
The output of one run of the program is given here.
INPUT:
Enter the number of constituent components of the mixture (n) : 5
Enter the coefficients, constant term and the sensitivity for each constituent of the mixture:
0.12 0.18 0.15 0.08 100 134.2 0.396
0.98 0.13 2.2 26.2 0.0 130.1 0.464
3.33 3.33 29.0 0.0 0.0 145.4 0.335
2.42 3.0 0.0 0.0 0.0 2.5 0.392
12.3 2.7 0.0 0.0 0.0 5.0 0.246
Press any key to continue:
Output:
Relative Contribution Partial Pressure Mole per cent
x[i] Par_pr[i](micron) Mol_pt[i]

0.27169 0.68608 2.03832
0.61417 1.32365 3.93252
4.91207 14.66290 43.56317
4.53997 11.58157 34.40859
1.32957 5.40475 16.05739
Press any key to continue:

Do you want to solve any other equation? Enter y or Y for yes and any other key for no : n
Note:
• While entering the data it should be kept in mind that there should be one blank space between any two consecutive
values and at the end of each line of data entry, press Enter.
• This program can handle a mixture containing a maximum of thirty-five components. This number can be increased
by changing the value of SIZE.
(n) Nature and Amount of Impurities The mass spectrum of the impure compound, say X, is recorded and
the relative abundance of the various peaks is noted. From the positions of the peaks, the nature of impurities in
the sample X is accomplished. In order to determine the amount of impurities in the sample X, the impurities in the
standard sample are determined by chemical methods and the mass spectrum of the standard sample is recorded.
Just by comparing the mass spectral data of the impure compound X with that of its standard sample, the amount
of impurities in the impure sample of the compound can be determined.
Problem 11.44: In the mass-spectrometric determination of impurities in gallium arsenide, the lines were
obtained for singly charged ions with mass numbers (m/e) equal to 16, 56, and 112 and the following values of
the signals, respectively, 15, 35, 48 mV. The standard sample of gallium arsenide in which these impurities were
determined by chemical means was found to contain 2 × 10−5, 5 × 10−4, and 3 × 10−4 per cent of the respective
impurities and gave signals at 10, 50 and 72 mV.
Determine the nature of these impurities and their contents in the sample. (in percent).
Solution The peaks at m/e : 16, 56 and 112 in the mass spectrum of gallium arsenide correspond to singly
charged oxygen (O+), iron (Fe+) and cadmium (Cd+) ions respectively. We know that the peak height is a measure
of the relative abundance of an ion. Accordingly,
2 10 5 15( )
the amount of O+ in the sample = = 3 0 × 10 −5 per cent
10( )
5 10 4 35( )
the amount of Fe+ in the sample = = 3 5 × 10 −4 per cent
50( )
3 10 4 48( )
and of Cd+ in the sample = = 2 0 × 10 −4 per cent
72( )
(o) Reaction Kinetics Mass spectrometry is used for the study of fast reactions, free radicals; equilibrium
processes and exchange reactions. By using variable rate of gas flow, the time interval for the disappearance of a
precursor ion and for the formation of a daughter ion can be calculated. From the peak height in the spectrum, the
concentration of species at different time intervals can easily be calculated. A plot of time t against ln(peak height)
will be a straight line with slope equal to k, the specific rate constant. The activation energy Ea of the process can
be determined from the Arrhenius equation,
k Ze − Ea / RT (11.57)
provided the experiment is performed at least at two different temperatures, say T1 and T2.
k 2 Ea ( )
Accordingly, ln = (11.58)
k1 RT1T 2
RT1T 2 k
or Ea = i ln 2 (11.59)
(T 2 T1 ) k1
11.16.2 Biological Applications of Mass Spectrometry

Although EIMS proved to be an invaluable method for the analysis of volatile compounds up to about 1000 Da, it is
now possible to use other ‘soft’ ionisation techniques for analysis of nonvolatile compounds such as proteins, nucleic
acids, and other compounds of biological importance with molecular masses up to and in excess 100,000 Da. Tech-
niques such as ESI, IS, MALDI and FAB mass spectrometry have proved to be very potential tools for analysis of pro-
tein molecular masses, enzyme-substrate complexes, antibody–antigen binding, drug-receptor interactions, and DNA
oligonucleotide sequence determinations. The ions produced by these techniques are separated by quadrupole mass
filters, ion traps, time of flight and Fourier transform-ion cyclotron methods. For example, the electron-spray process
proved to be very useful for the molecular mass of macromolecules. The initial droplets produced by the electron-spray
process usually have positive charges. Protonated or deprotonated droplets evaporate. In the nacked macromolecule
ions, entering the mass analyser, this charge is typically manifested as variable states of protonation at proton-accessi-
ble sites. The phenomenon results in ions with varying charges in the range +2 up to + 40 or even higher. According to
the multiple charge theory of Fenn (1989), the different charge states are interpreted as independent measurements of
molecular mass and an averaging method based on the solution of simultaneous equations provides accurate molecular
mass estimations of macromolecules. The molecular mass so determined from the sequence is the chemical molecular
mass, i.e. the average value based on the natural abundance of isotopes. The molecular masses of some typical systems
are listed here in Table 11.19.
Table 11.19 Molecular masses of some biological systems.
Bimolecule Molecular mass (Da or amu) Chares m/e
Insulin (bovine) 5,730 4−6 950−1450
Cytochrome C (horse
12,400 12−21 550−1100
heart)
Alcohol dehydrogenase
39,800 3 2−46 800−1300
(horse liver)
a-Amylase(bacterial) 54,700 35−58 880−1550
Conalbumin (Chicken egg) 76,000 49−64 1200–>1500
Fragmentation in Proteins and Peptides Proteins and peptides are linear polymers made up of twenty most
common amino acids linked with each other by peptide bonds. However, the proteins produced by ribosomes may
undergo covalent modifications, called post-translational modifications, after its incorporation of amino acids.
Over 200 such modifications have been detected already, and the most important are indicated below:
Post- translational modification Mass difference (Da)
Methylation 14.03
Propylation 42.08
Sulfation 80.06
Phosphorylation 79.98
Glycosylation 146.14
Carboxymethylation 58.04
Cysteinylation 119.14
The fragments that result are generally analysed by tandem mass spectrometry using soft ionisation techniques.
The first fragments that are identified are produced by cleavage of a bond in the main chain. The cleavage of a
bond in that peptide chain can occur in either of three types of bonds Ca⎯C, C⎯N or N⎯Ca which yields six types
of fragments that are labelled an, bn and cn, when a positive charge is kept by N–terminal side and xn, yn, and zn when
positive charge is retained by the C-terminal side. The cn and yn fragments implicate the transfer of two extra hydrogen
atoms, the first one responsible for the protonation and a second one originating from the other side of the peptide.
The subscript ‘n’ indicates the number of amino acids present in the fragments. Main fragmentation paths of peptides
in CID/MS/MS(CID/MSn) are:
+
an bn cn
R R R
H2N CH CO (NH CH CO)n NH CH COOH + H
xn yn zn
R Rn Rn R
+ +
H (HN CH CO)n-1 NH CH + CO HN CH CO (NH CH CO)n−1 OH
an xn
R Rn Rn R
+ +
H (HN CH CO)n-1 NH CH C O H3N CH CO (NH CH CO)n−1 OH
bn yn
R Rn Rn R
+ +
H (HN CH CO)n-1 NH CH CO NH3 CH CO (NH CH CO)n−1 OH
cn zn
The mass difference between consecutive ions within a series allows one to determine the identity of the consecutive
a/a and thus deduce the peptide sequence. There are two exceptions: Leu-Ileu which are isomers; and Gly-Lys, which
are isobars. Normally, the spectra show several incomplete series of ions that produce redundant data and make the
spectrum very complex.
There is also a marked difference between the fragmentation at high and low energies.
Two other types of fragments found in most spectra result from the cleavage of at least two internal bonds in the
peptide chain. This type of ions is often only weakly abundant and since they contain more than three or four a/a
residues, they appear in the spectrum among the low masses; the exception proline contains imino group is included in
a five-atom ring and thus has higher proton affinity than other peptide bonds. The second type of fragments requiring a
cleavage of two bonds or a multiple cleavage of peptidic chain appearing among the low masses in the spectrum. These
are immonium ions of a/a, labelled by a letter corresponding to the parent a/a code.
an bn cn +
R R
H2N CH CO (NH CH (CO)n NH CH COOH + H
xn yn zn
R R>1 R
+ +
H2N CH H2N CH (CO NH CH) m<n C O
Immonium ion Internal fragment
In addition to the ions described earlier, three new types of fragments that require the cleavage of the peptidic
chain and amino-acid lateral chain are highlighted only in the high energy spectra. These fragments are useful to
distinguish between isomers, e.g. Leu-Ileu.
Two of the fragment types originate from the cleavage of the bond between the b and g carbon atoms of the side
chain of a/a. These fragments are denoted by dn or wn, respectively, according to whether the positive charge resides
R″
H+ CHR′ H+ CHR′
H (NH CHR CO)n−1 NH CH H (NH CHR CO)n−1 NH CH + R″

dn
an+1
H Rn
NH CH CO (NH CHR CO)n−1 OH H Rn + NH CH CO (NH CHR CO)n−1 OH
R″ yn vn
R′ HC H+
CH CO (NH CHR CO)n−1 OH R″ + R′HC H+

zn + 1
CH CO (NH CHR CO)n−1 OH
wn
Fragmentation paths yielding the ions characteristic of a/a-a/a lateral chain
on N-terminal or the C-terminal fragment. They are useful to distinguish between isomers. Amino acids carrying an
aromatic group attached to the b-carbon either do not display these two fragments or have very low abundance.
The other type of fragments results from the complete loss of the side chain and is denoted by vn. This fragment
containing the C-terminal moiety is intense for a/a that do not easily yield wn type fragments. No equivalent con-
taining the N-terminal moiety has ever been observed.
The charge and fragmentation depend upon the nature of the peptide. The rule describing the relationship
between the position and charge localisation on the one hand and fragmentation process on the other is
Less localised ⎯Positive charge localisation → More localised

c
N-acetylated Peptides without His, Lys Arg N+ peptides

peptides without basic amino acids peptides peptides
basic a/a. anbndn
N-terminal
bnyn Important ions
C-terminal
yn vn wn
Charge and fragmentation with respect to nature of peptide.
Some of the strategies adopted for identification, sequencing, identification and localisation of post-translation
modification, verifying of structure and purity of peptides and proteins are listed here (1-6).
PROTEIN
Enzymatic cleavages
Enzyme I Enzyme II Enzyme III
Mixture of Mixture of Mixture of

peptides 1 peptides 2 peptides 3
Reversed-phase HPLC saparation of each mixture

in a few fractions: mixtures of a few peptides
with close lipophilicity
m fractions n fractions p fractions
For each fractons:

derivatisation if needed
MS: Molecular mass of each peptide
MS/MS: sequence of each peptide from
each fraction
Comparison of the results:

sequency deduction
1. Strategy followed in sequencing proteins using mass spectrometry.

Unknown protein(S)
Enzymatic digestion
Mixture of proteolytic peptides
Determination of mass
of each peptide by
MALDI or LC/ESI
Search
Determine intact peptide mass
masses of proteins database
by MALDI or ESI with peptide
masses
U
nk
no
n
ow
w
n
Kn
Search
Identify and Known protein sequence
characterise database with Sequence by MSn
protein peptide
sequences
Unknown
n
MS for additional
information about
protein sequence
to design
oligonucleotides
for cloning
2. Strategy for the identification of proteins based on mass spectrometry.
E S S
E
SH
E
Protein
Enzymatic digestion
or chemical cleavage
Peptide mixture
Reduction of
MS disulfide bridges
MS
S S SH + SH
S S
SH
Relative intensity
Relative intensity
m/e
m/e
3. Strategy used to locate the cysteines linked by a disulfide bridge.

Recombinant protein
MS of intact protein Structure verification

ldentification of
Cleavage modification
Purity evaluation
Peptides in mixture
LC + MS
MS of peptides
or Localisation of
in mixture
LC/MS modified peptides
LC + MS/MS
MS/MS of Confirmation of
or
peptides Peptides identity
LC/MS/MS
Localisation of modified
residues
MS combined with Detailed characterisation of

enzymatic or glycosylation and other
chemical treatment modifications
4. General strategy to characterise recombinant proteins.
+ +
ESI
+
+
Folded protein + m/e
+
+ ++ +
+ + + +
ESI +
+
+ + +
+
Unfolded protein + +
m/e
5. Characteristic charge distribution of native or denatured protein.
Antigen
Proteolytic
digestion Antibody
Proteolytic
Antibody digestion
Wash Elute
MS MS MS
Antigen digestion Non-epitope Epitope
m/e m/e m/e

6. Two different methods for the identification by mass spectrometry of an epitope recognised by a given antibody. The
selection by affinity for the antibody in performed either before or after the enzymatic cleavage.
The sequencing and cloning of a protein which inhibits the proliferation of capillary endothelial cells in vitro
has been achieved by electron spray-tandem mass spectrometry. Sequencing and DNA cloning of a peptide with
mle: 699.3 from endothelial cell growth inhibitor carried out by tandem mass spectrometry are compared with the
deduced sequence of amino acids here.
Nucleic acid sequence Amino-acid sequence obtained by

Deduced amino-acid sequence
of cloned gene mass spectrometry
ATA GTT GCA ATT Ileu—Val—Ala— Lxx—Val—Ala—
AAT GTT CCT AAA Ileu — Asn—Val—Pro−Lys Lxx—Asn—Val—Pro−Lys
Here, Lxx is Leu or Ileu.

Electron-spray mass spectra for a synthetic oligonucleotide show major peaks at m/e : 709.1 (M−6H)6−, 608.0,
531.6, 472.4, 425.1, 386.3 (M − 11H)11−. Based on the mass spectral data, the sequence of nucleotides in the
system has been found to be C—A—T—G—C—C—A—T−G−G—C—A—T−G.
The combination of desorption ionisation and MS/MS allows surface analysis of heterogenous materials with both
good sensitivity and selectivity in the real world also. The direct analysis of plant material is possible by SIMS. The
direct detection of the quarternary alkaloid muscarine in fungal tissue is an example. The mushroom tissue is effectively
sputtered into the gas phase by the primary ions of Ar+. The cation of muscarine shows a peak at m/e : 174 as expected.
Positive ion MS/MS spectrum of m/e : 180 generated from a dried aqueous extract of the cactus Trichocereus pacacana
by laser desorption mass spectrometry shows two major fragments which arise as illustrated from the quarternary
alkaloid candicine. In this cactus, the quarternary compound is present at about 4 parts per thousand dry mass.
CH2 CH2 N+(CH3)3

N (CH3)3
+OH
OH
m/e : 180 m/e : 121
H H
C N(CH3)3
C + HN+(CH3)3
H2
HO m/c : 60
SI/LD-MS/MS has been used for the identification and quantitation of modified bases of nucleic acids since SI
or LD mass spectrum of modified bases of nucleic acids contain intact ions corresponding to the nucleoside bases,
e.g. 7-methylguanosine (C11H15N5O5, MM = 297) is characterised by the peaks at m/e : 282, 166 (base peak) and
149. The quantitation is carried out with d3-labelled internal standard and an additional peak due to deuterated
moiety at m/e : 169 also appears in the spectrum.
EI-GCMS has proved to be a very powerful tool for the analysis of the branched fatty acids of human skin surface,
identification of volatile compounds in human urine, identification of drugs in body fluids such as gastric contents,
blood or urine in emergency cases of actuate poisoning, etc. For instance if EI-GCMS of the urine samples of an
overdose victim shows peaks at m/e : 370, 356 and 245, the victim is supposed to have consumed the drug melleril
(B) since the peaks are characteristic of parent drug (B) and its metabolites, (A) and (C) (Fig. 11.36).
H3C H
N N
CH2CH2 CH2CH2
H
N SCH3
N N
SCH3 SCH3
S
S S
(A) MM 245 (B) MM 370 (C) MM 356
Fig. 11.36 Structures of the drug melleril (B) and two of its metabolites (A) and (C) found in the urine
of an overdose victim.
GCMS has also been successfully applied to detect morphine in the bile extract, codeine in the urine extract,
amobarbital in the blood sample, etc., of drug addicts.
11.17 NEGATIVE-ION MASS SPECTROMETRY

Recall from sections 11.1−11.4 that when a molecule is subjected to high-energy electrons, a mixture of positive
and negative ions as well as neutral species are formed. The energy of the bombarding electron is generally much
higher than that of bonds which bind together the atoms in the molecule. Thus when high-energy electrons interact
with the molecules, not only does ionisation occur, but bonds are broken and fragments are formed giving rise to
ions, other than the intact molecular ion. Positive-ion EI spectra are most commonly recorded because many fewer
negative ions are formed by this particular ionisation technique than positive ions. Positive ions are propelled into
the analyser by maintaining the ion source at an electrical potential positive relative to the analyser and by focus-
sing with voltages applied to a lens system located between the source and analyser. The repeller electrodes assist
in focussing the ions into the analyser. Negative ions and electrons are attracted to the positively charged electron
trap. Negative-ion mass spectrometry differs from positive-ion mass spectrometry that in the former, the domi-
nant ionisation processes involved are slow electron attachment and/or dissociative attachment. The slow electrons
needed for such attachment processes result mainly from the primary electron bombardment by the following
mechanism.
[ ( y) [M]]+primary + 2e (
[M] )
The slow electrons required for the generation of stable carbanion can be obtained by the bombardment of
either the sample molecules themselves or by an added gas, such as nitrogen or argon. The masses of nega-
tive ions, so produced, are sorted out as a function of their relative abundance. The relative abundance of
ions is measured relative to the base peak (i.e. peak of highest intensity which is taken as 100 per cent abun-
dance). Negative ions are not only produced by electron bombardment but by a variety of chemical reactions
(observed in the negative chemical ionisation) described in Section 11.3. One of the limitations of negative-
ion EI mass spectrometry is its low sensitivity since carbanions formed from most of the organic compounds
are unstable. However, negative-ion chemical ionisation and desorption ionisation has solved the problem of
low sensitivity to a great extent and these days just like its elder sister, it is promising for molecular structure
determination not only of simple organic molecules but also of biological macromolecules, e.g. the negative-
ion spectra of electron-loving compounds, namely quinones, trifluoroacetates and nitrocompounds can sat-
isfactorily be scanned by CIMS which otherwise is very hard to scan satisfactorily by EIMS because of high
energy of the bombarding electrons. The molecular collisions in the CI process lower down the initial kinetic
energy of the bombarding electrons to such an extent that they are easily captured by the anions of the analyte
molecule to produce the anion radicals. As an example, the CH3O− produced in the CI source behaves as a
Brönsted base and deprotonates the sample molecule to generate an anion which captures an electron to give
the radical anion.
M CH 3O − ⎯⎯
→ (M H) − + CH 3OH
( )− + e ⎯ → ( )− •
The study of negative-ion spectra is especially useful in determining ion structures and energetics. Negative
metastable ions aid in deducing the structure of molecules. Although multi-charged negative ions of molecules
produced in the ionisation process are also well documented but to date no doubly charged negative atomic ions
except O 2− has been reported. The existence of O 2− has been proved by accelerating mass spectrometry (AMS).
Using 180-degree magnetic analyser and a Li vapour jet target, it is observed that no O 2− with a lifetime longer
than 10−9 second accompanying the O− ions are produced. Ca−, Dy−, etc., are well known.
Inorganic compounds which contain elements having high electron affinity, i.e. SO2, PCl3, IClO3, etc., or organic
compounds containing electron-loving groups, namely F, NO2, Cl, etc., are most suited to negative-ion mass spec-
trometry since derivatisation by such groups would enhance substantially the electron-capturing capacity of the
organic compounds. The electron affinities of some molecules, radicals, and atoms are listed in Table 11.20 for
Table 11.20 Electron affinity of some molecules, radicals and atoms.
System Electron affinity (in eV)/kJ mole−1
Molecules: Ion formed

BF3−1 2.65/255
(p-Benzoquinone) −
1.34/129
(Chloranil) −
2.40/231
(Fluoranil) −
2.27/218
NO2 −
3.91/377
O2− 0.45/43
SF6− 1.43/138
(Tetracyanoethylene)− 2.88/278
WF 6
−
2.74/264
UF 6
−
2.91/280
Radicals: Ion formed
CH3− 1.08/104
C2H5 −
0.89/86
C3H5 (Cyclopropyl)
−
0.344/33.125
C6H5 −
2.20/210
CF3 −
1.85/179
CCl3 −
1.22/117
SiF3 −
3.35/323
NH2 −
1.12/108
C6H5NH −
1.55/149
(C6H5)2N −
1.19/114
PH 2
−
1.60/154
OH −
1.83/176
CH3O −
0.38/36
CF3O −
1.35/131
SH −
2.19/211
CH3S −
1.32/127
CN −
3.17/305
SCN −
2.17/209
SeCN −
2.64/255
Atoms: Ion formed
Br− 3.363/324
−
C 1.12/108
−
Cl 3.613/348
F 3.448/333
−
H 0.80/77
−
*H (1s ) 2
0.747/72
I− 3.063/296
O− 1.466/142
S− 2.07/200
*Quantum mechanical calculation: The electron affinities of most simple alkyl

radicals are negative, and consequently the anions are unstable with respect to
their radicals.
ready reference. As in positive-ion mass spectrometry, fragmentation of the negative molecular ion occurs with
the elimination of neutral molecule or of a free radical. Ions other than M−1, i.e. [M⎯H]−, [M⎯2]−, [M⎯R]−
are also encountered in negative-ion mass spectra of molecules and they assist in accomplishing the fragmen-
tation pattern of molecules. Moreover, negative-ion mass spectrometry is complimentary to positive-ion mass
spectrometry.
Recently, a new tool based on negative-ion mass spectrometry has been used for the structure determination
of unknown molecules. Charge stripping of negative ion yields a decomposing positive ion, which produces a
dissociative charge inversion spectrum of the negative ion. This technique allows the study of positive ions not
available by conventional ionisation, it provides a fingerprint for the parent positive ion produced in the collision
process, and it may be used to determine the structure of the precursor negative ion. Finally, a combination of
negative ion spectrum and a charge inversion spectrum may be used as an analytical technique for the structure
determination of unknown molecules.
The collision-induced negative-ion mass spectra of some organic compounds are described here.
(a) Hydrocarbons Simple alkanes being weakly acidic are not readily deprotonated even by the strongest
• • •
base, —NH2. Most of the simple alkyl radicals except C H 3 , C2 H 5 and C3 H 5 (cyclopropyl radical) have negative
electron
•
affi
•
nity so their
•
anions are unstable relative to their respective radicals. But the anions corresponding
to H 3 C, C2 H 5 and C3 H 5 can be produced indirectly by decarboxylation of the respective carboxylate ions. As
an example,
CH 3COO CH 3 CO 2
m/e : 15
Olefins and benzyl derivatives can readily be deprotonated but benzyl anion undergoes collision-induced depro-
tonation. The negative-ion mass spectra of (i) C6H5C(C2H5)2 and its deuterated derivative (ii) C6H4DC(C2H5)2 show
peaks at m/e (i) 147, 131 and (ii) 148, 131. The peak at m/e : 131 in the spectra originates from the fragmentation
of deprotonated diethylbenzyl anion as follows:
C2H5 C2H5
−C C
CH2 CH2
CH3 CH3
H H
m/e : 147/148
D D
C6H4C (C2H5) = CH2 + CH4/CH3D

m/e : 131
The fragmentation pattern and isotope effect reveal that the loss of methane involves the terminal methyl group
and a ring hydrogen; the removal of ring hydrogen is rate determining (kH/kD = 3.75).
Similarly, the fragmentation of deprotonated 4-phenylbutene occurs to give a peak at m/e : 103 with the loss
of ethylene.
−
CH2 CH CH2
−
[(C6H5CH CH2)CH2H3] (C6H4) CH CH2 + CH2H4
m /e : 131 m/e : 103
(b) Aromatic Hydrocarbons Negative-ion mass spectra of benzene, napthalene and some polycyclic aro-
matic hydrocarbons are obtained using low-energy electrons. The spectra originate from electron capture with the
subsequent expulsion of protons from the hydrocarbon or resonance electron capture or both, i.e.
− •
+ e + H
Anthracene + e [Anthracene] −
Higher condensed systems, however give M− ions as base peaks.

Table 11.21 Negative-ion spectral data on some aromatic hydrocarbons.
Compound Formula n Base Peak Other Peaks

(m/e) (m/e)
Benzene C6H6 1 77 (M − 1) 175
Naphthalene C10H8 2 127 (M − 1) 126, 125
Anthracene C14H10 3 178 (M) 177
Phenanthrene C14H10 3 177 (M − 1) 178, 176, 175
Pyrene C16H10 4 202 (M) 201
Tetracene C18H12 4 228 (M) 227
The negative-ion spectral data on some aromatic hydrocarbons have been listed in Table 11.21.
The negative-ion mass spectra of aromatic hydrocarbons also show peak at M + 15. This is attributed to the addition
of oxygen atom (present in the ion source) which takes place according to one of the following mechanisms.
M O + e ⎫⎪
−
⎬ ⎯ →[M O] ⎯⎯
→[( M )O]− + H
−
M O ⎭⎪
M e →[M 1] H ⎯+ O →[( M )O]− + H
The ease of addition of oxygen atom in the polycyclic aromatic series follows the order: Tetracene >Anthracene >
Naphthalene. The peaks at m/e (peak height) : 258 (7 mm), 243 (16.5 mm), 228 (35 mm) in the spectra of tetracene
correspond to (M + 30), (M + 15) and M− respectively. In case of anthracene, the peaks at m/e (peak height): 178 (34
mm) and 193 (7 mm) are attributed to M− and (M + 15) respectively.
(c) Alkyl Halides The negative-ion mass spectra on the reactions of alkyl halides with strong base in the gas phase
reveal that the reactions proceed either via SN2 or elimination mechanism. However in alkyl fluorides, deprotonated
species have also been detected. As an example, the reaction of CD3O− base with ethylfluoride follows the scheme:
37 percent
−
F + CD3OH + C2H4
m/e : 19 Elimination
29 percent reactions
− −
CD3O + CH3CH2F F .... HOCD3 + C2H4
m/e : 54
34 percent
−
CH2CH2F + CD3OH
m/e : 47
The stability of b-fluoroethyl anion is attributed to the negative or anionic hyperconjugation. The anion will
give F− and C2H4 on collision activation.
F−
H H
C C F − + CH2 CH2
H H
(d) Ketones The negative-ion mass spectrum of deprotonated acetone shows peaks at m/e : 57, 56 and 41. The
peak at m/e : 57 corresponds to the acetate enolate ion H 3COCH 2− ). The other two peaks in the spectrum result
from the fragmentation of enolate ion by the following two processes.
• −
CH2COCH2 + H•
− m/e : 56. (i)
(CH3COCH2)
m/e : 57 − −
CH3(CH2 CO) HC CO + CH4
m/e : 41. (ii)
The peaks at m/e: 97, 95 and 69 in the mass spectrum of deprotonated cyclohexanone arise due to the follow-
ing scheme:
−
+ H2
O O
m/e : 95
− −
− H O
C6H11O
Cyclohexanone
enolate ion m/e : 97 + H2
m/e : 95
O O
C
−
HC CH2 + C2H2
−
H2C
m/e : 69
(e) Alkyl Nitrites The ion cyclotron resonance (ICR) spectra of alkyl nitrites (RONO) which contain at least
one hydrogen substituent on each of the a and b carbon atoms, show peaks corresponding to both RO− and
(RO−—H2) ions. As an example, in case of ethylnitrite, the (EtO−—H2) ion at m/e : 43 correspond to the enolate
CH 2 == CHO that is produced by the following three processes [(i)−(iii)]
→ CH 2 = CHO N O + H 2
C2 H 5 ONO + e ⎯⎯ (i)
C2 H 5 O + C2 H 5 ONO ⎯ → CH 2 = CHO C2 H 5 OH HNO (ii)
(C2 H 5 O )* ⎯ → CH 2 = CHO + H 2 (iii)
Excited ethoxide
negative ion
The ICR and theoretical studies suggest that the most plausible mechanism for the elimination of H2 from the
ethoxide ion is
C2H5O− → H−...HCH2CHO → H2...−CH2CHO → CH2CHO−...H2 → CH2CHO− + H2
The negative-ion mass spectra of alkoxide ion produced by dissociative secondary electron capture (the
energy of primary electron beam being 70 eV) from alkyl and deuterated alkyl nitrites using N2 as collision
gas (since it gives slightly higher isotope effects than those obtained from argon and helium), recorded in
Table 11.22 suggest that in cases where losses of H2 and CH4 are competitive, the step (i) is always more pro-
nounced than step (ii).
Table 11.22 Collision-induced losses of H2 and CH4 from RO- ions generated from alkyl nitrites.
Loss(es) m/e Loss (es) m/e

R in RO − R in RO −
Me – – CD3 – –
Et H2 43 CH3CD2 HD 44
CD3CH2 HD 45
CH3CH2CD2 HD 58
Pr H2 57
CH3CD2CH2 HD 58
i-Pr H2, CH4 57, 43 (CH3)2CD HD, CH3D 57, 43
Bu H2 71 (CD3)2CH HD, CD3H 62, 46
i-Bu H2, CH4 71, 57
t-Bu CH4 57
Me(CH2)4 H2 69
Me3CCH2 CH4 55
The enolate ions are produced by the following scheme:
H H
C C CH2 CHO− + H2
H H
m/e : 43
H O−
H D
C C CH2 CDO− + HD EtO−
H D
m/e : 44
H O−
D H
C C CD2 CHO− + HD
D H
m/e : 45
D O−
CH3 CH2 CH2 CH3 CH CHO− + H2

m/e : 57
O− PrO−
CH3 CH2 CD2 CH3 CH CDO− + HD
m/e : 58
O−
O− O−
H3C C H CH3 C + H2
CH3 CH2
m/e : 57
−
O O−
H3C C D CH3 C + HD
CH3 CH2 i-PrO−

m/e : 57
O−
−
H3C C H CH3CO + CH4
m/e : 43
CH3
O−
−
H3C C− D CH3CO + CH3D
m/e : 43
CH3
H CH3 CH3
+ CH4
H
C C
CH3
CH2 C
O− t−BuO−
H O− m/e : 57
m/e : 73
(f) Nitro Compounds The dissociative resonance electron capture negative-ion mass spectra of nitro aromat-
ics are interpreted by the following fragmentation scheme.
•
O
••
Ar N
—2
1—
O−
O O2 O
+ ·
Ar N Ar4 N •Ar + N
••
1
1—4
c− O
− O− O−
3
1—3 •• •−
Ar N O+O
The negative ion mass spectrum of nitrobenzene shows peaks at m/e : 123, 46, 16.
(g) Carboxylic Acids Collision activation negative-ion mass spectra of deprotonated carboxylic acids in the
gas phase using He as a collision gas reveal that the deprotonated acids exist as carboxylate and enolate anions
in significant amounts. The interconversion of enolate anion (I) to carboxylate anion (II) vice versa occurs under
conditions of single collision activation; Both the anions are characterised by their modes of fragmentation. As
examples, the single collision activation mass spectra of enolate- and carboxylate anions produced respectively
from (CH3)3SiCH2COOH and CH3SiOCOCH3 using CH3O−(CH3ONO) as a deprotonation-inducing base and He
as a collision or bath gas show similar peaks but with different per cent abundances at m/e : 59, 58, 44, 41, 17,
15, 14 : These peaks are attributed to the ion fragments formed from the parent anions by the following modes
of fragmentation:
H + CH2 COO–
O m/e : 58
HO C − CH3COO−
[CH3COO]– CH3 + CO2–
O
m/e : 44
(I) (II) m/e : 59
CH3– + CO2
m/e : 15
CH2 + H
HC CO– + H2O
– m/e : 41
[CH2COOH]– [HO(CH2CO)]
m/e : 59
OH– + CH2CO
m/e : 17
The observed spectra of both the anions are similar so far as their m/e values are concerned. However a
minor formation of CH 3− from (I) and this together with the formation of 41 from (II) implies that the car-
boxylic acid enolate anion and the acetate anion are interconvertible under identical conditions of collision
activation. The I-II transformation is possible only provided the barrier for 1, 3H+ - transfer is lower than that
for 1, 2 H+ - transfer. The conversion of a carboxylate anion by 1, 3H+- transfer to the enolate anion prior to
fragmentation is a standard reaction of a alkyl carboxylate anions (M B Stringer et.al, J. Chem. Soc., Perkin
Trans II (1987) 385)
The charge reversal, i.e. positive ion spectra of both the anions, are quite different; m/e values of common peaks
in both the spectra are same but their intensities are quite different. Base peak for I appears at m/e : 42 and at m/e :
44 for II. Major ions corresponding to I are CO +2 • (26 per cent) and HCO2+ (65 per cent) while for II are CH3+ (31 per
cent and CO +2 • (100 per cent). This confirms the initial difference between the structures of the two anions.
The negative ion mass spectrum of deprotonated (CH3)3SiCH2COOH in the gas phase using F− as a deprotonation
inducing base and in flowing after glow reveals that only enolate ion I is formed and it (m/e : 59) reacts readily with
CH3OD to form m/e : 60. Acetate anion does not react with CH3OD under these conditions.
CH2
F + (CH3)3SiCH2COOH HO C + (CH3)3 SiF
O
O
HOC + CH3OD
CH2
[(CH2CO2H)– DOCH3]
[CH2DCO2H (–OCH3)]
CH2DCO–2 + CH3OH
m/e:60
Gas-phase collision-induced mass spectra of deprotonated acetic acid using hydroxide (OH−) as a deprotona-
tion-inducing base show that significant amounts of the carboxylate and enolate anions are produced. However,
the unimolecular I–II transformation does not take place significantly.
O
CH3C O– + H2O
O
HO– + CH3C OH
O–
H2C C OH + H2O
(h) Alcohols and Phenols The collision-induced decompositions of simple alkoxides occur with the 1, 2-loss of
dihydrogen from alkoxide stepwise. For example, fragmentation of ethoxide takes place via hydride ion-aldehyde
complex. Steps A and B are both rate determining (i). The methoxide ion, in contrast, undergoes a low yield 1, 1
elimination of dihydrogen. Here again, steps C and D are both rate determining (ii).
CH 3CH 2 O A
[H (CH 3CHO)] B
→ CH 2 =CHO − + H 2
[H (CH 2 =CHO − )] ⎯⎯
⎯ (i)
CH 3O C
[H (H 2 CO)] D
[H (H C=O)] → HC=O + H 2 (ii)
The collision-induced negative-ion spectra of alkoxide PhCH(CH3) O − can be interpreted on the basis of decompo-
Ph
sition of Wittig product ion CH O–. The Wittig product ion undergoes competitive losses of methane, benzene
CH3
or aldehyde according to the following scheme.
– –
[Me (Ph CHO)] (C6H4) CHO + CH4
m/e : 105
(CH2CHO)– + C6H6
Ph
– m/e : 43
CH O– [Ph (Me CHO)]
Me
m/e : 121 (C6H5)– + CH3CHO
m/e : 77
O O
– –
H (PhC Me) (C6H4) CMe + H2
m/e : 119
−
The collisional-induced dissociations of ion PhCH2O − (and its deuterated analogues PhCHDO − , PhCD2 O ,
− •
C6D5CH2O ) in the gas phase occur with the competitive losses of H ; H2, CH2O and C6H6. The loss of hydrogen,
−
i.e. H2, takes place mainly to form (C6H4) CHO. The abi nitio calculations show that hydrogen loss occurs by the
following mechanism:
PhCH 2O − → [H
[ − ( PhCHO)] (C6 H 4 ) CHO + H 2
The losses of CH2O and C6H6 are accompanied or preceded by partial phenyl H-benzyl H interchange. The
negative-ion mass spectrum of PhCH2O− is rationalised by the scheme given below:
C6H5 + CH2O
[C6H5 (CH2O)] m/e : 77
HC = O + C6H6
m/e : 29
(C6H4) CH2OH C6H5CHOH

C6H5 CH2O m/e : 107
m/e : 107 •
PhCHO]
• + H
CH2O −
• •
(C6H4) + H
− −
[H2 (C6H4) CHO] (C6H4) CHO + H2
m/e : 105
[H (PhCHO)]
− −
[H2 (PhC=O)] PhC=O + H2
m/e : 105
The peaks at m/e : 121, 119, 103 and 91 in the collision activation negative-ion mass spectrum of PhCH2CH2O −
(and its deuterated analogues PhCH2CD2O − and PhCD2CH2 O − ) are attributed to the losses of H2, H2O, CH2O (losses
for PhCD2CH2O − are HD, HOD and CH2O while for PhCH2CD2O − are HD, H2O HOD and CD2O) have been rationa-
lised by the following mechanism of fragmentation.
− −
[Ph CH2(CH2O)] Ph CH2 + CH2O
− m/e : 91
Ph CH2CH2O
m/e : 121 −
− −
[H (Ph CH2 CHO)] [H2(Ph CH CHO)] PhCHCHO + H2
m/e : 119
−
(C6H4) CH = CH2 + H2O
− − − m/e : 103
Ph CH2CH2O Ph CH CH2OH [HO(Ph CH = CH2)]
Ph CH = CH− + H2O
m/e : 103
The collision-induced negative-ion mass spectra of ions C6H5(CH2)nO− where, n = 3 − 5 being complex, is
beyond the scope of this book. The ion Ph(CH2)3O− undergoes many fragmentations including losses of H2O and
CH2O and loss of H2. The loss of H2 occurs by both 1,2- and 1,3-eliminations. A number of minor fragmentations
occurs after partial interchange of phenyl hydrogens and hydrogen at position 2. The first example of specific
double-protons transfer is noted, viz. Ph(CH )3 O → C6 H 7 CH 2 CHCHO. Ions Ph(CH2)nO− (n = 2 – 5) all
−
decompose to produce PhCH 2 ions; when n = 3 − 5 the reactions may involve Smiles intermediates, i.e.
(CH2) n
− − −
Ph(CH2)n O Ph CH2 + O (CH2) n−1
O
For n = 3; peaks result in at m/e : 77, 91 and 105, i.e

77 91 105
Ph CH 2 CH 2 CH 2 O −
The other peaks are observed at m/e: 135, 133, 117 (–H2O), 115, 103, 79 (–C3H4O), 57 (–C6H6), 43 (–C7H8),
29 (HC = O) . The peaks at m/e: 79 ( C6 H 7− ) is formed by a specific double H-transfer, the first such reaction to be
reported for negative ions.
−
HC =O + C8H10
− −
Ph(CH2)3 O [C8H9 (CH2O)]
−
C8H9 + CH2O
Two possible structures for C8 H 9− ion are the following:

(i) Ion PhCH2CH2− Formed by the direct loss of CH2O. This seems like a lesser possibility, but may be stabilised
by negative hyperconjugation.
(ii) Spiro ion: Mass spectra of C8 H 9− is still not established,

Simple phenols are little studied. The phenoxide ion itself loses CO and CHO to give the fragment ions as
follows:
O
−
C5H5 + CO
m/e : 65
O
m/e : 93
H
o − •
C5H4 + CHO
m/e : 64
m/e : 93
(i) Ethers Deprotonated ethers undergo Wittig rearrangement involving either of the ion-neutral or radical–
radical intermediates shown in Scheme 1 followed by collision-induced products from the intermediates.
[R1 (RCHO)]
ion-neutral
R CH O R1 R
CH O
• • R1
[R1 (R CH O)]
radial-radial Wittig
production
Scheme 1 R = C6H5 or CH2 = CH—, R1 = allyl, benzyl, alkyl, or aryl.
In the condensed phase, aldehydes are often by-products of the reaction and the migratory tendency of sub-
stituents R1 is allyl benzyl > methyl > ethyl > phenyl. Since the migratory tendency is in the order of free radical
stabilities, it is suggested that the radical pair mechanism is more likely. The collision-induced negative-ion mass
spectra of ions PhCHOR and Ph(R)CH O − (R = alkyl or phenyl) are similar suggesting thereby that the Wittig
rearrangement also occurs in the gas phase. Major fragmentations are best interpreted in terms of the Wittig prod-
uct ion, i.e. R(R1)CH—O−. When R1 is an alkyl group > Me, the radical-radical intermediate is more pronounced,
since the electron affinity of R1 ≤ 0. In view of it let us now see how Wittig rearrangement aids in interpreting the
mass spectra of deprotonated ethers in the gas phase.
The reaction between simple dialkyl ethers R2O (R ≥ C2H5) and strong bases in the gas phase undergoes elimi-
nation reaction to give alkoxy anion rather than deprotonation process. For example,
HO− + C2H5OC2H5 → C2H5O− + C2H4 + H2O
On the other hand, the unsaturated ethers undergo deprotonation at either the allylic or benzylic positions to
form carbanions which undergo a number of collision-induced dissociations. For example, in allyl alkylethers,
the reaction proceeds as follows:
CH 2 = CH CH (CH ) OC2 H 5 ( n = − 7) Proton transfer
CH 2 C 2 ) n +1 OCH 2 CH 2− ⎯ →
CH(CH
Intermediate
CH 2 = CH(CH ) n O C2 H 4
For n = 1, the reaction is most pronounced when the proton transfer takes place via six-centre state.
H H2 H2
H CH CH2
C C C
H2C C O H2C C O
CH2
H H2
CH2
C
H2
CH2 = CH CH2 CH2 O− + C2H4
The collision-induced mass spectrum of a Wittig product ion Ph(CH3)CHO− (i.e. m/e : 121, 120, 106, 105, 77, 43)
suggests that the fragmentation of the ion occurs via ion-molecule intermediates by competitive elimination reac-
tions and via radical cleavage according to the following scheme.
[CH3−(PhCHO)] (C6H4)−CHO + CH4

ion-molecule m/e : 105
Ph intermediates
CH O
CH2CHO + C6H6
CH3
m/e : 43
[Ph(CH3CHO)]
(Ph)− + CH3CHO
m/e : 77
When R1 > CH3, the collisional activation spectrum always shows a pronounced loss of R1:
Radical Cleavage
PhCHOR1(R1 > CH3) •
[R1•(Ph – C H – O]
R1• + Ph – C•H – O
m/e : 106
The ion at m/e : 120 corresponds to (M—H)–.

The deprotonated diallylethers undergo collision induced reactions in the gas phase by 1, 2- and 1, 4-Wittig
rearrangements followed by oxy-Cope rearrangement to give the product.
–
O
Witting Oxy-Cope CHO
O –
Rearrangment
Rearrangment
Deprotonated benzyl ethers undergo collision-induced Wittig rearrangement as well as radical scissoring to
give respectively Wittig product ion and radical cleavage product ion
[R1(PhCHO)]
(ion-neutral Ph
PhCHOR1 Wittig intermediate)
CH O
rearrangment
Radical • • R1
cleavage [R1(PhCHO−)] Wittig
(radical-radical product ion
intermediate
• • Wittig ions)
(PhCHO) + R1
cleavage
product
ion
As has already been stated, the similarity between the mass spectra of ions PhCHOR
C and Ph(R)CHO is an
evidence for the occurrence of Wittig rearrangement in the gas phase.
Deprotonation of phenyl (diphenylmethyl) ether (PhOCHPh2) results in the ion PhOCPh C 2 (I) while that
of Ph3COH gives rise to the ion Ph3C O (II). The collision-induced negative-ion mass spectra of both the
anions are very similar except that there is a very low abundance peak corresponding to Ph O in the spec-
trum of PhOCPhC 2. This suggests that the complete fragmentation occurs through Wittig ion according to the
following scheme.
Ph2(C6H4) COH (C6H4) Ph + C7H6O

m/e : 153
Ph + Ph2CO
Ph3CO
m/e : 77
m/e : 259
O
[Ph (Ph2CO)] (C6H4) C Ph + C6H6

m/e : 181
O Ph
+ H2
m/e : 257 Ph C O + Ph – Ph
m/e : 105
The similarity between the spectra of anions (I) and (II) suggests the collision-induced transformation I → II.
The collision-induced mass spectrum of deprotonated allyl phenyl ether ( PhOCHCC H CH 2 )m/e (peak height
from common base line in mm) : 133(60), 131(59), 115(50), 106(13), 105(15), 103(15), 93(5), 77(43), 55(22),
27(2.7) is similar to that of Ph(CH2 =CH)CH O to a great extent. The major fragments in the spectra are formed
through the Wittig product ion and most fragments follow normal trend. The ions corresponding to the peaks at
m/e: 106 and 93 do not appear in the spectrum of the ion Ph(CH2==CH)CH O .
[CH2 = CH−(PhCHO)] CH2 = CH− + PhCHO

m/e : 27
Ph
CH O Ph + CH2 = CH — CHO
m/e : 77
CH2 = CH
m/e : 133 [Ph (CH2 = CH – CHO)]
H
CH2 = C = C + C6H6
m/e : 55 O
C
+ H2
O
m/e :131
H (Ph – C – CH = CH2)
Ph
C = C = CH2 + H2
O
m/e :131
The ions at m/e : 93 and 106 are formed through radical cleavage and Claisen rearrangement respectively.
+
• •
PhO CH CH = CH2 PhO + CH CH = CH2
m/e : 93
•
O O O O
− O + •C2H3
H
m/e : 106
The peak at m/e: 115 corresponds to (M—H2O) and results in by the following scheme.
−
O
Ph CH
H
C
CH2
−
OH (C6H4) CH = C = CH2 + H2O
− m/e : 115
Ph CH
−
[HO (Ph CH = C = CH2)]
C −
CH2 Ph — C = C = CH2 + H2O
m/e : 115
The deprotonated spectra of C6 D5 OCHCC H CH 2 , i.e. allyl phenyl (d5) ether fits in the above routes of
fragmentation. Moreover, the charge reversal (positive ion) mass spectra of PhOCHCHC = CH 2 and Ph (CH2 =
CH) CH O are different but each species has a contribution from the other. This suggests the initial difference in
structure of two anions and collision induced transformations of one form to the other form and vice versa.
(j) Amines The collision-activated fragmentation pattern of alkyl N− ions is similar to that of C− and O− species
to a great extent. The fragmentation pattern of deprotonated isopropyl amine follows the scheme:
−
(CH3)2 C = N + H2
− − m/e : 56
(CH3)2 CHNH → [H[(CH3)2C = NH)]
m/e : 58
−
CH3(CH2) C = NH + H2
m/e : 56
−
CH3CH = N + CH4
m/e : 42
[CH3−(CH3 CH=NH]
−
CH2CH = NH + CH4
m/e : 42
CH3− + CH3 CH=NH

m/e : 15
To summarise, in negative-ion mass spectrometry some generalisations regarding the different types of stable
carbanions formed in various classes of compounds are listed here in Table 11.23 .
Table 11.23 Stable carbanions in various types of organic compounds.
Stable carbanions Molecular systems

Ketones, aliphatic esters, aldehydes, alcohols, carboxylic acids, mercaptans
[M—H]−
and steroids.
Condensed aromatics (n ≥ 4; n is the number of condensed rings) or same ring
M−
systems in which C—H group is replaced by nitrogen atom
[M − 2]− Condensed aromatics (n < 4), compounds such as phenanthradine.
[M—R] −
Compounds in which radical R is bound to a heteroatom.
CN− Nitriles, lower molecular mass-nitrogen heterocyclics, amines and amides.
NO2 −
Nitro compounds
Cl , Cl2
− −
Chlorinated compounds
S , SH
− −
Low molecular mass sulphur compounds
Stable carbanions are not readily formed Lower molecular mass hydrocarbons, ethers and thiophenes. Not suitable for
(wide range of weak ion peaks) negative-ion mass spectra
(k) Negative-ion Mass Spectra of Biological Molecules The high-energy electrons cause fragmentation
of bioorganic polar molecules and their derivatives due to which it is difficult to observe their negative-ion
mass spectra in vivo and vitro (especially in intact form). However, this problem has been solved to a great
extent by the advent of new ionisation techniques, i.e. chemical ionisation and field ionisation. The derivatisa-
tion of molecules is carried out by electron-loving groups in order to enhance their electron affinities. Con-
sequently, the stable carbanions of bioorganic compounds can be produced using low-energy electrons. As
already stated in Section 11.4, the minimum energy required for the production of a stable negative ion can be
determined from the maximum in the ionisation efficiency curve of the molecular system under study. As an
example, let us discuss the EI negative ion mass spectra of p-nitrobenzyl-N-acetyl amino acid derivatives. The
negative ions are produced by the dissocaitive resonance capture of 2.9 eV electrons as follows:
O O
H3C C NH CH C O CH2 NO2 + −

e (2.9 eV)
O O
•−
H3C C NH CH C O CH2 NO2
(Intermediate anion radial)
O O − O O •
H3C C NH CH C O CH3 C NH CH C O
R R
m/e : 115 + R
+
(N-acetlycarboxylate anion)
• −
CH2 NO2 CH2 NO2
m/e : 136
Accordingly, the gas-phase mass spectra of (N-acetyl derivatives of) alanine, valine and proline amino acids
exihibit peaks at m/e values indicated below:
Amino acid m/e (Relative abundance as percent)
Alanine (R==CH3) 130(100), 136(11)
Valine (=CH (CH3)2) 158(87), 136(7)

Proline 156(32), – (not observed)
Just like positive ion–soft ionisation–MS/MS, negative ion–soft ionisation–MS/MS proved to be useful for
the identification and sequencing of polypeptides and their derivatives, polysaccharides and their derivatives,
etc. The nature of the precursor ion depends upon the type of derivatisation. Thus the molecular mass related
ions such as (M⎯H)−, (M′ + Na)−, (M′ − 2Na)− −, etc.(where M and M′ have their usual meanings) act as pre-
cursor ions. The fragmentation patterns of trisaccharides based on negative ion ESI-CID-MS/MS are given in
schemes 1 and 2.
241
−SO3 + H+
−HSO4−
223 160 197
CH2OH
O
HO
HOH2C 2−
([M⬘ − 2 Na]2−)
OSO3Na O
O
OC3H7
NaO3SO (m /e : 364.5)
O
CH3 O NHAc
HO
HO
OH
291.5
(i) Scheme 1 Fragmentation patterns of negative ion ESI-CID-MS/MS of disulphated Lex-type trisaccharide
propylglycoside 2 (M′ 775, as the disodium salt) having (M′−2Na)2− at m/e : (M′ −2Na)2− as the precursor ion.
241 315 590
HOH2C
O
−C3H7OH
HO
HOH2C −
OSO3Na O
O
OC3H7 (−Na)
OH
(m/e: 650)
O
CH3 O
NHAc
HO HO
OH
504 444
−C3H7OH
(ii) Scheme 2 Fragmentation patterns of the negative ion ESI-CID-MS/MS of sulphated Lex-trisaccharides
having (M′—Na)− as the precursor ion.
PROBLEMS
1. The mass spectrum of a naturally occurring silicon (Si) 8. Using the data in Problem 7, determine the cyclo-
sample yields the following data: tron frequencies corresponding to the radicals if the
strength of the applied magnetic field is 15000 G.
Isotopic mass number Peak height [Ans. 3.52, 3.26, 3.05, 1.83, 1.76, 1.69, and 1.63 MHz]
28 92 9. The amino acid analysis of a peptide shows the presence
of six different amino acids, namely leucine, glutamic acid,
29 5 glutamine, valine, proline and tyrosine. The hexapeptide
30 3 has been first acetylated (with acetic anhydride/methanol)
and then per methylated (upon treatment with dimethyl
Calculate the average atomic mass of Si. sulfoxide anion ( 2 3 ) followed by methyl iodide
[Ans: 28.11] in order to obtain per methyl-N-acetyl peptide. The mass
2. From the mass spectrum of naturally occurring neon, the spectrum of the volatile N-acyl peptide methyl ester exhib-
following data are obtained. its characteristic peaks at m/e: 929 (M+•), 707, 610, 497,
327, 192, 170, 142 and 121. Accomplish the sequence of
Isotopic mass number Peak height amino acids in the hexapeptide chain .
[Ans: Leu-Glu-Gln-Val-Pro-Tyr]
20.0 91
+
22.0 9 • Hint : [ Tyrosine peaks:] CH3O CH2 : H3CO CH CH COOCH3
(m/e : 121) (m/e : 192)

Calculate the average isotopic atomic mass of neon.
[Ans: 20.18] The peaks at m/e : 142 results via loss of CO from the
3. The mass spectrum of nitrogen produced from the air sequence ion at m/e : 170]
is characterised by the presence of the peaks of the 10. The values of appearance potentials of C2H5+ ion in
isotopic forms of nitrogen I4N14N and 14N15N, which ethane and propane are respectively 15.2 and 14.5 eV.
are equal to 68 and 0.5 mm, respectively. Determine What will be the values of ionisation potentials of the
•
the percentage content of nitrogen 15N. C2H5 radical in ethane and propane if the values of
[Ans: 0.37 per cent] dissociation energies of the C2H5—H and C2H5—CH3
4. Pick the even and odd electron ions corresponding to bonds are respectively 98 and 84 kcal mole−1?
the following molecular formulas: CH2N2, C2H4O, C10H23N, [Ans: 10.95 eV, 10.86 eV]
C4H4N2, C6H6BrN, C3H7Cl3Si, CH3NO2, CH3CN. 11. The sensitivity coefficients (in divisions/micron) corre-
5. The mass spectrum of nitrobenzene shows the presence sponding to the analysis peaks of n-, (m/e: 56), iso-,
of particles with the following mass numbers: 30, 46, 61.5, (m/e : 74), sec-, (m/e : 45) and tertbutyl (m/e : 59) alco-
77 and 123. Which of the following sets of the products of hols are 11.51, 12.05, 26.98 and 20.93 respectively.
decomposition of nitrobenzene does this mass spectrum Determine the relative contributions of the alcohols
correspond to? (in mole per cent) in an unknown mixture of isomeric
(a) C6H5NO2+, NO2+, NO2+, C6H5+, C6H52+. butyl alcohols from its mass spectral data given
(b) NO+, NO2+, C6H5+, C6H4NO+, C6H5NO22+ below:
(c) NO22+, NO+, C6H5NO2+, C6H52+, C6H5NO+ m/e: n-butyl Percentage percentage
(d) C6H5NO2+, C6H5NO22+, NO+, NO2+,C6H5+ alcohol component sec-butyl tert-butyl of unknown
(e) C6H4+, NO2+, NO2+, C6H52+, C6H5NO22+ iso-butyl alcohol alcohol mixture
alcohol
1
[Ans: (d),[C6H5NO22 + ] = 61.5 +] 56 90.58 2.46 1.02 1.47 126.7
2
74 0.79 9.06 0.29 — 148.0
6. Calculate the flight times of singly charged (a) N2 molecule,
(b) O2 molecule, (c) CO2 molecule, and (d) Ar+ ion for drift 45 6.59 5.03 100.00 0.59 322.6
length of 1 m and acceleration voltage of 2800 V. 59 0.26 4.98 17.78 100.00 301.5
[Ans: (a) 7.07 ms; (b) 7.56 ms; (c) 8.86 ms; (d) 8.45 ms]
7. In the radio-frequency mass spectrum of olefin C2H4,
there are observed lines corresponding to the ionised [Ans: n-24.4, iso-25.8, sec-24.8 and tert- 25.0mole percent
molecules and ions indicated below (relative abundance, 12. Predict the molecular formulas of the compounds
in per cent): containing C, H, N or O whose exact molecular
masses are 60.0210, 45.0214, 102.0678, 190.9540
Radical … CH CH2 CH3 C2H C2H2 C2H3 C2H4 (Cl), 334.0873(S), 180.0630, and 84.0936.
[Ans: C2H4O2, CH3NO, C5H10O2, C6H3Cl2NO2,
Relative
C17H18O5S, C6H12O6 and C6H12]
abundance … 0.7 1.8 0.3 7.3 50 54.5 100
in per cent 13. A peptide has been found to contain five types of amino
acids, viz., Gly (1 molecule), Ala (2 molecules), Val
Determine the radio frequencies corresponding to the
(2 molecules), Leu (2 molecules) and Pro (2 molecules).
radicals if the accelerating voltage is 4000 volts and
The mass spectrum of N-acylated (i.e. C17H35CO) methyl
the spacing between the adjacent grids is 1 cm.
ester of the nona peptide exhibits major peaks at m/e :
[Ans: 9.04, 8.71, 8.41, 6.52, 6.39, 6.27, and 6.16 MHz] 1115 (M+•), 1084, 985, 888, 817, 720, 621, 508, 451, 338
and 267. Determine the sequence of amino acids in the

nona peptide. m/e M M+2 M+4 M+6 M+8
• Hint: Stearoyl = C17H35CO (m/e: 267), OCH3(m/e : 31). Per cent

27 45 21 3 ….
Use the molecular mass data from Table 11. l8 abundance
[Ans: Ala-Leu-Gly-Leu-Val-Pro-Ala-Pro-Val]
[Ans: 3; Cl = 2, Br = 1]
14. The mass spectra of o or m or p nitro-analines show
prominent peaks at m/e : 138 (M+•), 108, 92, 80 and 65. 20. (a) The mass spectrum of an unknown compound has
What ions might these peaks correspond to? a molecular ion peak with relative intensity 43.27 per
cent and a M + 1 peak with a relative intensity of 3.81
+ +
per cent. What might be the number of carbon atoms in
• Hint : C6H4 N H2 → + + HCN Ans : NO2+• , C6H4N H2 the molecule? (b) Determine the molecular formulas of
the isobaric compounds containing C, H, O or N unless
NH2 otherwise indicated corresponding to the exact molecu-
O+
lar masses (amu) given below: 122.1096, 122.0845,
122.0732, 122.0368, 122.0579, and 122.0225(S).
+
C5H5+, NH2, + NH 2, [Ans: (a) 8 (b) C9H14, C7H10N2, C8H10O,
C7H6O2, C4H10O4, C4H10S2.]
15. The mass spectrum of benzaldehyde shows peaks at 21. Which of the following set of peaks correspond to
m/e : 105, 78, 77 and 51. What ions might these peaks l-methoxybutane (a), 2-methoxy butane (b) and 2-meth-
correspond to? oxy-2-methyl propane (c)
+ (A) (B) (C)

C O
Ans: , C6H+6, C6H+5, C4H+3 Peak
m/e m/e peak m/e peak
hight
16. Deduce the structural formulas of the compounds cor- hight hight
responding to the mass spectral data indicated below: (in mm) (in mm) (in mm)
(a) Group A B C D 88 4 88 2.5 88 3

73 45 73 6 56 7
123, 121, 86, 84,
m/e 49, 47 37, 35 57 14 59 45 45 45
119,117 82
Peak • Hint: (C) CH3— CH == OH
1, 20, 45,
height 1, 9, 14 6, 19 6, 24 m / :45
48 m/e : M+•—CH3OH =56
(in mm)
[Ans: (A) → (c); (B) → (a), (C) → (b)]
85, 57,
(b) m/e: 166, 164 137, 135 109, 107 22. The mass spectra of 2-pentanone and 2-hexanol
43
show prominent peaks at m/e : 43, 71, 58 and m/e :
Peak 45, 87, 84 respectively. What ions might these peaks
high (in 1, 1 21, 21 3, 3 22, 6, 35 correspond to?
mm) • Hint: m/e : (102 – 18) = 84; Elimination by OH group
CH2 CH2
of alcohol and a g-hydrogen.
• Hint: (b)
CH2 CH2 + +
Ans: CH3C O C3H7C O
•• ••
+
Br CH2 C CH3 and CH3CH OH
(+) ••
m/e : 135, 137 +
H O+ H2C CH2 •
[Ans: (a) CCl4 and (b) 1-bromohexane) +
C4H9CH = O H
17. Determine the molecular formulas of the isobaric com- •• (CH2)4
• +
pounds corresponding to the exact molecular masses H2C (CH2)4 CH2
indicated below: 31.9898, 32.0375 and 32.0262
Hint: In the first case, mass defect is negative.
[Ans: O2, N2H4, CH4O] 23. Given below are the characteristic peaks of the mass
spectra of some aromatic compounds.
18. Determine the molecular formulas of the isobaric
compounds containing C, H, O or N unless otherwise (a) (b) (c) (d)
indicated corresponding to the following exact molec- NC NH2 Cl OCH3
ular masses: 27.9949, 28.0313 and 28.0061.
[Ans: CO, C2H4, N2] OCH3
19. Deduce the number and nature of halogen atoms pres-

ent in a molecule from the mass spectral data indicated
below: C (CH3)2 Br NH2
93 → 65 + 28 (≡CO) : 45.4
m/e m/e m/e m/e M+•—CH3 = 121; Strong molecular ion peak-aromatic
component.
138
(C6H5O + CH3) = CO;
159 171/173(1 : 1) 127/129(3 : 1) 123 O 93, C8H8O2
IR peak at about <
144 92 100/102 (3 : 1) 95 1700 indicates C=O
group. m/e
116 — 92 77 O
Assign these peaks to the ions in the respective mole- O C CH3

cules.
[Ans: (a) (M+•—CH3); (144+ – C2H4); (b) (M+•—Br)
O
+•
(c) (M+•—HCN); (M+•—Cl); (d) (M+•—CH3), m/e : +C CH3 = 43 ; O+ CO + + C5H5
+
C3H3 + C2H2; M –(CH2 C O) [C6H6O]+ •
m/e : 65 m/e : 39
+
(123+–CO), (95+–H2O) = 77 ]
26. The mass spectrum of sulphur vapour shows peaks
at m/e (peak height):32(25 mm), 64(Base peak, 47
mm), 96(12 mm), 128(21 mm), 160(8 mm), 192(6 mm),
24. (a) Deduce the structural formula of a compound with 224(3 mm), 256(18 mm). What ions might these peaks
molecular formula C6H13O2Br from the mass spectra correspond to?
data indicated below: [Ans: S+, S2+, S3+, ... S8+]
• Hint: On bombardment with electrons, a puckered
m/e 196/198 167/169 151/153 123/125 103 ring of 8 sulphur atoms in crystalline sulphur breaks
Peak height up firstly into chains of atoms; these then succes-
1/1.3 1.8/2 19/18 28/27 51 sively break to give the various S to S7 units shown by
(in mm)
the mass spectrum, S2+ is the most stable form.
Metastable peaks: 100.2, 102.2, 54.7.
(b) The mass spectrum of CO2 shows peaks at m/e :
44, 28, 32, 22, 16, 12. Minor peaks are observed S
S S •
at m/e : 13 and 45. What ions might these peaks Δ S S S S S S S S
•
S S
correspond to?
S S S8 biradical
⎧153 → 125 + 28(≡ C2H4 ) : 102.1 S
• Hint: ( ) ⎨
⎩151 → 123 + 28(≡ C2H4 ) : 100.2 •
Then S8 + S8 • S16 etc.
The changes in the viscosity of molten sulphur as
196 → 103 + 93 ( = CH2Br): 54.1, M—C2H5 = 167, temperature rises have been found to be due to the
M—OC2H5 = 151, presence of biradical chains which can be detected
by ESR technique. The longer the sulphur chains, the
H CH2 more viscous the molten sulphur.
M — CH2Br = 103, 151 – C2H4 = 123 — C — O — CH2 . 27. The molecular mass of a certain gas, as determined on
a high-resolution mass spectrometer is 27.9949. Identify
— C — OH + C2H4
the gas from one of the following compounds which all
have molecular masses of about 28:
OC2H5
Thus there are two OC2H5 groups. C6H13O2Br C + CH2 Br = H N2, C2H4, H2CN, CO
OC2H5 [Ans. C16O]
28. The mass spectrum of a compound exhibits significant
[Ans: (a) BrCH2CH(OC2H5)2 (b) CO2+, CO+, peaks at m/e : 60 (M+•), 42, 31. What might be the struc-
O2+, 1 (CO22+), O+, +C, 13C+, l3CO2+ ] ture of the compound?
2 •Hint: M+•—H2O; M+•—C2H5, [Ans: CH3—CH2—CH2OH]
25. Deduce the structural formula of a compound with 29. The molecular mass of a compound, as determined on a
molecular formula C8H8O2 from the mass spectral data high-resolution mass spectrometer is 32.0375. Identify the
given below: compound from one of the following compounds which all
have molecular masses about 32.
m/e 136(M+•) 121 94 93 65 43 39 O2, N2H4, CH4O
50.8 [Ans: N2H4]
Peak height
35 (Base 15 5 28 17 12.2 30. Electron spray mass spectrum of arginine in
(in mm)
peak) methanol-water (50 : 50) solution containing tetra-
methyl ammonium iodide shows peaks at m/e :
Metastable peaks: 71.6; 45.5. IR shows peak at about < 349(1 nm), 207(4 nm), 175(32 nm), 74(37 nm). What
1700 cm–1. ions might these lines correspond to?
[Ans: C6H5O—COCH3] [Ans: M2H+ ,MCH3OH+2 ,MH
MH+ ,(CH )4 N;+ M = 174 ]
• Hint: 121 → 93 + 28 (≡CO) : 71.6
31. The significant peaks in the mass spectrum of a compound H

with structural formula N +
= CH2 +• +• N CH2 — C H — CH3
+ (M — C(CH3)2;(M — CH3)
C2H5 C (CH2)4 CH C2H5 N OCH3 N OCH3
O O OCH3 [Ans: (a) 1, 3, C9H14N2O]

H2C CH2 N CH2 — CH — (CH3)2
are indicated below: (b) 2-isobutyl-3-methoxy-pyrazine; N OCH3

Problem 33 Deduce the molecular as well as the
m/e 230(M+•) 201 169 139 101 99 73 57 45 structural formula of the compounds A, B and C from
their mass spectral data, i.e. m/e, and exact molecular
Peak masses indicated below:
height — 9 2 2 40 3.5 8 7 2
(in nm) A(298.2862) B(116.0834) C(116.0834)
What ions might these lines correspond to? 298(M+•) 116(M+•) 116(M+•)
Hint: The activation energy for formation of highly sta-
267 101 101
bilised oxoniun ion m/e: 101 is so small that forma-
tion of m/e : 73 and other primary processes compete 239 87 87
very poorly.
237 73 56
M+•—OCH3 (m/e : 199+) 225 57 43
199+—2H2CO → 139+ 223 43 29
211 — —
[Ans: C2H5 — C
•O O +
•
209 — —
197 — —
(m/e : 101);
195 — —
C2H5 — C
59 — —
+O O
CH2
31 — —
+
OCH3 — CH — C2H5 (m/e : 73) ; 101+ O C2H5C O+ (m/e : 57);
CH2
OCH3 O
M+•— C2H5 (m/e : 201+); 201+—CH3OH (m/e : 169+);
+ [Ans: A: C17H35COOCH3; B: CH3COCH — C2H5; C: C2H5CHOC — CH3]
C2H5—CH=OH (m/e : 59+)
CH3
↓ :CH2
34. Dibenzyl ether undergoes a Wittig rearrangement in
+
the condensed phase, but in the gas phase it forms no
CH3—CH=OH (m/e : 45)
stable deprotonated form and no Wittig rearrangement is
32. Chromatography of the steam volatile bell pepper oil noted. It reacts with HO– to give only PhCH2– most prob-
separated a peak which possessed a potent odour, ably by dissociation of unstable PhCH2OCHPh. The gas
characteristic of bell peppers. The exact molecular phase negative ion mass spectral data of the Wittig prod-
mass of the heterocyclic potent odorant, one drop uct Ph(PhCH2) CHO– (A) and its deuterated analogue
(0.05 ml) of which is sufficient to impart bell pepper Ph(PhCD2)CHO– (B) are indicated below:
flavour to a large swimming pool is 166.1102, (a) Cal-
culate the number of rings, double bonds and the A … m/e : 197, 195, 129, 105, 91, 77
molecular formula of the compound. (b) Deduce the
structural formula of the odorant in question from the B … m/e : 199, 196, 131, 105, 93, 77
mass spectral data indicated below,
What might be ions corresponding to the mass spectral
data in question?
124(base
m/e: 166(M+ •) 151 135 −
peak) Ans: Ph CHO−, Ph−, PhCH CHO−, (C6H4) CHO,
109 95 94 93 81 PhCH2 O
Peak height 2 12.5 0.8 – 81
PhCH2−, Ph C C− H Ph.
(in mm) 1 8 16 9 7
35. Mass spectrometric analysis of the vapours in ther-

The substituents are ortho to one another in the molecule. modynamic equilibrium with powdered molybdenum
• Hint: (M+ •—OCH3), trioxide (MoO3) at 850 K as sampled from Kundsen
effusion cell has shown that the vapour phase consists mole–1.
predominantly of Mo3O9; Mo4O12 and Mo5O15 molecules. • Hint: ΔHf(P+) = Appearance potential (P+) + ΔHf(PH3) –
Determine the heats of sublimation of tri-, tetra- and ΔHf(H2) – ΔHf(H).
pentamers of MoO3 from the data given below: [Ans: (a) PH3 → P+ + H + H2, → PH+ + H2,
→ +
2 + H, → PH+3 .
Molecular Temperature Vapour pressure
(b) P+, PH+, PD+ or PH+2 , PH3+, PDH2+, PD2H+, PD3+ and
species (in K) (in atomsphere)
ΔHf+ (kcal mole–1) : P+ = 346, PH+ = 308, PH+2 = 254
and PH+3 = 237.
Mo3O9 973 1.26 ×10–4
37. The equilibrium partial pressures of the gaseous spe-
914 1.19 ×10–5 cies (viz., Cr, CrO, CrO2, CrO3*, O2 and O) formed in
the dissociative vapourisation of Cr2O3(s) have been
967 9.0 ×10–5 measured mass spectrometrically under neutral and
861 1.64 ×10–6 oxidising conditions. The partial pressures of Cr in
equilibrium with Cr2O3 under both the conditions have
Mo4O12 973 6.71 ×10–5 been indicated below:
913 4.94 ×10–6 Oxodising condition: (CrO3* is formed under this condition
only)
905 4.31 ×10–6
Mo5O15 943 1.74 ×10–6 T(K) 1890 1911 1942

916 2.38 ×10–7
PCr × 108 (atom) 3.24 5.91 11.1
908 1.77 ×10 –7
Neutral condition (CrO3* is not formed under this condition)

[Ans: Mo3O9 = 81, Mo4O12 ≈ 94 and Mo5O15 ≈ 105 kcal
mole–1 at 850 K] T(K) 1839 1862 1883 1887 1917 1925 1936 1939
36. The relative abundances and appearance potentials of
the principal ions in the mass spectrum of (a) phosphine, PCr×107
4.30 7.08 10.8 10.8 18.8 22.0 29.0 29.3
and (b) a mixture of phosphine and deuterated phos- (atom)
phines are listed below:
1959 1963 1998 2029 2059
70 eV relative Appearance Probable
(a) m/e 39.8 46.0 77.5 119 183
abundance potential (eV) Processes
31 24.1 17.2 —
Calculate the heat of vapourisation of Cr formed by the
32 100.0 13.3 — decomposition of Cr2O3(s).
33 21.8 13.2 —
−ΔHvap
• Hint: In P = + const; ΔHvap is in J mole–1 deg–1.
34 75.2 10.2 — 8.3143T
70 eV relative Appearance Ionic spe- [Ans: 94.2 kcal mole−1]
(b) m/e
abundance potential (eV) cies 38. LD/SI-MS/MS of 7-methylguanosinc exhibits peaks
at m/e: 282 (intaction) and 166 (base peak) in its
31 39.8 17.1 —
SI mass spectrum. What might be the ions corre-
32 85.8 13.3 — sponding to these peaks? The sturctural formulas of
guanoscne is given below:
33 100.0 13.1 — O
34 33.8 10.1 — HN N
35 58.0 10.1 — H2N N N

HOCH2 O
36 47.8 10.2 —
H H
37 22.9 10.1 — H H
HO OH
Guanosine
Fill in the blank columns in (a) and (b) and also compute
the heat of formation (ΔHf+) of the ionic species formed
in (a). ΔHf(H2) = 0.0, ΔHf(H) = 52.09, ΔHf(PH3) = 2.21 kcal
Suggested Readings
BOOKS
1. Abraham, A and Bleaney, B; Electron Paramagnetic Resonance of Transition Ions; Oxford Press, 1970
2. Abraham, R J and Loffus, P; Proton and Carbon-13 NMR Spectroscopy, An Integrated Approach; Heydon,
1978
3. Alger, R S; Electron Paramagnetic Resonance: Technique and Application; Wiley Interscience, 1968
4. Alpert, N A; Keiser, W E and Szymanski, H A; Theory and Practice of Infra-red Spectroscopy; Plenum
Press, 1970
5. Anderson, A; The Raman Effect, Vol. 1: Principles; Marcel Dekker, 1971
6. Anderson, A; The Raman Effect, Vol. 2: Principles; Marcel Dekker, 1973
7. Banwell, C N; Fundamentals of Molecular Spectroscopy, Tata McGraw-Hill, 1994
8. Barnes, A J and Orville-Thomas, W J (eds.); Vibrational Spectroscopy: Modern Trends; Elsevier, 1977
9. Becker, Edwin D; High Resolution NMR: Theory and Chemical Applications, 2nd Ed.; Academic Press,
1980
10. Bellany, L J; Advances in Infra-red Group Frequencies; Methuen, 1968
11. Bellany, L J; The Infra-red Spectra of Complex Molecules; Chapman and Hall, 1975
12. Bersohn, M and Baird, J C; An Introduction to Electron Paramagnetic Resonance; Benjamin, 1966
13. Breiglab, G; Elecktronen-donator-acceptor-komplexe, Springer-Verlag, 1961
14. Cantor, C R and Schimmel, P R; Biophysical Chemistry, Part II: Techniques for the Study of Biological
Structure and Function; W H Freeman, 1980
15. Casy, A F; PMR Spectroscopy in Medicinal and Biological Chemistry; Academic Press, 1971
16. Chang, R; Basic Principles of Spectroscopy; McGraw-Hill, 1971
17. Chantry, G W; Modern Aspects of Microwave Spectroscopy; Academic Press, 1979
18. Chapman, D; Biological Membranes, Vol. 2; D Chapman and DFH Wallach (eds.), Academic Press, 1973
19. Chapman, D; The Structure of Lipids; Methuen, 1965
20. Cohen, J S; Magnetic Resonance in Biology, Vol. 1, John Wiley, 1976
21. Darnell, D W and Wilkins, R G; Methods for Determining Metals in Environments in Proteins, Structure
and Function of Metal Proteins; Elesveir, 1980
22. Drago, R S; Physical Methods in Inorganic Chemistry; Reinhold Publishing Corporation, 1965
23. Dunitz, J D; Konstanze, P H; Ibers, J A; Jagensen, C K; Neilands, J B; Reimen, D and William, R J P;
Structure and Bonding, Vol. 17; Metal Bonding in Proteins; Springer-Verlag, 1973
24. During, J R; Vibrational Spectra and Structure, Vol. 4; Elsevier Scientific Publishing Company, 1975
25. Edmondson, D E; Biological Magnetic Resonance; Academic Press, 1978
26. Farrar, T C and Becker, E D; Pulse and Fourier Transform NMR, Introduction to Theory and Methods;
Academic Press, 1971
27. Ferraro, J R and Basile, L J; Fourier Transform Infra-red Spectroscopy: Application to Chemical System;
Academic Press, Vol.1, 1978; Vol. 2, 1979
28. Finch A F; Dickson, N and Bentley, F F; Chemical Applications of Far Infra-red Spectroscopy; Academic
Press, 1970
29. Forster, R; Organic Charge Transfer Complexes; Academic Press, 1969
30. Gerson, F; High Resolution ESR Spectroscopy; John Wiley, 1971
31. Gilson, T R and Hendra, P J; Laser Raman Spectroscopy; John Wiley, 1971
32. Grody, W and Cook, R L; Microwave Molecular Spectra, Techniques of Organic Chemistry; Vol. 9, John
Wiley, 1970
33. Grody, W; Smith, W V and Trambarulo, R; Microwave Spectroscopy; John Wiley, 1966
34. Harrich, N J; Internal Reflection Spectroscopy; Interscience, 1967
35. Herzberg, G; Infra-red and Raman Spectra of Polyatomic Molecules, Molecular Spectra and Molecular
Structure, Vol. 2; Van Norstrand, 1945
36. Herzberg, G; Spectra of Diatomic Molecules, Molecular Spectra and Molecular Structure, Vol.1, 2nd ed.;
Van Norstrand, 1950
37. Herzberg, G; Infra-red and Raman Spectra of Polyatomic Molecules, Molecular Spectra and Molecular
Structure, Vol. 2; Van Norstrand, 1945
38. Hester, R E and Girling, R B; Spectroscopy of Biological Molecules; Thomas Graham House, 1991
39. Ingram, D J E; Spectroscopy at Radio and Microwave Frequencies, 2nd ed.; Butterworth, 1967
40. Jaffe, H H and Orchin, M; Theory and Applications of Ultraviolet Spectroscopy; John Wiley, 1962
41. Kevan, L and Schwartz, R N; Time-domain Electron Spin Resonance, John Wiley, 1979
42. Kortum, G; Reflectance Spectroscopy, Springer, 1969
43. Kroto, H W; Molecular Rotation Spectra, Wiley-Interscience, 1975
44. Lindman, B and Forson, Sture; Chlorine, Bromine and Iodine NMR: Physico-chemical and Biological
Applications; Springer-Verlag, 1976
45. Loader, J; Basic Laser Raman Spectroscopy; Heydon, 1970
46. Miyazwa, T in G D Fasman (ed.), Series 1, Poly-a-amino Acids, Biological Macromolecules; Dekker,
1967
47. Mooney, E F; An Introduction in 19FNMR Spectroscopy; Heydon, 1970
48. Mooney, E F; Annual Review of NMR Spectroscopy, Vol. 2; Academic Press, 1969
49. Nakamoto, K; Infra-red Spectra of Inorganic and Coordination Compounds, 3rd ed.; John Wiley, 1978
50. Polpe, J A; Schneider, W G and Bernstein, H J; High Resolution NMR, McGraw-Hill, 1959
51. Ranby, B and Rabek, J F; ESR Spectroscopy in Polymer Research; Springer-Verlag, 1977
52. Rao, C N R; Chemical Applications of Infra-red Spectroscopy; Academic Press, 1963
53. Rosenwaig, A; Photoacoustic and Photoacoustic Spectroscopy; John Wiley, 1980
54. Ross, S D; Inorganic Infra-red and Raman Spectra; McGraw-Hill, 1972
55. Sanders, Jermy K M and Hunter, Brian K; Modern NMR Spectroscopy; Oxford University Press, 1988
56. Shaw, D; Fourier Transform NMR Spectroscopy; Elsevier, 1976
57. Sheppard, N in H W Thompson (ed.), Advances in Spectroscopy; Wiley-Interscience, 1959
58. Siesler, H W and Holland-Mortitz, K; Infra-red and Raman Spectroscopy of Polymers; Marcel Dekker,
1980
59. Silverstein, R M; Bassler, G C and Morril, T C; Spectrometric Identification of Organic Compounds; John
Wiley, 1991
60. Slichter, C P; Principles of Magnetic Resonance, 2nd ed.; Springer-Verlag, Berlin, 1978
61. Smith, A L; Applied Infra-red Spectroscopy; John Wiley, 1979
62. Smith, H J and Rawallo, F N; A Non-mathematical Approach to Basic MRI; Medical Physics Publishing,
1989
63. Steele, D; The Interpretation of Vibrational Spectra; Chapman and Hill, 1971
64. Sugden, T M and Kenny, N C; Microwave Spectroscopy of Gases, Van Nostrand, 1965
65. Suzuki, S; Iwashita, S; Tsuboe, M and Shimanouchi, T; Molecular ‘Structure and Spectroscopy’ Proc.
Intern Symp., Science Council of Japan, 1962, paper A110
66. Timasheff, S N and Fasman, G D; Structure and Stability of Biological Molecules; Marcel Dekker, 1980
Suggested Readings 941
67. Tobin, M C; Laser Raman Spectroscopy; Wiley-Interscience, 1971

68. Towens, C H and Schawlow, A L; Microwave Spectroscopy, McGraw Hill, 1955
69. Vinogradov, S N and Linnell, R H; Hydrogen Bonding; Van Nostrand, 1971
70. Wallach, D F H; Gordon, A S; Graham, J M and Fernbach, B R in F Snell, J Wolken, G J Inverson and J
Lam (eds.); Physical Principles of Biological Membranes; Gordon and Breach, 1970
71. Wehri, F W and Wirhthlin, T; Interpretation of Carbon-13 NMR Spectra; Heydon, 1976
72. Wertz, J E and Boulton, J R; Electron Spin Resonance: Elementary Theory and Practical Applications;
McGraw-Hill, 1972
73. West, W; Chemical Applications of Spectroscopy, Techniques of Organic Chemistry, Vol. IX; Interscience,
1968
74. Williams, D H and Fleming, I; Spectroscopic Methods in Organic Chemistry, 3rd ed.; McGraw-Hill, 1980
75. Wilson, E B; Decius, J C and Cross, P C; Molecular Vibrations; McGraw-Hill, 1955
76. Wollrab, J E; Rotational Spectra and Molecular Structure; Academic Press, 1967
77. Woodward, L A; Raman Spectroscopy Theory and Practice, H A Szymanski (ed.); Plenum Press, 1967
JOURNALS
1. Aleneastro, R B D E and Sandorfy, C; Can J Chem., 51 (1973) 1443

2. Anderson, J H Jr; Surface Science; 3 (1965) 290
3. Boyd, D R J and Thompson, H W; Trans Faraday Soc., 48 (1952) 502
4. Bulkin, B J and Krishnamachari, J; J. Am. Chem. Soc., 94 (1972) 1109
5. Chapman, D; Kamat, V B and Levene, R J; Science, 160 (1968) 314
6. Farrar, C; Anal Chem, 59 (1987) 749A
7. Frommherz; Peters, P J; Muldner, H G and Otting, W; Biochem. Biophys. Acta, 274 (1972) 644
8. Fukushima, F and Miyazawa, T; Paper presented at the Annual Meeting of the Chemical Society of Japan,
Tokyo, 1964
9. Gambhir, P and Nagarajan, S; PINSA, 65A (1999) 731
10. Gardner, R A and Petruci, R H; J. Am. Chem. Soc., 82 (1960) 5051
11. Gardner, R A; J. Phys. Chem., 64 (1960) 1120
12. Gill, D J; Kilponen, R G and Rimai, L; Nature, 227 (1970) 743
13. Graham, J M and Wallach, D F H; Biochem. Biophys. Acta, 193 (1969) 225
14. Graham, J M and Wallach, D F H; Biochem. Biophys. Acta, 241 (1971) 180
15. Green, D H and Salton, M R J; Biochem. Biophys. Acta, 221 (1970) 139
16. Haurie, M and Novak, A; J. Chem. Phys., 62 (1965) 137
17. Hiraishi, J and Shinoda, T; Bull. Chem. Soc. Japan, 48 (1975) 2385
18. Itoh, K; Nakahara, T; Shimanouchi, T; Oya, M; Uno, K and Iwakura, K; Biopolymers, 6 (1968) 1759
19. Jones, G I L; Lister, D G; Qwen, N L; Gerry, M C L and Palmieri, P; J. Mol. Spectroscopy, 60 (1976)
348
20. Kasendo, O and Huyskens, T H; J. Phys. Chem., 88 (1984) 2636
21. Kinney, J B and Staley, R H; Anal. Chem., 55 (1983) 343
22. Kiselev, A V; Surface Science, 3 (1965) 292
23. Koenig, J L and Sutton, P L; Biopolymers, 8 (1969) 167
24. Kreuzer, L B; Anal. Chem.,46 (1974) 239A
25. Krimm, S J; Mol. Biol., 4 (1962) 528
26. Lippert, J L and Peticolas, W L; Biochem. Biophys. Acta, 282 (1972) 8
27. Lippert, J L and Peticolas, W L; Proc. Nat. Acad. Sci. (US), 68 (1971) 1572
28. Lipscomb, W N; Acc. Chem. Res., 3 (1970) 81
29. Lord, R C and Yu, N T; J. Mol. Biol., 50 (1970) 509

30. Maddy, A H and Malcolm, B R; Science, 153 (1966) 213
31. Malcolm, B R; Biochem. J., 110 (1968) 733
32. Malcolm, B R; Biopolymers, 9 (1970) 911
33. Malcolm, B R; Nature, 195 (1962) 901
36. Manoharan, P T and Chaandramouli, G V R; PINSA, 65A (1999) 613
37. McCleland, J F; Anal. Chem., 55 (1983) 89A
38. McCrory, G A and Scheddel, R T; Anal. Chem., 30 (1958) 1162
39. Miyazawa, T and Scheddel, E R; J. Am. Soc. Chem., 83 (1961) 712
40. Miyazawa, T; Masuda, T and Fukushima, K; J. Polymer Sci., 62 (1962) 562
41. Phillips, D C; Sci. American, 215 (5) (1966) 78
42. Pliskin, W A and Eischens, R P; Z. Phys. Chem. (Frankfurt), 24 (1960) 11
43. Ragunathan, P; PINSA, 65A (1999) 699
44. Rakov; Res. Mol. Spectroscopy (Proceedings of the P N Lebedev, Physics Institute), 27 (1965) 109
45. Randhawa, H S; Walter, W and Meese, C O; J. Mol. Struct., 37 (1977) 187
46. Randhawa, H S and Walter, W; J. Mol. Struct., 37 (1976) 303
47. Randhawa, H S and Walter, W; J. Raman Spectroscopy, 11 (1981) 138
48. Randhawa, H S; J. Mol. Struct., 3 (1977) 25
49. Randhawa, H S and Rao, C N R; J. Cryst. Mol. Struct., 3 (1973) 309
50. Randhawa, H S; Reddy, N V R and Rao, C N R; Biopolymers, 13 (1974) 565
51. Randhawa, H S; Z. Phys. Chem., (Leipzig) 266 (1985) 1032
52. Rimai, L; Kilponen, R G and Gill, C J; J. Am. Chem. Soc., 92 (1970) 3824
53. Rimai, L; Gill, D J and Parsons, J J; J. Am. Chem. Soc., 93 (1971) 1353
54. Sarin, V N; Jain, Y S and Bist, H D; Thermochem. Acta, 6 (1973) 39
55. Schmid, H; Z Phys. Chem. (Leipzig) 259 (1978) 147
56. Susi, H; Timasheff, S N and Stevens, L; J. Biol. Chem., 242 (1967) 1553
57. Thompson, W K and Hall, D G; Trans Faraday Soc., 63 (1967) 1553
58. Timasheff, S N and Susi, H; J. Biol. Chem., 241 (1966) 249
59. Timasheff, S N; Susi, H and Stevens, L; J. Biol. Chem., 242 (1967) 5407
60. Tschalmer, H; J. Chem. Phys., 22 (1954) 1745
61. Wallach, D F H and Gordon, A; Federation Proc., 27 (1968) 1163
62. Wallach, D F H and Zahler, P H; Biochem. Biophys. Acta, 150 (1968) 186
63. Wallach, D F H and Zahler, P H; Proc. Nat. Acad. Sci., 56 (1966) 1552
64. Wallach, D F H; J. Gen. Physiol., 54 (1969) 35
65. Wallach, D F H; Graham, J M and Fernbach, B R; Arch. Biochem. Biophys., 131 (1969) 322
66. Wallach, D F H; Graham, J M and Oseroff, A; FEBS Letters, 7 (1970) 330
67. Walter, W and Kubersky, H P; J. Mol. Struc., 11 (1972) 207
68. Watson, C; Anal. Chem., 55 (1983) 1165A
69. Williams, V Z; Anal. Chem., 29 (1957) 1551
Nobel Prizes
Nobel Prize in Physics and Chemistry Related to Molecular Spectroscopy

and Structure (1901–2005)
• 1902: (jointly to) HENDRIK ANTOON LORENTZ and PIETER ZEEMAN for influence of magnetic field
of magnetism upon radiation phenomena
• 1914: MAX VON LAUE for diffraction of X-rays by crystals
• 1915: (jointly to) SIR WILLIAM HENRY BRAGG and SIR WILLIAM LAWRENCE BRAGG for analy-
sis of crystal structure by X-rays
• 1919: JOHANNES STARK for discovery of the Doppler effect in canal rays and the splitting of spectral lines
in electric fields
• 1924: KARL MANNE GEORG SIEGBAHN for his discoveries and research in the field of X-ray spec-
troscopy
• 1930: SIR CHANDRA SEKHARA VENKATA RAMAN for work on the scattering of light and for discovery
of the effect named after him
• 1952: (jointly to) FELIX BLOCH and EDWARD MILLS PURCELL for development of new methods for
nuclear magnetic precision measurements and discoveries in connection therewith
• 1961: (divided equally between) ROBERT HOFSTADTER for pioneering studies of electron scattering in
atomic nuclei and thereby achieved discoveries concerning the structure of the nucleons; RUDOLF LUDWIG
MOSSBAUER for researches concerning the resonance absorption of gamma radiation
• 1964: *DOROTHY CROWFOOT HODGKIN was awarded the Nobel Prize for chemistry for her determina-
tion by X-ray techniques of the three-dimensional structure of important biological substances, for example,
cholesterol, penicillin, Vitamin B12, and zinc insulin (which contains almost 800 atoms)
• 1971: *G HERZBERG for work on electron structure and geometry of molecules particularly of free radicals
(molecular spectroscopy)
• 1981: The Prize was awarded by one half jointly to NICOLAA BLOEMBERGEN and ARTHUR L SCHAW-
LOW for their contribution to the development of laser spectroscopy and the other half to KAI M SIEGBAHN
for his contribution to the development of high resolution electron spectroscopy
• 1992: *RICHARD R ERNST for his contribution to Fourier transform nuclear magnetic resonance spectros-
copy
• 1994: For pioneering contributions to the development of neutron scattering techniques for studies of condensed
matter to BERTRAM N BROCKHOUSE for the development of neutron spectroscopy and CLIFFORD
G SHULL for the development of the neutron diffraction technique
• 2002: *The prize was awarded by one half jointly to JOHN B FENN and K TANAKA for developing soft
desorption ionization method for mass spectrometric analysis of biological macromolecules and the other half
to K WUTHRICH for his contribution to the development of nuclear magnetic resonance spectroscopy for
determining the three-dimensional structure of biological macromolecules in solution
• 2005: The Nobel Prize was awarded by 2/3rd jointly to JOHN L HALL and THEODOR W HAENNSCH for
their contribution to the development of laser based-precision spectroscopy including the optical frequency
comb technique while ROY J GLAUBER won 1/3rd of the prize for his contribution to the quantum theory of
optical coherence

S K Dogra - Molecular Spectroscopy-McGraw-Hill Education (2012)

Uploaded by

Copyright:

Available Formats

S K Dogra - Molecular Spectroscopy-McGraw-Hill Education (2012)

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

S K Dogra - Molecular Spectroscopy-McGraw-Hill Education (2012)

Uploaded by

Copyright:

Available Formats

Molecular Spectroscopy

About the Author

Tata McGraw Hill Education Private Limited

McGraw -Hill Offices

Typeset at Mukesh Technologies Pvt. Ltd., Puducherry.

2.8 Thermal Distribution of Population Among the Rotational Levels 58

3.15.3 Theoretical Aspects of PR-separation and IQ /ITotal in Different

4. RAMAN SPECTROSCOPY 224

4.12.2 Mechanism of Tautomerism and Polymerisation 248

5. ELECTRONIC SPECTROSCOPY 256

5.16.1 Energy Levels in Transition Metal Complexes 317

6. PHOTOELECTRON SPECTROSCOPY 457

6.8.1 Rules to Predict the Intensities of Ionisation Peaks 466

7. NUCLEAR MAGNETIC RESONANCE SPECTROSCOPY 507

7.31.1 Effect of Temperature 553

8. NUCLEAR QUADRUPOLE RESONANCE SPECTROSCOPY 610

8.12 Effects of Various Factors on Nuclear Quadrupole Resonance Frequency 633

10. MÖSSBAUER SPECTROSCOPY 701

10.5 Line Width of Mössbauer Line and Factors Affecting It 707

11. MASS SPECTROMETRY 790

11.5.1 Magnetic Field Deﬂection Mass Spectrometry 802

Suggested Readings 939

Overview of the subject

A teacher should produce less heat and more light.

1.1 THE NATURE OF ELECTROMAGNETIC RADIATIONS

Direction of Wave Motion

(b) Wave Number The reciprocal of wavelength is called wave number.

Table 1.1 Conversion factors of energy units.

Units cm-1 Ergs/molecule Cal/mole eV

1 cm−1 1 1.9865 × 1016 2.859 1.2398

1.2 MOLECULAR SPECTROSCOPY AND SPECTRAL REGIONS

Branch Phenomenon Wavelength Information

1.3 BORN–OPPENHEIMER APPROXIMATION

1.4 SCHRÖDINGER WAVE EQUATION

1.5 GENERAL CONDITION OF RESONANCE

1.6 BOLTZMANN DISTRIBUTION LAW AND POPULATION

1.7 WIDTHS AND SHAPES OF SPECTRAL BANDS

where tp is the contribution to bandwidth from pressure effect and is given by

Normal Absorption Curve

First Derivative Curve

Second Derivative Curve

Fig. 1.8 Comparison of (a) absorption, and (b) derivative curves.

1.8 INTENSITY OF THE SPECTRAL BANDS (OR QUANTUM MECHANICAL

Table 1.3 Terms and symbols used in Beer–Lambert law.

Current Symbol Deﬁnition Current Name Obsolete Symbol Obsolete Name

1.9 SPECTRUM AND BASIC ELEMENTS OF A SINGLE-BEAM AND

Source Slit Splitter Sample Detector Analyzer

1.10 SIGNAL-TO-NOISE RATIO

1.11 RESOLVING POWER OF A SPECTROMETER AND ITS RELATION

1.12 LIMIT OF SENSITIVITY OF THE SPECTROSCOPIC METHOD

1.13 FOURIER TRANSFORM (FT) AND COMPUTER AVERAGE

1.14.1 What is Stimulated Emission?

Now, when the photon strikes an excited atom which R

A2; A2 >A1) in and out of phase is shown in Fig. 1.17. The

deﬁned by optical resonant cavity which in its simplest A2 − A1

1.14.2 Wavelength Range and Power Output of Lasers

Table 1.4 Wavelengths of some commercial lasers.

The lasing range covered by various dyes runs from CO2

Pumping by 1.3-mm laster

1.15 PLANE-POLARISED RADIATION AND DICHROISM