IDENTIFICATION

OF MATERIALS
VIA PHYSICAL PROPERTIES
CHEMICAL TESTS AND MICROSCOPY

BY

A. A. BENEDETTI-PICHLER
QUEENS COLLEGE OF THE
UNIVERSITY OF THE CITY OF NEW YORK

WITH 2 PLATES AND 75 FIGURES

WIEN
SPRINGER-VERLAG
1964
 ISBN-13: 978-3-7091-8109-6 e-ISBN-13: 978-3-7091-8107-2
DOl: 10.1007/978-3-7091-8107-2

ALLE RECHTE, INSBESONDERE DAS DER UBERSETZUNG

IN FREMDE SPRACHEN, VORBEHALTEN
OHNE SCHRIFTLICHE GENEHMIGUNG DES VERLAGES
1ST ES AUCH NICHT GESTATTET, DIESES BUCH ODER TEILE DARAUS
AUF PHOTOMECHANISCHEM WEGE (PHOTOKOPIE, MIKROKOPIE)
ODER SONSTWIE ZU VERVIELFA.LTIGEN
ALL RIGHTS INCLUDING TRANSLATION INTO OTHER LANGUAGES RESERVED
NO PART OF THIS BOOK MAY BE REPRODUCED IN ANY FORM,
BY PHOTOSTAT, MICROFILM, OR ANY OTHER MEANS, WITHOUT WRITTEN
PERMISSION FROM THE PUBLISHERS
© 1964 BY SPRINGER-VERLAG/WIEN
LIBRARY OF CONGRESS CATALOG CARD NUMBER 64-16107

SOFTCOVER REPRINT OF THE HARDCOVER 1ST EDITION 1964

 Preface
This book has been written for the practicing chemist whose occasional
task may be qualitative analysis. It deals with the investigation of things
as they are without any limitations to the scope. It emphasizes the
identification of materials - inorganic, organic, organized (biological),
common, rare, described or not described in the accessible literature -
as they actually occur in nature and industry, or are met in the investigation
of mishaps and crime.
The description of techniques - macro to submicro - and the practice
exercises have been included since the teaching of these arts is rarely a
part of academic curricula and it happens with increasing frequency that
chemists have to acquire them "on the job".
In the systematic procedure given, emphasis is placed upon the investiga-
tion of minute specimens and upon acute reasoning that continuously
weighs all accumulating evi9.ence. The work begins with the consideration
of the history of the material under investigation. Especially when specks
of all organic substance shall be identified, it should be realized that the
discovery of the source - and consequently of the possibilities involved -
may be the most valuable clue to an efficient solution of the problem.
Because of the increasing practical importance of the identification
of small specimens, the systematic scheme starts with the application
of non-destructive tests. Inspection of solids under the microscope reveals
material of biological origin, and references to the literature are given
for· a continuation of the search. The appearance under the microscope
may also permit to recognize man-made materials (artifacts), and the
microscope may be used as an aid in studying color, fluorescence, hardness,
refraction, and crystallographic data. Tests for radio activity and ferro-
magnetism are applied, and the density may be determined or estimated.
If the definitely non-destructive tests do not permit recognition, the
solid is first heated to 300 0 C and then to higher temperatures. Its behavior
and the testing of products of decomposition will usually reveal whether
the material is inorganic or organic. The observed transition temperatures
of organic substances may permit prompt identification by the performance
of "eutectic" and "mixed" melting points requiring very little material.
Simple inorganic substances may be recognized sufficiently by the accumulat-
ing evidence to permit arriving at final decisions by decisive confirmatory
tests.
IV Preface

At the same time, enough information has been obtained to decide

whether instrumental or chemical methods are best suited to continue the
investigation. Advice is supplied to aid in arriving at an intelligent decision,
and directions are given for the chemical analysis of inorganic materials.
Instructions for the separation of analytical groups are outlined. The
separation of the tantalum group of NOYES and BRAY is included, but
otherwise it is assumed that the isolation of a group will suffice to identify
the members - rare as well as common - by specific tests.
It is the opinion of the author that the book may also serve as a basis
for an interesting laboratory course. The traditional course in qualitative
analysis, which usually gravitates toward mechanical, lengthy, and tedious
separations, is getting discarded to the detriment of the training in inorganic
chemistry. The new approach can bring new life into the laboratory practice
of qualitative analysis since it is based upon nicety of observation and
penetrating reasoning which often lead to surprisingly quick identification.
The recommended frequent use of the microscope - and a very inexpensive
one will do - will contribute to the enjoyment of the work and familiarize
the student of chemistry with a very valuable tool.
The book is based upon a critical evaluation of the experience gained
in 40 years of teaching and occasional practice of qualitative analysis,
micro analysis, and chemical microscopy. It makes use of the contributions
of about three generations of analytical chemists, and the more glaring
appropriations are duly acknowledged by references to the literature.
The basic idea for the systematic scheme of identification stems from a
recollection of an ingenious and simple course in qualitative analysis given
to us by Professor KARL WITMANN, around 1910 at the 1. Staatsrealschule
in Graz, Austria. For inspiration, I am primarily in debt to him and to
my teacher FRIEDRICH EMICH.
Furthermore, I am indebted to my late friend NICHOLAS D. CHERONIS
for encouragement; to the analytical staffs of the Technical Service
Laboratory of the Socony Mobil Oil Company and the Technical Center
of General Motors for being kept aware of practical needs and procedures;
and certainly also to my associates, coworkers, and students for criticism,
suggestions, help, and the testing of procedures.
Special gratitude goes to the publisher and to all those who had a
part in the production of the book for the understanding shown and the
special care taken.

New¥ork, December 1963

A. A. Benedetti-Pichler
 Contents
l'age
Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
The Task of Qualitative Analysis.. . . . . . . . . . . . . . . . . . . . . . . . . . 1
Principles and Definitions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
Specificity and Sensitivity............................... . . . 4
The Method of Qualitative Analysis.... . . . . . . . . . . . . . . . . . . . . . 6
The Minimum Size of the Sample for Chemical Analysis ..... 7
Properties Affected by the Size of the System............... 9
Apparatus, Technique, and Scale of Work. . . . . . . . . . . . . . . . . .. 11
Observation of Properties. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
Selection of Procedure....... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18

Part I: Technique of Observation and Manipulation . . . . . . . . . . . . . . . . . . .. 20

Use of Optical Aids .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
The Microscope......... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 20
Experiment I. Inspecting and Cleaning the Microscope. . . .. 28
Illumination of the Specimen ............................ :.. 34
Experiment 2. Illumination of Microscopical Specimens..... 35
Experiment 3. Calibration of Eyepiece Micrometer; Working
Distances and Fields of Vision . . . . . . . . . . . .. 37
The Immersion Method for the Determination of Refraction.. . .. 39
Experiment 4. Phenomena Caused by Differences in Refrac-
tion. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 41
Observation of Schlieren................................... 44
Experiment 5. Visual Observation of Schlieren . . . . . . . . . . . . 45
Use of Polarized Light .................................... 47
Experiment 6. Testing and Adjusting a Polarizing Microscope 50
Experiment 7. Isotropic and Anisotropic Substances in
Polarized Light..... . . . . . . . . . . . . . . . . . . . . . . 52
Experiment 8. Determination of Vibration Directions in Rela-
tion to Profile. . . . . . . . . . . . . . . . . . . . . . . . . . .. 55
Experiment 9. Use of Compensators. . . . . . . . . . . . . . . . . . . . . .. 57
Experiment 10. On the Determination of the Refractive
Indices of Anisotropic Materials. . . . . . . . . . . 60
Experiment II. Observation of Pleochroism... . . . . . . . . . . .. 61
Experiment 12. Observation of Axial Figures ............. 61
Experiment 13. Transition Phenomena in Polarized Light. . . . 63
Experiment 14. Preserving Microscopical Preparations. . . . . .. 65
Technique of Experimentation and Observation . . . . . . . . . . . . . . . . . . . . 67
Basic Rules... ...... .... . .. . ..... ..... .. ..... . ... . .. . ... .. 67
Work on the Gram Scale . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68
Experiment 15. Phenomena Observed Upon Heating in an
Inert Atmosphere . . . . . . . . . . . . . . . . . . . . . . .. 75
VI Contents

Experiment 16. Phenomena Observed Upon Heating in a

Stream of Air . . . . . . . . . . . . . . . . . . . . . . . . . .. 77
Experiment 17. Heating Upon the Charcoal Block......... 78
Experiment 18. Performance of Flame Tests .............. 79
Work on the Centigram Scale................................. 81
Experiment 19. Preparation of Capillaries, Capillary Pipets,
and Microburners ........................ 99
Experiment 20. Emich's Method of Fractional Distillation .. 101
Experiment 21. Emich's Method for the Determination of
Boiling Points . . . . . . . . . . . . . . . . . . . . . . . . . .. 102
Work on the Milligram Scale.................................. 104
Experiment 22. Calibration of Capillary Pipets and Centrifugal
Pipets ................................... 125
Experiment 23. Preparation and Calibration of Platinum
Loops and Hooks.... . . . . . . . . . . . . . . . . . . .. 127
Techniques of the Submilligram Scale .......................... 128
Spot Tests ............................................... '. 128
Experiment 24. Test for Mercuric Mercury and Lead ...... 131
Experiment 25. Chromate Test for Silver . . . . . . . . . . . . . . . .. 132
Experiment 26. Tests for Bismuth and Antimony.......... 132
Experiment 27. Test for Copper, Nickel, and Cobalt ....... 134
Experiment 28. Test for Cadmium ....................... 134
Experiment 29. Precipitation of Silver Arsenate . . . . . . . . . .. 135
Experiment 30. Molybdenum Blue Test for Tin ........... 135
Slide Tests................................................ 136
Experiment 31. Silver Dichromate ....................... 139
Experiment 32. Recrystallization of Silver Chloride . . . . . . .. 139
Experiment 33. Lead Iodide...................... . . . . . .. 140
Experiment 34. Potassium-Lead-Copper Nitrite ............ 141
Experiment 35. Cesium Iodobismuthite and Cesium Iodoanti- ,
monite .................................. 142
Experiment 36. Bismuth Cobalticyanide Pentahydrate ...... 143
Experiment 37. The Mercurithiocyanates of Copper, Zinc,
Cadmium, and Cobalt .................... 144
Experiment 38. Test for Cadmium with Brucine and Bromide 146
Experiment 39. Test for Mercury with Zinc, Copper, and Thio-
cyanate . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 146
Experiment 40. Magnesium-Ammonium Arsenate and Silver
Arsenate ................................ 147
Experiment 41. Rubidium Chlorostannate ............ '..... 148
Experiment 42. Estim;ttion of the Relative Quantities of the
Metals in a Slurry Containing Arsenic, Anti-
mony, Copper, and Silver................ 149
Working Upon the Surface of a Slide ....................... 150
Experiment 43. Conversion of Silver Chloride to Silver Di-
chromate. Handling Precipitates and Solu-
tions, Fusion, and Electrolytic Reduction .. 151
Experiment 44. Separation of Bismuth and Lead. Evapora-
tion, Extraction of "Invisible" Residues ... 153
 Contents VII
Experiment 45. Separation of Silver, Lead, and Mercurous
Mercury, Sublimation, Extraction with Boiling
Solutions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 156
Experiment 46. Test for Ammonium Ion, Use of the Gas
Reaction Cell............................ 158
Working in Capillaries ..................................... 159
Experiment 47. Lead Sulfate, Triple Nitrite, Lead Chromate.
The Capillary as Adjunct to Working Upon
the Microscope Slide ..................... 160
Experiment 48. Recrystallization of Lead Iodide .......... 161
Experiment 49. Isolation of Metallic Mercury, Conversion to
Iodide .................................. 162
Experiment 50. Bettendorff's Test for Arsenic. . . . . . . . . . . . . .. 164
Experiment 51. Oxidation of Arsenic to Arsenic Acid. Carius'
Treatment in Capillaries .................. 165
Experiment 52. Conversion of Aniline to Acetanilide ....... 167
Experiment 53. Conversion of Urea to Symmetrical Diphenyl
Urea .................................... 170
Experiment 54. Purification of Benzene; Separation by Par-
tial Melting in the Capillary . . . . . . . . . . . . .. 171
Experiment 55. Cupric Ammonia Complex, Observation of
Color in the Capillary .................... 175
Working in Filter Paper ................................... 177
Experiment 56. Ring-Oven Technique for Extraction and
Evaporation in Paper .... . . . . . . . . . . . . . . .. 180
Particles of Ion Exchange Resins as Reaction Media ......... 182
Working on Textile Fibers and Wires ....................... 183
Experiment 57. Turmeric Test for Boric Acid... . . . . . . . . .. 188
Experiment 58. Test for Bismuth, Precipitation of the
Sulfide Upon the Fiber and Conversion to
Sulfate, Chromate, and Elemental Bismuth 190
Experiment 59. Bead Test for Cobalt .................... 191
Experiment 60. Luminescence Test for Bismuth, Antimony,
and Manganese ...... . . . . . . . . . . . . . . . . . . .. 193
Work on the Microgram Scale ................................. 193
Apparatus ................................................ 194
Technique .............•............ ,...................... 202
Experiment 61. Mechanical Separation of the Components of
a Powder ............................... 219
Experiment 62. Precipitation of Silver Dichromate Upon the
Platform of the Condenser Rod .. . . . . . . . .. 221
Experiment 63. Estimation of the Quantities of Arsenic and
Antimony in a Solution of Unknown Con-
centration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 222
Additional Practice Experiments for the Chosen Scale of Work 224
Experiment 64. Study of Chemical Behavior . . . . . . . . . . . . .. 224
Experiment 65. Analysis of Two Unknown Solutions ...... 225
Experiment 66. Identification of Simple Compounds of the
Common Metals of the Hydrogen Sulfide
Group .................................. 225
VIII Contents

Experiment H7. Identification of Simple Inorganic Com-

pounds .................................. 226
Experiment 68. Identification of Simple Compounds ....... 227
Experiment 69. Identification of Simple Compounds .. . . . .. 227
Experiment 70. Identification of Materials as They Occur in
Nature, Industry, and Research ........... 227

Part II: Systematic Analysis. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 228

Choice of Materials and Cleaning ........................... 228
Sampling for Analysis ........................................... 230
Systematic Procedure of Analysis.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 234
The History of the Sample ................................. 234
Description of Sample and Record of Investigation ............ 235
Preliminary Inspection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 235
The Sample is a Liquid. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 236
Identification of Organized Matter .......................... 240
Identification of Artifacts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 242
Well-Developed Crystals ................................... 243
Solids of Random Shape and Structure . . . . . . . . . . . . . . . . . . . .. 244
Non-Destructive Testing.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 244
Action Upon Light, Color.................................. 244
Investigation of Crystals and Crystal Fragments. . . . . . . . . . . . . .. 247
Testing for Radioactive Decay ............................. 253
Testing for Ferromagnetism ................................ 254
Odor ..................................................... 256
Hardness ................................................. 258
Refractive Index .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 2HO
Density .................................................. 2H2
Classification Tests ........................................... 2H8
Observation of Transition Points Below 350 C .............. 0
2()9
Ignition Above 300 0 C ..................................... 274
Solubility ................................................ 285
Performance of Solubility Tests ......................... 287
Review: Inorganic Substances . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 2!H
Organic Substances................................ 294
Elemental Analysis of Organic Substances . . . . . . . . . . . . . . . . . .. 299
Testing with Dilute Sulfuric Acid. . . . . . . . . . . . . . . . . . . . . . . . . .. 301
Test with Concentrated Sulfuric Acid .. . . . . . . . . . . . . . . . . . . . .. 309
Flame Tests .............................................. 313
Bead Tests ............................................... 315
Review of Findings. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 317
Dissolution of the Sample.................................. 320
Treatment of Substances Insoluble in Acids ................. 322
Confirmatory Tests . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 324
Group I A: Alkali Metals. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 325
No.3: Lithium, H.939 ................................... 325
No. 11: Sodium, 22.9898 ............ , ................... 32H
No. 19: Potassium, 39.102 .............................. 32H
No. 37: Rubidium, 85.47 ................................ 327
No. 55: Cesium, 132.905 ................................ 328
 Contents IX

Group II A: Alkaline Earths ............................... 329

No.4: Beryllium, 9.0122 ..........••••...•.••••••..••••• 329
No. 12: Magnesium, 24.312 . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. a29
No. 20: Calcium, 40.08 ................................. 330
No. 38: Strontium, 87.62 ............................... 330
No. 56: Barium, 137.34 ................................. 331
Group III B: Scandium Group ............................. 331
No. 21: Scandium, 44.956 ............................... 331
No. 39: Yttrium, and the Lanthanides, Nos. 57 to 71 ...... 332
No. 58: Cerium, 140.12 ................................. 332
No. 63: Europium, 151.96 ............................... 333
No. 70: Ytterbium, 173.04 .............................. 333
Group IV B: Titanium Group............................... 333
No. 22: Titanium, 47.90 ................................. 333
No. 40: Zirconium, 91.22 ................................ 334
No. 90: Thorium, 232.038 ............................... 335
Group V B: Vanadium Group ............................... 335
No. 23: Vanadium, 50.942 .............................. 335
No. 41 : Niobium, 92.906 .. , . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 335
No. 73: Tantalum, 180.948 ., . . . . . . . . . . . . . . . . . . . . . . . . . . .. 336
Group VI B: Chromium Group .............................. 336
No. 24: Chromium, 51.996 ............................... 336
No. 42: Molybdenum, 95.94 .............................. 337
No. 74: Wolfram (Tungsten), 183.85 ...................... 337
No. 92: Uranium, 238.03 ................................ 338
Group VII B: Manganese Group. . . . . . . . . . . . . . . . . . . . . . . . . . . . .338
No. 25: Manganese, 54.9380 ............................. 338
No. 75: Rhenium, 186.2 ................................ 340
Group VIII: Fe-Ni Triad .................................. , 340
No. 26: Iron, 55.847 .................................... 340
No. 27: Cobalt, 58.9332 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 341
No. 28: Nickel, 58.71 ................................... 342
Group VIII: Ru-Pd Triad .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 343
No. 44: Ruthenium, 101.07 ............................. 343
No. 45: Rhodium, 102.905............................... 344
No. 46: Palladium, 105.4 ................................ 344
Group VIII: Os-Pt Triad. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 345
No. 76: Osmium, 190.2 ................................. 345
No. 77: Iridium, 192.2 " ................................ 346
No. 78: Platinum, 195.09 ................................ 346
Group I B: Copper Group .................................. 347
No. 29: Copper, 63.54 ................................... 347
No. 47: Silver, 107.870 .................................. 347
No. 79: Gold, 196.967 .................................. 348
Group II B: Zinc Group ................................... 349
No. 30: Zinc, 65.37 ..................................... 349
No. 48: Cadmium, 112.40 ................................ 351
No. 80: Mercury, 200.59 ................................ 351
Group III A: Boron-Thallium Group ........................ 351
No.5: Boron, 10.811.................................... 351
x Contents

No. 13: Aluminium, 26.9815 ............................. 352

No. 31: Gallium, 69.72 ................................. 353
No. 49: Indium, 114.82 ................................. 354
No. 81: Thallium, 204.37 ............................... , 355
Group IV A: Carbon-Lead Group ........................... 355
No.6: Carbon, 12.01115 ................................ 355
No. 14: Silicon, 28.086 .................................. 362
No. 32: Germanium, 72.59 ............................... 363
No. 50: Tin, 118.69 .................................... 364
No. 82: Lead, 207.19 .................................... 365
Group V A: Nitrogen-Bismuth Group . . . . . . . . . . . . . . . . . . . . . . .. 365
No.7: Nitrogen, 14.0067 ................................ 365
No. 15: Phosphorus, 30.9738 ............................. 369
No. 33: Arsenic, 74.9216 ................................ 373
No. 51: Antimony, 121.75 ............................... 374
No. 83: Bismuth, 208.980 ............................... 375
Group VI A: Oxygen-Polonium Group ....................... 375
No.8: Oxygen, 15.9994 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 375
No. 16: Sulfur, 32.064 .................................. 377
No. 34: Selenium, 78.96 ................................ 382
No. 52: Tellurium, 127.60 ............................... 383
Group VII A: Halogen Group. . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 385
No.9: Fluorine, 18.9984 ................................. 385
No. 17: Chlorine, 35.453 ................................ 387
No. 35: Bromine, 79.909 ................................ 391
No. 53: Iodine, 126.9044 ................................ 393
Separations .................................................. 395
Systematic Schemes for the Detection of Cations .............. 396
The Classical Scheme . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 396
Outline for the Separation of the Analytical Groups ........ 397
Observations and Notes ................................ 400
Analysis of Metals and Alloys Attacked by Nitric Acid ...... 403
Separation of the Analytical Groups of NOYES and BRAY .. , 404
Systematic Search for Anions .............................. 422
I. Only Alkali Metals or (and) Ammonium Are Present ...... 424
II. Nonmetallic Materials Readily Dissolved or Decomposed
by Water or Acids . .. . .. .. .. . .. .. .. .. .. .. .. . . .. .. ... 427
III. Nonmetallic Refractory Materials .................... 427
Final Review of Observations and Report ......................... 428

Appendix ......................................................... , 430

Test Solutions ................................................ 430
Preparation of Unknowns .............................. ~ ...... 431
Reagents .................................................... 432
Table 1. Color of Some Inorganic Substances ................... 432
Table 2. Substances Crystallizing in the Cubic System ........... 433
Elements ... . . . . . . . . . . . . . . . . . . . • . . . . . . . •. . . . . . . . . . . .. 433
Inorganic Compounds. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 433
Organic Compounds .. , ................................ 435
 Content,; XI

Table 3. Substances Crystallizing in the Hexagonal System ........ 435

Elelnents ................. . . . . . . . . . . . . . . . . . . . . . . . . . .. 435
Inorganic Compounds................................. 436
Organic Compounds ................................... 437
Table 4. List of Common Inorganic Compounds in the Order of Their
Melting Points ....................................... 438
Table .5. Inorganic Substances that Sublime, Arranged According to
Color ............................................... 457
Table 6. Inorganic Solids which Burst into Flame when Heated in
Air, Ignition Temperatures in Centigrades .. . . . . . . . . . . .. 458
Tahle 7. List of Solids which Explode on Heating .............. 459
Inorganic Compounds ................................. 459
Organic Com pounds ................................. " 460
Tahle 8. Inorganic Solids Moderately Soluble in Water at Room
Temperature ......................................... 461

Literature ...................................................... 463

General Reference Books ..................................... 463
Theory of Chemical Analysis .................................. 463
Reagents .................................................... 464
Standard Tests and Procedures of Qualitative Analysis .......... 464
Chromatography and Ion Exchange.. . . . . . . . . . . . . . . . . . . . . . . . . .. 465
Instrumental Methods ........................................ 466
Chemical Microscopy ......................................... 466
Slide Tests and Spot Tests ................................... 468
Micro Analysis and Microtechnique ............................ 469
Miscellaneous Applications of Microtechnique .................... 470
Mineralogy ................................................... 471
Journals .................................................... 472
Reports. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 483
Theses ...................................................... 483
"Unpublished Experiments ..................................... 483
Private Communications ....................................... 483
Demonstrations of Microgram Technique by Dr. M. CEFOLA . . . . .. 484
Meetings. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 484
Addresses ................................................... 484

Sub j e c tIn d ex. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 485

 Introduction
The Task of Qualitative Analysis
The purpose of qualitative analysis is the discovery of the essential
nature of the object under investigation. A procedure of qualitative
analysis, which shall be useful in actual analytical practice, must be
generally applicable. It must give information without regard whether
the object is a mixture or a pure substance; rare or common; simple or
complex; organized, organic or inorganic; and whether it is known and
described or as yet "unknown" and not yet mentioned in the literature.
In addition to this, the procedure must be applicable to very small amounts
of sample since qualitative analysis is frequently needed to determine
the nature of minute objects.
It is impossible to ever present a complete practical scheme of qualitative
analysis between the covers of a book since the practicing analyst may
have to draw on any and every part of our knowledge of material things.
A sensible approach, however, will give some useful information under all
conditions and at the same time provide a high probability that simple
problems will be solved satisfactorily and efficiently. The amount of
information that can be gained with a suitable scheme will depend on the
nature of the object. Obviously, qualitative identification must fail
whenever distinguishing features are missing, and it may be impossible
to indicate more than the general class of matter (lanthanides, mixture
of fatty acids, proteid substance).
It is also plain that lack of material may seriously limit the scope of
investigation and the amount of information that can be gained. Assuming,
however, the existence of suitable apparatus and technique (145), one
may say that the amount of material available for testing is only a minor
factor in estimating the difficulty of the task. The time and labor needed
depend in the first place upon the amount of information available concerning
the sample and upon the amount of knowledge to be gained by the analysis.
Under otherwise equal conditions, it will depend mainly upon the complexity
of the object of investigation and upon the number and kind of separations
required, whereby the emphasis should be on the latter since the number
of constituents in the sample is by far not as important as their nature
and their mass ratios. '
If the object is a substance which is not yet described in the literature (146),
qualitative analysis may· or may not reveal this fact. It may furnish
Benedetti-Pichler, Identification 1
2 Introduction

descriptive material, possibly some data concerning physical and chemical

properties, the elemental composition as to kind, and perhaps evidence
concerning functional groups and the general class of substances to which
it belongs. As a rule, however, precise determination of percentage composi~
tion, number of functional groups, molecular weight, etc., are required to
give a new substance its proper name, i. e., to assign it the correct niche in
our classification of matter. If some unique behavior, physical or chemical,
is discovered in the course of the investigation, this will facilitate the
recognition of the substance when it occurs again.
This book outlines a plan of investigation for the use of the practicing
chemist. Emphasis is placed upon a proper start which assures accumulation
of useful information with small expenditure of sample material and the
quick identification of common simple materials. References to the liter~
ature serve to widen the field of investigation in all directions toward
the limits of our knowledge. Descriptions of technique are included to
permit the investigation not only of small samples but of truly minute
objects.

Principles and Definitions

In the strict sense, analysis (lhd).I'I1I(J = resolution) should apply only
to the method of taking apart or breaking down things for the purpose
of investigation. Destruction of the sample, however, is not necessarily
a part of qualitative analysis.
The modifier qualitative, too, should be accepted with a rather loose
meaning. It indicates that no effort will be made toward a precise determina-
tion of the mass ratios if it should turn out that the sample is a mixture
of several chemical specimens. On the one hand, the purely "qualitative"
finding that the sample consists of sodium chloride implies the quantitative
facts that it is composed of 60.67 per cent chlorine and 39.33 per cent
sodium and represents practically lOO per cent of the combination of the
two. On the other hand, a statement that the sample consists of sodium,
calcium, sulfate, chloride, and water would not permit identifying the
object of analysis. It could be gypsum contaminated by some rock salt,
or table salt contaminated with gypsum, or some hydrate of calcium
chloride mixed with some sodium sulfat.e, or a concentrated or dilute
.solution of some combination of the four possible salts, etc. Obviously,
an approximate statement of the ratio of the ingredients is required before
the object of inve8tigation can be given a name; a merely qualitative
stat.ement does not suffice in such instances where several chemical specimens
are found in the sample.
The primary task of qualitative analysis, the disclosure of the general
nature of the sample, is usually accomplished by a rather crude classification
according to mass ratio. Customarily one distinguishes majors (more than
 Principles and Definitions 3

5 per cent of the whole), minors (between 0.1 to 5 per cent), and traces
(less than 0.1 per cent of the whole).
Whereas observations of mass ratios are frequently required, the
performance of laborious quantitative determinations is avoided since the
general nature of the object, as a rule, may be recognized without precise
knowledge of ratios. In most instances it will suffice to know the major
and minor constituents. A more exacting qualitative task, the identification
of a special steel, of a special batch of copper, or of a reagent taken from
a certain bottle, etc., may be solved by a study of the trace constituents
or may require precise quantitative determinations.
There is no sharp dividing line between analysis and micro analysis.
One speaks of micromethods whenever the experimentation proceeds
. with significantly less material than is customarily used for the purpose.
If "significantly" is translated to mean ten times, there still remains the
interpretation of "customarily", which will change with time and circum-
stance. In addition, there is the fact of the continuously varying amount
of matter (sample) taken for testing. Schemes of separation for inorganic
qualitative analysis usually start with one decigram to one gram of sample,
but only milligrams and fractions of milligrams are used for the final
confirmatory tests; the quantities decrease even to micrograms when
flame tests, spot tests, fluorescence phenomena, etc., are included among
the final tests.
By sample is always meant that part of the object of investigation,
which is being treated or observed at the particular stage of experimentation.
It is always assumed that it is truly representative of the whole object.
Because of the impossibility of giving a precise meaning to the term
micro and derived terms, EMleR suggested to indicate the scale of work
by stating the approximate mass of the sample used at the time. The
magnitude of mass is indicated with the use of the prefixes of the metric
system. Accordingly one may state that schemes of separation in inorganic
analysis are customarily started on a decigram or gram scale, whereas
preliminary tests and confirmatory tests are carried out on either the
centigram, milligram, or submilligram scales. If the total amount of sample
required or the amount of sample required for the separation of cations
(these two quantities are in the approximate ratio of two to one) is taken
for criterion, the terms semimicro, micro, and ultramicro may be interpreted
as refering to inorganic qualitative analysis with decigram, milligram,
and microgram samples, respectively.
The subdivision into inorganic and organic analysis complies with the
fact that the limitations of human capability call for specialization. Since
the objects of analysis are not necessarily labelled organic or inorganic,
a procedure must be followed that leads to the recognition of organic
substances in time so that the task may be turned over to a specialist if
4 Introduction

the already established facts and some additional simple tests, which are
recommended, do not suffice for a solution of the problem. The practical
analyst cannot adopt the rules of game of the typical school texts on
qualitative analysis, which restrict the search to a small selection of inorganic
cations and anions and may even require that the sample for analysis is
an aqueous solution of a certain concentration range.

Specificity and Sensitivity

So far, chemical reactions cannot be observed as such. The objects
of observation are the physical properties of the reacting system before,
during, and after the reaction. The reactions used for recognizing substances
produce either one or both of the two following effects:
a) Changes affecting the transmission of light by the appearance or
disappearance of phases (boundary lines): changes from homogeneity
to heterogeneity and vice versa. Into this category belong precipitation,
evolution of gas, dissolution, condensation, evaporation, transition to a
different state of aggregation, etc.
b) Distinct changes affecting the absorption or emission of light, i. e.,
changes of color, fluorescence, phosphorescence.
The practical value of an analytical test or procedure depends mainly
upon thp. nicety of discrimination, its selectivity, and upon the limiting
proportions. Tests (procedures, reagents) to which only one substance
responds are called specific. Tests which are given only by a small number
of substances are classified as selective (1241).
Sensitivity of a test (procedure, reagent) is to be described by listing
the limiting concentration, the limit of identification, and the limiting
proportions (850). The first two usually apply only in the absence of
interfering substances and become less favorable in the presence of such.
The limiting concentration (L. C.) is defined as the lowest concentration
of substance which always produces a positive test. Its dimension is mass
per volume (856, 859).
The limit of identification (L. I.) is the smallest absolute quantity
(mass or volume) which always gives a positive test (850).
The limiting proportions (L. P.) are the smallest mass ratios (substance
soughtfsubstance interfering) which still always permit to get a positive
test (1137).
If perceptibility, the limitation imposed by our senses, is not considered,
one may assume that limiting concentrations and limiting proportions
are determined by essentially chemical phenomena: rate of reaction, shift
of equilibrium, side reactions, establishment of metastable equilibria,
complex formation, catalysis, solubility, coprecipitation, etc. Thus it is
reasonable to expect that limiting concentrations and the important
limiting proportions will not depend upon the size of the reacting system.
 Specificity and Sensitivity 5

This assumption is supported by the available evidence, and it may be

expected that the limiting proportions (and limiting concentrations)
determined for gram samples will also hold on the microgram or nanogram
scales and vice versa. The limits of identification, of course, depend upon
the absolute size of the sample or the volume of solution taken for the test
and may be improved by reducing the scale of work. The relation L. 1. =
= L. C. X volume should hold within a considerable range.
As a rule, the limiting proportions are less favorable than the limits
of identification or limiting concentrations which apply only after the
substance has been isolated in a more or less pure state. It is important
to realize that negative' confirmatory tests must be interpreted with
reference to the sensitivity criteria. A negative test does not prove more
than that the amount of substance, if it should be present, is leRs than
indicated by the limit of identification or limiting proportions. If separations
have preceded the test, the criterion is provided by the operation with
the least favorable limiting proportion. Absence in an absolute sense
cannot be proven by an analytical test; at times, it may be concluded
from the history of the object under investigation.
Limiting concentrations, limits of identification, and limiting proportions
cannot be determined with high accuracy since the conditions which always
give a positive test and those which always lead to a negative result have
no sharp borderline. They are separated by a range of conditions, which
produces uncertainty as to the outcome: some trials give positive and some'
negative results (22, 150, 1580). This implies that the limiting conditions
"always" leading to a positive result must be established by a sufficient
number of experiments.
Suggestions for the brief notation of sensitivity criteria originated
with GILLIS (899), FLASCHKA (341), and MALISSA (895, 898). For this
book, the simple notation according to HAHN (962) has been adopted:
a) If a solid sample is being tested, the exponent (negative logarithm)
of the limit of identification given in gram, pL. 1., is listed in parenthesis:
(6.0).
b) If a solution is being tested, the parenthetical notation indicates
the subtraction pL. C. - pL. 1.: (4.0 - 7.0), whereby the L. C. is given
in gram per milliliter and the L. 1. in gram. Performance of the subtraction
leads to the logarithm of the volume of test solution given in milliliter:
0.001 ml = l,ul. The notation consequently reveals also the scale of
performance.
c) To indicate the limiting proportions, the interfering substances are
listed with the logarithm of their limiting mass per unit mass of substance
tested for:
(6.0; Fe, Co, Ni: 2; U: 1) means that l,ug of X may still be detected
in presence of 100,ug of Fe, Co, or Ni and in presence of 10,ug of U.
6 Introduction

(4.2 - 7.5; Ca: 2.5) indicates that O.03,ug X, L. C. = 60,ug X/ml,

may be detected in a drop of 0.5,ul in presence of 9,ug Ca.
HAHN suggests to give the figures with one decimal.

The Method of Qualitative Analysis

Two principal steps may be recognized when considering the performance
of a qualitative analysis (463): the preliminary treatment of the sample
material and the performance of the tests that identify the substances which
are present. This classification applies whether all tests are performed on
one portion of the sample, or various tests require separate portions of
the sample; in either instance, the material taken for analysis usually
must undergo some treatment before the decisive tests may be applied.
The final tests, which are known as conf1"rmatory tests, obviously must
be based upon the observation of some phenomenon that, under the
prevailing conditions, is produced only by the substance under consideration.
The obse~ed phenomenon may originate with a chemical, physical, or
biological property that is characteristic of the substance (145); the intensity
of the phenomenon is usually directly related to the mass of substance
so that a crude estimation of quantity becomes possible. As a rule, con-
firmatory tests as well as estimations of quantity may be performed quite
rapidly and without difficulty, once substances have been obtained in
a reasonably pure state.
The treatment of the sample material preliminary to the confirmatory
tests may involve transfers, disintegration, mixing, aliquot partitions,
preparation of solution(s), and involved procedures for the separation of
constituents. It may require nothing beyond transfer of the sample or
part of it to some apparatus for treatment or observation if either the
sample has a simple composition, or very specific and sensitive tests are
available for obtaining the desired knowledge. It should be considered
in this connection that the task may require merely a decision on absence
or presence of a certain substance.
Whenever little or nothing is known concerning the nature of the
sample, its composition is more or less complex, and sensitive as well as
specific tests are missing, then it becomes necessary that isolation precede
the identification of the various constituents. Isolation implies separation
and transfer in space, i. e., mechanical operations which require that
the substances to be separated have been collected in different phases.
Separable phases may exist to begin with. As a rule, however, they
must be established for the purpose of separation. Only three ways are
available for the creation of a phase: (1) the phase may be added mechanically;
(2) it may be caused to appear as a consequence of a change in temperature
or (and) pressure; and (3) it may be made to separate as the product of
a chemical reaction proceeding in the system. Obviously, one will try
 The Minimum Size of the Sample for Chemical Analysis 7
to obtain a new phase of such a kind that some of the substances to be
separated collect in it in their entirety, whereas the others remain al-
together outside of it. Noticeable deviations from this ideal are quite
common as a consequence of finite distribution ratios as well as adsorption
phenomena and are responsible for the difficulty of getting perfect
separations.
Apparatus and manipulative technique for the purely mechanical part
of the separations must depend upon the nature of the phases to be separated.
The three states of matter give five combinations which represent the
natural divisions for a discussion of mechanical separations: solid-solid,
solid-liquid, solid-gas, liquid-liquid, and liquid-gas. When suitable apparatus
and proper techniques are applied, the purely mechanical part of the
separations should not introduce errors or cause losses of a magnitude
that would render them significant in qualitative analysis.

The Minimum Size of the Sample for Chemical Analysis

The various species of matter are identified by their properties.
Obviously, the microchemist must be concerned about the effects of
diminishing size upon the properties of matter and the means for their
observation.
It follows from the atomistic concept of matter that its properties
should remain unchanged down to the size of the fundamental units.
These already contain the correct kind, ratio, and arrangement in space
of the building stones, that determine the particular behavior. Therefore,
the fundamental unit should represent the size of the smallest sample
suitable for the study of physical properties since the particular material
and the properties belonging to it cannot exist in a still smaller quantity
of matter. Masses larger than the fundamental unit are simply accumulations
of the-latter and should not present any novel features excepting incidental
phenomena such as size and shape, that can be assumed by any material.
The size of the fundamental unit that still permits qualitative identifica-
tion on the basis of the typical properties depends greatly upon the type
of material. In the instance of a coarse conglomerate, it may take a piece
of several kilograms to reveal the essential characteristics; with crystalline
rocks, a fraction of a gram may be required. The fundamental unit of
mixtures must represent a reasonably close approximation of the composition
of the whole (13, 14). Of suspensions, emulsions, colloidal sols, and
molecular dispersions, a volume will be required which includes a fairly
characteristic percentage of the dispersed matter. In the instance of
chemical specimens (substances) the fundamental unit may be identified
with the molecule or the unit cube of a crystal lattice.
When dealing with chemical specimens (solids and liquids), the funda-
mental unit representing the critical limit to identity and identification
8 Introduction

can hardly be imagined to exceed 1000 atoms corresponding to a mass of

1000 X 1.7 X 10-24 W gram or 3.4 X 10-19 gram if an average molecular
weight is assumed such as W = 200.
After remembering that the energy content of one molecule or one
fundamental unit need not represent the temperature of the bulk and that
the collision of light with just one fundamental unit may not lead to the
commonly accepted value for the refraction, one may be inclined to take
into account the essentially statistical nature of the phenomena exhibited
by matter. Accordingly one may adopt a safety factor of 1000 and expect
that this number of fundamental units wQuld give a satisfactorily close
approach to the average effects observed with large quantities of matter
and thus permit identification by observation of the characteristic behavior
with about 10-15 to 10-16 g (0.003 pg) of substance (145).
There is no doubt that such and smaller quantities will suffice for
identification if the phenomena are striking and originate in the atom
or in the molecule. Thus it has been assumed long ago (1149) that quanti-
tative determinations via the radiation count are possible with 10-15 g
of highly radioactive elements, which implies that one hundredth of this
amount or less (10-17 to 10-18 g) should suffice for qualitative identification.
EMICH (150) demonstrated that 10-11i g hydrogen are sufficient to produce
the line H(X, and PANETH and PETERS (652, 1201) devised gasvolumetric
apparatus that permitted to identify 2 X 10-14 g helium by means of the
glow discharge. With the use of the microscope, fluorescence may be
studied on tiny objects; thus taking hemispherical drops of l-,um (,u)
diameter, one should be able to recognize 2.5 X 10-16 g fluorescein in
0.1 per cent solution.
It appears safe to assume that general theoretical reasoning permits
reducing the scale of chemical experimentation to 0.001 pg = 10-11i g of
mass and possibly less. One may expect with quantities down to this
limit that matter still has the customarily accepted values for density
and specific volume, refraction, optical rotation, absorption of radiation,
magnetic susceptibility, conductivity, electrode potential, etc. One may
also expect that the customarily known equilibria will establish themselves
in reaction mixtures down to this size. This has important practical
consequences.
In most microchemical work, one is very far from approaching the
critical size limit. Therefore one may feel certain that there will be no
fundamental change in physical properties and in chemical equilibria
when the available quantity shrinks to milligrams, micrograms, nanograms,
or picograms. Since equilibria are determined, aside from temperature
and pressure, solely by the concentration of the reactants, working directions
may be simply transposed from anyone scale to any other by multiplying
 Properties Affected by the Size of the System 9

temperatures, and pressures without change. Any test or scheme of separa-

tion may be carried out with micrograms exactly as described for the gram
scale by just reducing all weights and volumes to one millionth, by taking
micrograms instead of grams, nanograms instead of milligrams, and nano-
liters in place of milliliters. Experiences gained on any scale hold for
every other scale of work; the utilization depends merely upon application
of proper techniques in observation and experimentation. The "chemistry"
and the theory of qualitative analysis may be applied as found in the
general literature (13-82).
It has been demonstrated by RACHELE (424) that the critical size limit
may be closely approached by the suitable performance of common test
tube tests. Droplets of test solution and reagent solution were deposited
by means of a micromanipulator in a film of paraffin oil that had been
deposited upon the underside of the top plate (cover slip) of a moist chamber.
The operation was performed in the field of a high-power microscope so
that it was possible to measure the diameter of the spherical drop of test
solution and to ascertain that both drops were clear. Drops of different
size merged spontaneously if deposited close to each other, whereupon
the separation of a precipitate or the development of a coloration was
observed. Thus the limit of identification was reached with 10-14 g of
barium ion supplied in a drop of 1 pI (12 ftm diameter) of 0.001 per cent
solution. With magnifications of about 400 diameters, a fine crystalline
precipitate could be observed with transmitted light as well as with dark-
field illumination. The Prussian blue test was reliably obtained with
0.4 pg of ferric ion supplied by drops of 4-ftm diameter (0.04 pI volume)
of 1 per cent solution. In a like manner, ANNA G. LOSCALZO (1261) could
identify 0.02 pg silver as silver chloride, 0.2 pg cobalt as cobalt-mercuric
thiocyanate, and 0.5 pg iron as ferric thiocyanate.
In all instances it was felt that the phenomena were probably obtained
with still smaller quantities, but that it became impossible to observe
them with the means employed.

Properties Affected by the Size of the System

It is to be expected that all properties that do not reside in the
fundamental unit may vary with the amount of matter. Thus all statistically
established equilibria must be affected even though our insufficiently
precise methods of observation fail to reveal the differences.
Most of all one may expect to find deviations from the customary
experience with properties or equilibria acting on or across phase boundaries.
This is to be expected because the relative amount of surface must increase
when mass and volume diminish.
It may be readily demonstrated that the specific surface area, defined
as surface area per unit volume, A/v, must be inversely proportional to
10 Introduction

the cube root of the volume (or mass) provided that the shape and its
proportions are retained. In the instances of the tetrahedron, the cube,
and the sphere, the equations are A/v = 7.2/vl/a (tetrahedron), A/v = 6/v1 /.
(cube), and A/v = 4.8!v 1/. (sphere).
The decrease of mass and volume and the corresponding increase of
specific surface area affect the surface tension and as a consequence the
vapor tension, solubility, melting point, and boiling point. The changes
are insignificant until the particle diameters drop below l/-lm, i. e., the
volume below 10-12 ml and the mass below 10-11 to 10-12 g. Noticeable
increases of vapor tension and solubility are expected with decidedly
smaller amounts of matter.
In the instance of barium sulfate, DUNDON (22) estimates a nearly
hundredfold increase of the solubility when the diameter of the pa,rticles
dropped to 0.4 /-lm corresponding to a mass of a cube-shaped particle of
3 X 10-13 g. The solubility should have been still higher in the experiments
of RACHELE (424) where only about 2 X 10-14 g of barium sulfate were
available at the limit of identification. The published photomicrograph
indicates at least four large particles, and if five of them were present,
their diameter could not have exceeded O.I/-lm. RACHELE, however,
assumes that these particles were aggregates and that a large number
of still finer particles collected at the interface with the paraffin oil. The
findings are not necessarily in contradiction. Aside from the fact that
there is no definite assurance concerning the absence of finer particles
in the experiments of DUNDON, excess of sulfate ion in RACHELE'S tests
may have more than compensated for increased solubility. Finally, some-
where near the critical size limit, there must be some reversal in the
solubility-particle diameter relation to permit the formation and growth
of nuclei.
In addition to these effects, a large specific surface area may have a
noticeable influence upon the composition of matter. In the instance of
a palpable collection of matter (crystalline or liquid), the mass of the
substances adsorbed on the surface may be alltogether negligible when
compared ~th the bulk. The ratio of adsorbed foreign matter to the
mass of the bulk must increase, however, when the latter diminishes.
If the mass decreases to include just a small number of unit cubes or mole-
cules, one may very well imagine that the amount of adsorbed matter
could assume such a proportion as to change the identity of the accumula-
tion. If no precautions are taken, adsorption might materially affect the
performance of chemical separations on a very small scale. Fortunately,
however, adsorbed matter is, as a rule, readily removed by washing or
displacement.
So far, microchemical experimentation has produced no evidence of
a change of chemical or physical behavior as compared with that customarily
 Apparatus, 'l'echmque, and ):Scale of Work 11

observed. This is probably due to the fact that most microchemical work
has been performed with quantities far above the critical size limit. It
means that our chemical knowledge applies without change down to at
least the picogram scale and that any deviations observed are probably
due to faulty technique. On the other hand, it will be wise to be on the
lookout for changes in the behavior of matter when entering the critical
size range of less than 10-14 g = 0.01 pg.

Apparatus, Technique, and Scale of Work

Whereas it may be assumed that the properties of matter and equilibria
remain essentially unchanged throughout the whole size range of applied
micro analysis, it is quite obvious that the technique of experimentation
must be varied depending upon the amount of matter to be dealt with.
Apparatus and procedures suitable for work with milligrams of material
may become quite impractical when adopted for use on either the gram
or the microgram scale. When the inescapable necessity of adjusting
apparatus and technique to the amount of matter is compared with the
universal validity of our chemical knowledge over the whole field of applied
micro experimentation, it appears only proper to substitute terms like
microtechnique of chemical experimentation and microtechnique of chemical
analysis for the established but misleading composites microchemistry and
micro analysis. The expression microchemistry might be more properly
reserved for the behavior of matter below the critical size range, where
the customary statistical equilibria are no longer attained.
Whatever the scale of work may be, proper apparatus and technique
will give efficiency. In other words, they will assure attainment of the
purpose with a minimum expenditure of skill, effort, and time. In addition
to these requirements, the micro analyst must be concerned with the
economical use of the sample material. In this respect it may be desirable
to have the possibility of recovering the material under investigation in
some form, but even this is not enough. It is important to keep this material
always confined in a known small space so that its manifestations give
an intensity to the phenomena assuring their reliable recognition by the
striking contrast with the behavior of the surroundings. Accordingly
apparatus and technique must be designed to retain the material under
investigation in the smallest practicable and, if possible, approximately
equidimensional space. The success of this reasoning is demonstrated by
the various techniques for the performance of slide tests in very small
areas: on the plateau of 0.2-mm diameter and 0.03-mm2 area with a L. I.
of 10 ng (p. 2l7); on the Brenneis electrode of 0.025-mm diameter and
0.0005-mm2 area with a L. 1. of about 1 ng; and in the confined field of
the electron microscope of 0.0001-mm2 area with a L.1. of 50 pg (934,
964).
12 Introduction

It is in keeping with this same principle that centrifugal force is widely

used for collecting matter under investigation into a small space. Filtration
is usually replaced by decantation; this avoids exposure of liquids to the
large surface of filter mats and does away with the often quite difficu1t
task of separating the collected solid from the filter material. Consequently,
centrifuge tubes of 3-ml to l-,ul capacity are widely used and are standard
equipment for work on the centigram, milligram, submilligram, and micro-
gram scales.
Of particular interest are capacity and shape of that part of the apparatus,
in which the matter under investigation is collected. The small volume
assures high concentration and corresponding intensity of action; approach
of the occupied space to the shape of a sphere reduces the area of contact
with surrounding matter and the extent of undesirable consequences such
as adhesion, exchange of matter, and adsorption. Exposure to air would
invite vaporization of volatile matter, absorption of water vapor and
laboratory fumes, and contamination by dust. Mechanical collection
would become difficult and inefficient if the area of contact with the
apparatus were large. Extensive contact with the wall of the apparatus
would also increase the amount of contamination by solvent action and
magnify losses due to adsorption and ion exchange.
A suitable approach to the spherical shape is obtained by filling a circular
cylinder to a height equal to its diameter, D. The sphere and such a cylinder
give the same specific surface area, 6/D, which may be readily tested
with the use of the equations for surface area and volume of the sphere,
:It D2 and :It D3/6, and of the equidimensional cylinder, l.5:1t D2 and :It V/4.

For obvious reasons, cylindrical apparatus will be given a height exceeding

its diameter, but the latter should be such that the contents will approach
the shape of the equidimensional cylinder. To obtain this condition, it
is necessary to select the bore of the cylindrical apparatus according to
the scale of work. To this end, Table I lists the diameters of the equi-
dimensional cylinders as a function of their volumes. The working scales
are named after the approximate mass of solid sample available. The
corresponding volumes in the second column are the volumes of solution
occurring in the course of a qualitative analysis. If one gram of sample
is available, the largest volume of up to 300 ml may occur during the
precipitation with hydrogen sulfide, and confirmatory tests may be
performed with I-ml aliquots. The corresponding cylindrical apparatus
would be beakers of 7.5-cm diameter and test tubes of 11 mm bore. The
table shows that D becomes inconveniently small for volumes of 1 nl =
= 10-6 ml corresponding to test-tube tests on the microgram scale.
Manipulation in the apparatus is materially facilitated by giving it
the shape of a cone which may be considered to represent an infinite number
of circular cylinders of zero height and continuously decreasing diameter.
 Apparatus, Technique, and Scale of Work 13

Table I. Dimensions of Apparatus as a Function of Its Capacity

Diameter (and Height)

Height of Circular
Volume of Equidimensional
Scale of Work Cone of 20 Degrees
v Circular Cylinder 3.13 v1/•
1.08 v1/•

1300 ml 7,4 em 21 em
100 ml 5.1 em 14.5 em
10 ml 2,4 em 6.7 em
Gram Scale
3 ml 1.6 em 4.5 em

1 ml 11 mm 31 mm
Centigram 0.3 ml 7.4mm 21 mm
Scale 0.1 ml 5 mm 15mm
0.03 ml 3,4mm 10mm
Milligram
Scale
10 ,ul 2,4 mm 6.8mm
3 ,ul 1.6 mm 4.5mm
1 ,ul 1.1 mm 3.1 mm

1
0.3,u1 0.74mm 2.1 mm
Microgram
Scale 100 nl 0.5 mm 1.5 mm
1 nl 0.11 mm 0.31 mm
Nanogram { 0.3 nl 0.074 mm 0.21 mm
Scale I 1 pI 0.011 mm 0.03 mm

~B F-=::
U
, 0 I
I
I
I
I

The cone has also the same advantage as a staircase of cylinders of

diminishing diameter: that a separating phase of small volume may occupy
an approximately equidimensional space, while this condition also holds
for the whole bulk of the system. This simultaneity cannot be obtained
with cylindrical apparatus truncated at a right angle; but it could be
rather closely approached with a cone having an angle of 60 degrees.
The taper of centrifuge tubes usually has an angle of less than 60 degrees,
which has the advantage of decreasing the specific surface area,
3 (1 + sin IX/2}/r cos IX/2 = 3.05 (1 + sin IX/2}/v1/• (tg IX/2)'/· cos IX/2
which for a given volume v reaches infinity for IX = 0° and IX = 180°
and becomes a minimum of 6.11 v-1/. for IX = 35 The relative surface
0 .

varies from 6.48 v-1/. for 20° to 6.35 v-1/. for 60°, and the angle of the cone
may be suitably chosen between these limits. Of course, the specific
 14 Introduction

surface area will increase directly with v -1/. when the volume of the system
becomes smaller.
The right-hand column of Table I lists the heights to which a circular
cone having an angle of 20 degrees must be filled to represent the listed
volumes v. The computed dimensions of the table show that, in general,
the shapes of centrifuge tubes have been selected with good sense. A taper
of 20- to 25-mm length fitted to a cylindrical tube of 4- to 6-mm bore is
appropriate for the needs of work on the milligram scale. For ease of
manipulation, especially cleaning, the tip of centrifuge cones is usually
rounded off as indicated in Fig. 22. This is no longer feasible for cones
of the microgram or nanogram scales, and consequently one may prefer
to give their tapers somewhat wider angles.
Very obvious general observations concerning type of apparatus may
be made with regard to the handling of substance of high vapor tension.
When dealing with small amounts of volatile matter, loss will result in a
short time if evaporation is not prevented; volatile solvents may vaporize
so fast from small drops of solutions that working becomes impossible.
Naturally, the vapor pressure that may be tolerated depends upon the
scale of work. In the instance of aqueous solutions, the volatility of the
solvent gives little concern on the milligram scale; only when slide tests
are performed in a dry atmosphere, the rapid drying up of test drops may
cause some annoyance. On the microgram scale, where the volumes range
from 1 to 300 nl, experimentation with aqueous solutions becomes impossible
if evaporation is not prevented.
If the material under investigation is volatile, work on a small scale
requires that the volatile material is enclosed in a restricted space. The
size of this space may be readily estimated by means of the gas law. One
unit of volume of a liquid or solid substance of molecular weight M,
density d, and vapor tension p mm will occupy 62000 d Tip M units of
volume at T Kelvin or 18· 106 d/P20 M units of volume at 20° C. If not
more than one per cent shall be lost by vaporization at 20° C, l,ul (I cubic
millimeter) of the liquid or solid substance must be confined into a
space of not more than 180000 d/P20 M,ul. The volumes are for
carbon disulfide, chloroform............ 1O,u1
ethyl alcohol ......................... 70,ul
acetic acid ........................... 0.25 ml
water . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0.57 ml
iodine, camphor, valerie acid........... 3 ml
naphthalene .......................... 30 ml
The corresponding volume for water at boiling temperature is 16 ,ul, and
it follows that water and dilute aqueous solutions will not show a decrease
of volume when heated to 100° C if they are sealed in a container and the
 Apparatus, Technique, and Scale of Work 15

gas space is not more than 16 times of the volume of the liquid (uniform
capillary of 17-mm length and I-mm height of liquid column).
The evaporation of solutions may be prevented also by maintaining
an atmosphere that is saturated with the vapor of the solvent. For work
under the microscope, this is simply accomplished by using a moist chamber,
Fig. 60, and entirely satisfactory on the microgram scale, i. e., solution
volumes down to 1 nl = 0.0011'1. RACHELE (424) found, however, that
drops of a few micrometers in diameter (0.01 to 0.1 pI or 10-7 to 10-81'1)
evaporated within a few seconds even when the moist chamber was com-
pletely sealed and contained air so saturated with water vapor that the
inside surfaces of the chamber were covered with condensate. Obviously
the equilibrium with the large drops of the condensate was still altogether
insufficient to prevent the rapid evaporation of the far smaller drops of
dilute salt solutions. The phenomenon was suppressed by increasing the
electrolyte concentration of the small drops to 1 per cent or adding 0.6 to
2.5 per cent of glycerol, which gave a sufficient reduction of the vapor
pressure of the small drops to establish equilibrium with the atmosphere
of the cell. Interesting is the fact that this precaution had to be retained
even when the small drops were suspended inside a film of paraffin oil
mounted in the moist chamber.
The feature of micromanipulation that is most apt to attract the
attention of the casual observer is connected with the increase of surface
forces on the small scale. Pipets and siphons operate without use of suction;
containers may be inverted or even dropped without spilling of contents; etc.
In general, the capillary attraction frequently facilitates manipulation,
and the tendency toward the spherical shape dictated by surface tension
is an aid in keeping matter confined into a small space. The relatively
large fraction of liquid left behind when emptying glass capillary
tubing is mostly a consequence of the large specific surface area (= 4jD)
of the cylinder surface, which increases proportionally with the shrinkage
of the diameter D. The amount of residual liquid approaches a minimum
when the outflow is slow, a fact that is well established for the customary
pipets and burets of the analytical laboratory. Even with very slow
delivery, the hold-up in glass capillary tubing may become very large
and cause losses that could become inconvenient even for qualitative
work. Experiments of LOSCALZO (437) indicate that the fraction of residual
liquid over total liquid reaches a maximum of about 0.25 when the inner
diameter of the tube is between 0.2 and 0.3 mm. When the meniscus
travelled between 0.005 to 0.08 mm per second, the following percentages
of residual liquid were determined in experiments with water: diameter
of bore, D = 0.26 mm, 24% residue; D = 0.22 mm, 24%; D = 0.16 mm,
20%; D = 0.14 mm, 13%; D = 0.12 mm, 12%; and D = 0.11 mm, 10%.
The maximum for a bore of about 0.25 mm may be explained by a decrease
16 Introduction

of the thickness of the residual film in narrower capillaries. It should be

obvious that, on emptying, the amount of residual liquid will be of similar
magnitude with small apparatus of any shape; one might expect it to
become most unfavorable when collecting liquid that has spread on a
plane surface.
The above statements apply to clean glass surfaces. The amount of
residual liquid as well as that of solid particles suspended in it will greatly
depend upon the nature of the surface of the container and the presence
or absence of surface active substances in the liquid. The technique of
using apparatus must be adjusted depending upon the behavior of liquids
upon its surface. Vice versa, the surface of apparatus may have to be
given a special treatment to assure the success of the technique to be
applied. So far, relatively little use has been made of the advantages
offered by surfaces that repel liquids, but their importance will probably
increase with the refinement of technique. The spreading of drops on the
surface of clean glass slides may be merely annoying when working on the
milligram scale; it must be prevented by the use of a repelling barrier
when working with micrograms, p. 201; and it would lead to complete
loss of material on a still smaller scale of work. Since very small apparatus
(cylindrical, conical, etc.) would not admit liquid that is repelled by its
surface, one will have to work with more or less spherical drops that either
adhere to a repelling surface or float in an inert medium, the techniques
tried by RACHELE (424) on the subpicogram scale.

Observation of Properties
Common sense indicates that it will be the easier to observe a' property,
the larger the object; it follows that all available material should be collected
into one body to give as much bulk as can be obtained for the observation.
All phases of microtechnique are designed to serve this end. Spot tests
and fiber tests are performed so that the colored matter is concentrated
into a compact body. Precipitations on the slide are made to take place
under conditions favoring the separation of few large crystals, which
facilitates the observation of their shape as well as optical and other
properties. The principle may be quite rigorously stated for the observation
of the absorption of radiation and color.
The law of BOUGUER-LAMBERT-BEER gives the absorbance A as a
function of absorptivity a, breadth of layer b, and concentration of the
absorbing matter c.
A =abc.
Since concentration is mass m over volume v and the volume may be
expressed as the product of breadth times cross-section area iX, it follows that
A =ma/iX.
 Observation of Properties 17

For a given test and method of observation, there will be a smallest

absorbance effect Ao that still may be recognized with certainty. The
amount of matter, m o, corresponding to that intensity of the phenomenon
is given by
mo = IX Ao/a
and represents the limit of identification, L. I. Obviously the test may
be refined, i. e., the L. I. reduced in direct proportion to the reduction
of the cross-section area occupied by the absorbing matter. Use of this
relation is made wherever absorption of radiation serves the identification:
spot tests, bead tests, coloriscopic capillary, microspectrophotometry,
micropolarimetry, and fiber tests.
The equation may also be used for an estimation of the sensitivity of
tests based upon the perception of color or characteristic absorption. If
two color tests use the same conditions of observation (same cross-section
area), one may expect that the limits of identification will be in a ratio
inverse to that of the respective absorptivities
(m o).,: (m o)" = a,,: am
since the limiting absorbances Ao will be similar. Obviously, this will
hold for all types of radiation that obey BEER'S law, and it will be the more
nearly correct, the smaller the frequency range of the radiation.
Of special importance for the microscopic observation of color is the
fact that absorbance does not vary with the intensity of the light. To
obtain a bright image of an absorbing object, it is necessary to increase
the intensity of the illumination in proportion to the square of the lineal
magnification, but this does not affect the contrast of the color of object
and surroundings. This makes it possible to collect just enough coloring
matter into an area of microscopic dimensions to obtain a perceptible
color effect and then to use the microscope or microprojection to get a
greatly enlarged image of that area without diminishing the color contrast.
It is assumed that the intensity of the illumination can be increased without
changing the composition of the radiation.
In reducing the scale of working, apparently unsurmountable difficulties
have been met in the observation of phenomena. The equation above
indicates that a certain minimum mass of colored substance has to act
on the beam of light, i. e., is required per unit of cross-section area if the
color shall become visible. As a consequence it is found that faint colorations
which are readily visible in thick layers of matter cannot be. seen in thin
sheets or small droplets of the same. If dye indicators shall be used, they
must be added in much higher concentrations than customarily
tolerated (437). Furthermore, when working with droplets of 7- to l4-pm
diameter floating in a film of paraffin oil, RACHELE (424) found that the
observation of heterogeneity as well as that of color was hampered by
Benedetti·Pichler, Identification 2
18 Introduction

refractive phenomena. With transmitted light, the drops were transparent

but surrounded by a heavy dark rim which closed up with very small
drops so that these appeared as black disks. When changing to darkfield
illumination, drops smaller than 7 p.m in diameter became bright disks
and the recognition of particles of a precipitate in the interior became
uncertain. Removing this difficulty would require the discovery of an
otherwise suitable liquid with a refractive index close to that of water.

Selection of Procedure
An ideal collection of microtechniques would offer apparatus and
manipulative procedures for the efficient performance of any desired
operation and on any desired scale. At this time, only the framework of
the structure is available, and the experimenter must generally be on the
lookout for possible improvements of the described technique. Deviations
from tested techniques should not be made, however, without careful
consideration of all possible consequences.
Usually there are good reasons for every detail of the directions. Thus,
whereas the order in which reagents are taken into a capillary may be
immaterial from the point of view of the chemical equilibrium, it may
have been chosen to prevent contamination of a particular reagent, to
avoid side reactions, to facilitate or to prevent mixing, or merely to assure
proper rinsing of the tip before sealing it.
(',oncerning the choice of method, there are usually various chemical
approaches for the solution of an analytical task, and they may require
different operations. Depending upon the scale of work, certain operations
may be awkward and inefficient, and it will be wise to avoid chemical
procedures that require such. In general, it will be best to keep away also
from schemes which need a large amount of manipulation that has to be
done by a skilled or experienced operator and to favor procedures using
operations that proceed spontaneously without much manipulation and
supervision.
Evaporation, distillation, sublimation, and diffusion consume much
time on a large scale and are therefore avoided in qualitative analysis.
When working on a small scale, they require little time and become
eminently practical, for they proceed spontaneously with little supervision.
On the centigram scale, it may still be practical to neutralize a solution
by adding small portions of acid and base, but the "dropwise" addition
of reagent becomes rather inconvenient already on the milligram scale
that would require "drops" of 0.05 pI volume if a macroprocedure is to
be faithfully copied. On the small scale, it is far more efficient to adjust
pH by exposing the solution to an atmosphere containing a known concentra-
tion of hydrogen chloride or ammonia gas, obtained by equilibrating air
with a large volume hydrochloric acid or ammonia solution of known
 Selection of Procedure 19

concentration (immersion of a test drop into the air space of the reagent
bottle will often suffice). If this is not practical, it is usually possible to
evaporate the solution to dryness after making certain that it contains
an excess of volatile acid or base and to dissolve the residue, possibly
after ignition, in a medium of the desired composition and pH. This
latter procedure has the additional advantage of preventing the accumulation
of salts in the solution.
Careful adjustment of equilibria, i. e . , control of concentrations and
measuring of reagents, is essential regardless of the scale of work. Special
attention is required when working on an unfamiliar scale since the intuitive
choice of correct proportions becomes unreliable under the conditions.
The measuring of test substance, solutions, and reagents is not only
necessary for the proper adjustment of the conditions, it is also required
for the estimation of the quantities of substances found.
The obvious choice is the operationally simplest procedure which has
the required sensitivity. From a purely technical view point, the simplest
procedure may be the one with the smallest numbers of transfers. Especially
the transfer of solids is fraught with the danger of loss. As far as the
sensitivity is concerned, it is necessary to recall that it may be determined
by some step in the separation preceding the final test. Of course, only
an approximate estimate may be possible at this time since .the limiting
proportions depend upon the composition, essential facts of which may
still be unknown.
 Part I

Technique of Observation and Manipulation

Use of Optical Aids

The Microscope
The microscope is what its name implies, an instrument for viewing
small things. It usually has the following parts:
I. the Stand consisting of,
Base, usually horseshoe-shaped, which supports the
Pillar or Column, often joined by a
Tilting Mechanism to the
Arm which carries the substage, the stage, and the body tube so that
the distance between the two may by changed by rack-and-pinion (coarse
adjustment) and possibly also micrometer motion (fine adjustment);
II. the Body tube containing the optical system that produces the
enlarged image of the object and consequently represents the microscope
proper consisting of:
Objective giving an enlarged image of the object in the plane of the field
diaphragm of the eyepiece,
Eyepiece which further enlarges the image given by the objective, and
Drawtube permitting variation of the distance between eyepiece and
objective;
III. the Stage supporting the object under investigation and needed
auxiliary devices; and
IV. the Substage which usually may be moved mechanically and
carries the apparatus for illumination with transmitted light, which may
consist of:
Mirror with one plane and one concave side,
Condenser, a single lens or a lens combination, and
Diaphragm.
The formation of the image is purely schematically illustrated since
presentation with correct proportions leads to fine detail which is difficult
to discern.
 The Microscope 21

In the left upper corner of Fig. 1 is indicated that the light emerging
from the eyepiece of any optical apparatus forms a double cone, of which
the cross section of narrowest diameter is known as RAMSDEN disk. Every
observer automatically places his eye so that the iris of the eye is around
the RAMSDEN disk, because failure to do this has the unpleasant effect
that part of the light rays are cut off by the iris and consequently only

SIJge

'c: l1itw/ imJge

Fig. 1. Action of the Compound Microscope

pa,rt of the image is received by the retina and visible. This annoying
phenomenon may be experienced with high-power eyepieces having the
RAMSDEN disk close to the eye lens so that the eye touches the latter
before the iris arrives in the desired position; it also may happen with
auxiliary devices (color filter or cap analyzer) on top of the eyepiece,
which prevent the eye from attaining the proper position. With properly
designed eyepieces, the eye is held at a definite distance from the eye Jens,
which not only fixes the focal length of the rigid combination of eye lens
and eye, but also determines the distance of the receiving screen (retina)
22 Use of Optical Aids

from the eye lens. Since focal length and distance of the receiving screen
are fixed, also the location of the object is determined: it is the conjugated
image plane in the eyepiece half way between eye lens and field lens and
indicated in the upper right corner of Fig. l.
The location of this image plane will vary somewhat depending upon
the characteristics of the eye. This does not impair the action of a field
diaphragm in the eyepiece, set into an image plane corresponding to a
"normal" eye. On the other hand, whenever the image plane is accurately
predetermined by placing a scale, cross hairs, or other rulings in the eyepiece,
then the eye lens is mounted in the eyepiece so that the distance between
eye lens and rulings can be changed until a sharp image of the latter is
received upon the retina.
Obviously, the reasoning changes very little when the image given by
the eye lens is received upon a screen or upon a photographic plate instead
of upon the retina. In either instance, the action of the eye is omitted
and the magnification may be greatly varied by changing the distance
between the eye lens and the receiving surface. Furthermore, in micro-
projection and photomicrography the final image is received on a plane
surface, whereas the retina is curved. Especially in photomicrography it
may become desirable to use specially corrected objectives and eyepieces
for the flattening of the final image: Homals of 'ZEISS (88).
When rulings in the image plane of the eyepiece are once focused upon
the retina, projection screen, or photographic plate, it is obvious that
anything appearing in the image plane of the eyepiece will be portrayed
upon the receiving screen (retina, plate). If the lenses of the objective
together with the field lens of the eyepiece produce an image of the object
in the image plane of the eyepiece, a conjugated image of the object will
also appear upon the screen. This will happen during focusing when a
distance between object and objective has been reached which places the
inverted image of the object into the image plane of the eyepiece.
To prevent undesirable contact of the objective with the object, the
latter is first closely approached with the objective while watching from
the side. Then the eye is brought to the eyepiece or the screen is observed
while the distance between object and objective is increased. An infinitely
large image is formed at infinity when the object arrives at the principal
focus of the objective combination. After this, the image is rapidly pulled
in from infinity into the body tube while the distance from object to
objective is increased to twice the focal length. Especially with high-power
objectives of short focal length, the image must move large distances with
very high speed while the object-objective distance grows by fractions
of a millimeter, and the image may appear in the conjugated image plane
of the eyepiece and on the retina or screen for such a short time that it
cannot be perceived.
 The Microscope 23

When the object is in focus, the plane of the object, the plane of the
rulings in the eyepiece, and the plane in which the final image appears
(retina, screen, plate) become conjugated image planes, i. e., any object
in one of these planes has an image in all others, provided that light proceeds
in both directions. If the object is located outside the principal focus of
the condenser, there must be another conjugated image plane on the
other side of the condenser lenses (Fig. 1 assumes that it is the plane of
the iris diaphragm of the illuminating lamp), and any objects or structures
in this plane will give images in the object plane and consequently in the
image plane of the eyepiece and on the retina, screen, or photographic plate.
Provided that the lenses of the microscope condenser are highly corrected,
this makes it possible to include into the final image of the small object
also reduced likenesses of relatively large objects (pointers, scales, rulings)
placed into the conjugated image plane on the far side of the microscope
condenser (88).
Fig. 1 shows that, because of double inversion, the image received on
retina, screen, or photographic plate is right side up. For this reason,
the image received upon the retina is interpreted corresponding to an
inverted virtual image.
As an aid to better understanding, one may copy Fig. 1 on a large
sheet of paper by actual construction of the images using the rules of
elementary optics and neglecting the effect of the field lens of the eyepiece.
Magnification. The total magnification of a microscope is the product
of the individual magnification8 of the objective aM the eyepiece being used;
it is the ratio of the diameter of the virtual image (Fig. 1), imagined 250 mm
in front of the eye of the observer, divided by the diameter of the object.
The magnification of objectives and eyepieces is usually indicated on
the mounting. If the focal length is given, the magnification may be
estimated as follows,
tube length
magnification of objective = focal length-of obJective'
250mm
magnification of eyepiece = ._-
focal length of eyepiece'
Most manufacturers use a standard tube length of 160 mm. If the
microscope is provided with a draw tube, usually permitting a variation
of the tube length from 140 mm to 180 mm, the tube length may be read
off a scale engraved upon the outside of the draw tube.
Total magnifications of less than 100 are considered "low", and of
more than 500, "high"; magnifications of 100 to 500 are called "medium".
Micro analysis very rarely requires magnifications above 250, and, con-
sequently, the following suggestions for the use and care of the microscope
consider mostly low-power microscopy.
24 Use of Optical Aids

Eyepiece. The eyepieces shown in Fig. 1 are of the Huygenian type

suited for use with common achromatic objectives. Micrometer eyepieces
with adjustable eye lens are preferable for most chemical work.
Eyepieces are readily exchanged by slipping them in and out of the
draw tube or the body tube of the microscope. The eyepiece remains
safely in the microscope as long as the instrument is not inverted. Eyepieces
which are not in use should be kept in the rack provided in every micro-
scope case. One eyepiece must always remain in the tube of the microscope
to keep the dust out. .
Objective. The quality of the final image depends largely upon the
perfection of the objective, and consequently it is imperative to handle
objectives with care and to keep them clean at all times. When not attached
to the microscope, they must be kept in the containers provided by the
manufacturer.
Dry objectives with magnifications from 2 to 20 will suffice for most
chemical work. They usually have severallenses mounted in a metal tube with
a screw thread for attaching to the body tube or objective changing devices.
Objective Changers. The component parts of objective changers, such
as the clamp changing device (clutch) or the more precisely working sliding
objective changer, may be attached to objectives and body tubes of standard
make. Use of a revolving nose piece has the disadvantage that all attached
objectives are continuously exposed to the objectionable agents frequently
present in chemical work (fumes, heat).
Aperture. The angle (hatched in Fig. 2b) between the most divergent
rays used for image formation is called the angular aperture, A. A., of the
objective. The numerical aperture,
N. A. = n· sin (A. A.j2)
is computed from the angular aperture and the refractive index n of the
medium occupying the space between the preparation and the front lens
of the objective.
Aperture determines the amount of light which an objective may grab
and consequently the brightness of the final image. It al!'lo determines
the power of resolution of the objective. Obviously, a small fraction of
the light originating from a gross feature of the object will !'Iuffice to give
an image of it. It is necessary, however, to employ the widest possible
cone of rays emanating from a minor heterogeneity in the object in the
hope of revealing its existence.
Front Lenses are those lenses of objective and condenser whkh are
closest to the object on the stage. Their exposed surfaces, which are always
plane, must be carefully guarded !'Iince they are most 1 kely to become
soiled, corroded, or mechanically damaged by contact with or proximity
of the object.
 The Microscope 25

Inspection of Fig. 2 shows that front lenses of objectives having low

magnification must have relatively large diameters to get a reasonable
aperture. This provides a simple means for recognizing low-power objectives.
Working Distance of an objective is the distance in millimeters between
the front lens and the object when the latter is in focus. The working
distance of a low-power objective is somewhat less than its focal length;
that of a high-power objective, only a fraction of the focal length. Obviously,
long working distances for the investigation of the interior of thick objects
or the convenient operation of tools between object and front lens can
be obtained only with low-power objectives.

Fig. 2. Use of Plane and Concave Mirrors. The diameters of the preparations and
the angular apertures of the objectives are indicated. The shaded areas show the
bundles of light that are used for the formation of the image

Field Diaphragms are used to limit the visible area without affecting
its brightness. They are placed in the conjugated image planes to obtain
sharp images of their edges giving a sharp circular boundary to the micro-
scopic field.
Aperture Diaphragms are placed into a focal plane of their respective
lens system so that they uniformly restrict the amount of light used in
forming an image without affecting the size of the field.
The aperture diaphragm in the lower focal plane of the condenser (Fig. 1)
serves to regulate the brightness of the image. A conjugated image of this
diaphragm forms in the upper focal plane of the objective, where another
aperture diaphragm may be used for improving the quality of the image
formed by the objective. Images of both aperture diaphragms are finally
formed in the RAMSDEN disk and may be observed there with the use of
a low-power magnifying lens.
Bastard Diaphragms and Baffles. The former are located in positions
intermediate to focal planes and image planes. They affect both field
26 Use of Optical Aids

and apertU"e, and they should be avoided since they do not permit
independent regulation of these two factors. Baffles may be located
wherever they serve best in the elimination of stray light, but their openings
should be so large that they permit passage of all image-forming rays
and cannot act as diaphragms.
Selection of· Microscopes. Whereas the simplest toy microscope will
suffice for many chemical tasks, efficient performance of work suggests
the use of several instruments selected according to the special needs of
the laboratory.
A chemical microscope as specified by CHAMOT and MASON (88) will
serve for most work of a routine nature. A very desirable companion is
a binocular microscope of the Greenough type with objective magnifications
from about one to eight, which may be used on a microscope stand or held
by some type of universal stand to serve as binocular magnifier.
Very satisfactory is a large stand with wide body tube and calibrated
rotating stage, as used in biological research. If properly chosen, it will
permit photography and projection as well as the use of polarizing equipment
and a great variety of illuminating devices. It should be possible to ex-
change the stage for a rotating stage with built-in mechanical movements
vertical to the axis of the microscope, to use.a spectroscopic eyepiece,
or to change to binocular viewing.
A simple microscope may be set aside for use with a heating stage
and provided with an illuminating device that meets the requirements
of the heating stage.
For convenience, microscope tables may be designed which hold the
instrument so that the stage appears at the level of the table top and
which are equipped with push buttons and voltage regulators or rheostats
to provide the various types of illumination for direct observation, photo-
graphy, and projection.
Microprojection and Photomicrography. Specially corrected lenses which
flatten the image are desirable in photomicrography; the use of reasonably
monochromatic (green) light may do away with the need for elaborate
correction for chromatic aberration. The common problem of projection
and micrography is the need for strong illumination and simultaneous
exclusion of stray light. It is obvious that the latter will befog photographic
emulsions. Only the experiment seems to be able to bring the realization
that projection (especially with polarized light) requires a completely dark
room if high magnifications shall be used.
The source should be small to permit efficient collection of the emitted
light with condenser lens (and concave mirror). A tungsten arc or an
incandescent lamp with concentrated filament furnish steady illumination
for photomicrography. They are also satisfactory for microprojection if
an image of only 10- to 25-cm diameter is needed. The crater of the positive
 The Microscope 27

electrode of a d. c. carbon arc is most suitable for the projection of large

images. A close approach to a point source of high brilliancy are the
Concentrated-(zirconium)-Arc lamps which are offered with 0.075- to
2.5-mm diameter of the arc ana 2 to 300 Watt power.
Stray light is effectively reduced by using critical illumination as
schematically indicated in Fig. 1. The light of the source is gathered by
a condenser with iris diaphragm. The substage condenser is focused to
obtain an image of the opening of the diaphragm of the auxiliary condenser
in the plane of the object, whereupon the iris diaphragm is closed until
only the field of vision receives light. This reduces the width of the
illuminating beam of light and, consequently, the stray light to the unavoid-
able minimum. In practice proceed as follows.
Place upon the stage a preparation of good transparency (thin layer
of thymol crystals from Expt. 13). Adjust the source, condenser, and
mirror with all diaphragms wide open until the light beam passes the
preparation and enters the objective so that a circular image ~ppears upon
the screen (ground glass plate). By the standard procedure (p.30),
focus the preparation to obtain a sharp image upon the screen and, if
necessary, readjust mirror, source, condenser, and substage condenser
until the field is evenly illuminated. Close the diaphragm of the auxiliary
condenser to leave an opening of about I-cm diameter and focus with
the substage condenser until a sharp image of the opening of this diaphragm
is obtained upon the screen; be certain not to change the focus of the
microscope. The diaphragm of the auxiliary condenser is now in a conjugated
image plane, and its sharp image appears in the object plane and the image
plane in the eyepiece as well as upon the screen. Open and close the
diaphragm and see its image changing upon the screen. Move the mirror
until the image of the opening of the diaphragm is concentric with the
circular boundary of the screen image. Finally open the diaphragm of
the auxiliary condenser wide enough so that the image of its opening is
just outside the field. If rulings contained in the eyepiece shall be used,
focus with the eye lens until they appear on the screen. Then again focus
the preparation and readjust the position of the substage condenser if
,this should be necessary to restore critical illumination.
Since the size of the field of vision depends upon the objective magnifica-
tion, it is obvious that critical illumination must be restored after each
change of objective. To maintain an image of a given size, the amount of
light needed for its satisfactory illumination must be concentrated into
an area (field of vision) which becomes the smaller, the higher the objective
magnification. Very soon, the intensity of the unavoidable infrared
radiation becomes so high that the object of investigation is endangered;
solids may explode, melt, sublime, char, or burn; liquid preparations start
to boil. To avoid this, cells filled with 5 % solution of copper sulfate penta-
28 Use of Optical Aids Expt.l

hydrate or an acidified saturated solution of ferrous sulfate are inserted

between source and mirror of the microscope; plates of special infrared
absorbing glass may be used in addition.
Consult the special literature for details (88, 106, 1240); ROYER (708,
1240) writes on the determination of exposure for color photography;
COURTNEy-PRATT and HUGGINS (1075) on 100000 frames per second
cinematography.
Microscope Lamps. A microscope lamp for chemical use should permit
quick adjustment in height so that one may readily change from work
by transmitted to work by reflected light. It should permit focusing a
bright small spot of light upon the object and have a slot for the insertion
of ground-glass plate and color (daylight) filters, singly and in pairs. It
should be light in weight, small in size, and it should not heat up so that
it becomes difficult to handle. A telescoping tube containing a low-wattage
bulb and a condenser lens could be mounted in a well-balanced manner
on a ball-and-socket universal joint which may be moved along a vertical rod.
As a substitute, desk lamps with gooseneck arm are quite satisfactory.
The use of spot lamps for automobiles and motorcycles as well as of electric
bicycle headlights may be considered. A rather efficient and handy micro-
scope lamp may be obtained with a lens bulb (Ma7da 222, 0.55 watt)
for pen lights. The glass at the top of the tiny bulb is fused into a bead
which acts as a powerful condenser lens. Potential of 2.2 volts may be
supplied to several of such bulbs via an ordinary 2.5-volt filament trans-
former and a 10-ohm variable rheostat in series with the bulbs.

Experiment 1
Inspecting and Cleaning the Microscope
Lens paper; carnel's-hair bru"h, roun:, bathe in 1 rnl of acetone and allow
to d y; cotton and tooth picks; clean rag.
Remove the microscope from its case and place it on a clean table top_
When carrying a microscope, grasp its arm with the right hand and support
the base of the stand with the palm of the left hand. This prevents accidents
such as smashing the lower part of the instrument into the edge of a table
top.
Inspect inside and outside of the case. Check the inventory for objectives,
containers, eyepieces, racks, and other accessories, and then proceed to
the testing of the miCIoscope.
1. Adjustment ot Eyepiece, Method ot Viewing. Micrometer and cross-
hair eyepieces should have an adjustable eye lens. Remove the eyepiece
from the microscope, hold it against a suitable source of light (try a window,
a white wall, lamps) and move the eye lens until a sharp image of the
rulings is obtained; use spectacles if they correct for astigmatism. Keep
Expt.l The Microscope 29

the eye lens in this position and return the eyepiece into the tube of the
microscope.
In the instance of binocular microscopes and binocular eyepiece
attachments, at least one of the paired eyepieces should have an adjustable
eye lens. In addition, there must be some means for adjusting the distance
between the eyes (interpupillary distance). Use spectacles to correct for
astigmatism. Cover the eyepiece having the adjustable eye lens and focus
as directed under (5) and (6). When sharp focus has been obtained, transfer
the cover to the other eyepiece and, without changing the position of the
microscope tube, focus with the eye lens of the eyepiece until again a
sharp image is obtained. It is understood that the left eye always is before
the left eyepiece and the right, before the right. Finally remove the cover
and view through both eyepieces. In general, two circular images will
be seen, that overlap for the most part. Operate the adjustment of the
interpupillary distance until the two images fuse into one.
Binocular vision facilitates the recognition of three-dimensional
structures and prevents abuse of the eyes by permitting them to function
as intended by nature. Since both eyes must be kept centered with respect
to two eyepoints, head and neck must be held very rigid, which may cause
muscular fatigue (88).
2. For monocular observation make it a principle to frequently alternate
the use of left and right eye and to relax the eyes at frequent intervals
by looking at distant objects. Both eyes should be kept open while looking
into the microscope; the unoccupied eye might be protected by a black
shade, which may prevent development of the unfortunate habit that
one eye goes automatically blind when viewing through an instrument
and binocular vision with instruments becomes impossible.
When viewing through the eyepiece of any optical instrument, relax
the eye so that it remains focused at infinity and make no effort to "see".
Let the instrument (eye lens) do the focusing. If very much time will
be spent with looking into an instrument, obtain a binocular eyepiece
attachment or use projection of the image upon a screen, which last solution
IS most convenient in every respect.
3. Clean both sides of the mirror with lens paper.
4. Preliminary Adjustment of Illumination. If the substage has a
diaphragm, open it wide. Place a lamp in front of the microscope and
adjust the concave mirror until the light is sent through the hole in the
stage to the front lens of the objective. If a microscope lamp is not available,
use sunlight reflected from white clouds or bright walls, or the light from
ceiling lamps. In such instances, try the plain and the concave mirror,
and use the one which gives the better illumination.
5. Place a clean slide on the stage so that its edge runs across the center
of the hole in the stage, and then focus with the standard procedure:
30 Use of Optical Aids Expt.l

a) Estimate the working distance of the attached objective from the

data available. Information on focal lengths and magnifications is often
posted inside the carrying case.
Observe the objective from the side with the eyes held at the level
of the microscope stage. Lower the tube with the coarse adjustment until
the front lens of the objective has approached the preparation (top surface
of the slide) to approximately half of the estimated working distance.
With a high-power objective, go as near as possible to the preparation
without touching it.
b) Look into the eyepiece, and slowly raise the tube by means of the
coarse adjustment until a sharp image of the upper edge of the slide is
obtained. If no image appears, one of three things may have happened:
the tube has been raised too fast, and the image has been missed; the
front lens of the objective never was below the working distance; or the
slide is not in the proper position on the stage. Make the necessary adjust-
ments, and focus again as directed under (a) and (b).
It may happen to a beginner that he raises the tube so high that rack
and pinion disengage (properly designed microscopes have a stop to prevent
this). The meshing of the gears must be done very carefully to avoid
stripping.
The fine adjustment should not be used for focusing with low-power
objectives, but reserved for small final adjustments in the study of detail
and for the focusing with high-powers. It should be kept engaged near
the center of the micrometer motion to provide a maximum of adjustment
in both directions. Lost motion and lag indicate wear or poor workman-
ship. The fine adjustment should be graduated for the measurement of
vertical displacement.
6. Final Adjustment of Illumination. Adjust the mirror and try the
plane and the concave surface until the most effective and even illumination
of the whole field is obtained. If the illumination is too strong, place a
ground-glass plate or a sheet of paper in front of the mirror or move a
lamp farther away from the microscope.
If the substage carries a condenser and diaphragm, move the condenser
up and down to find the position which gives the best even illumination
of the whole field. Finally adjust the brightness of the field by opening
or closing the substage diaphragm.
7. Examination of the Image. The edge of the slide should appear
sharply defined; otherwise the field should be empty. Films of dirt on the
lenses of the microscope proper will impair the sharpness of the image;
the edge of the slide will appear blurred or clouded by a haze, and refocusing
will give no improvement. Particles adhering to lens surfaces are indicated
by the presence of more or less sharply defined spots in the field of vision.
The location of the imperfections may be found as follows.
Expt.l The Microscope 31

a) Move the slide. Images of dust particles and fingerprints adhering

to the slide move simultaneously. Clean the slide until these imperfections
disappear.
b) Rotate the eyepiece; a cross-hair eyepiece must be slightly lifted
to make this possible. This imparts rotary motion to images produced
by dust or dirt on the glass surfaces of the eyepiece.
c) Move the mirror slightly. If a condenser is used, move it up and down.
This will reveal images of irregularities in the light source (the condenser
may be focused upon the frosted glass or the filament of the bulb, etc.)
or images of foreign matter adhering to the condenser lenses. Eliminate
the former by changing the distance of the light source from the microscope;
the latter, by cleaning the lenses of the condenser and the aperture of the
substage diaphragm.
d) Imperfections not located in either of the aforementioned parts of
the optical system must originate in the objective. Corrosion, dirt, and
dust on the lenses of the objective destroy the sharpness and clarity of
the image of the object, but they are only vaguely or not perceived when
looking into the eyepiece of the microscope.
e) Imperfections which are in sharp focus simultaneously with the
object (edge of the slide) must be located in one of the conjugated image
planes: plane of the field diaphragm of the eyepiece (micrometer scale or
cross hairs), plane of the preparation viewed, source of illumination or
some plane between microscope condenser and lamp.
Repeat the tests with the other objectives and eyepieces belonging
to the microscope. Finally inspect and, if necessary, clean the parts as
directed in the following. Start by carefully washing and drying the hands.
8. Eyepiece. Remove the eyepiece from the drawtube of the microscope,
unscrew the mountings of eye lens and field lens, and place the parts
upon a sheet of clean paper. Inspect all lens surfaces for corrosion, scratches,
and dirt by holding them so that light is reflected from them to the eye.
A perfect surface looks like a clean, faultless mirror. Rotate each lens
around its optic axis during inspection. Imperfections will rotate with
the lens, whereas images and reflections will remain more or less stationary.
Clean lens surfaces by first brushing or blowing off dust to make certain
that abrasive particles cannot make scratches when wiping. Then breathe
upon them and wipe off the condensed moisture with doubled layers of
lens paper. Do not apply much pressure, or oil from the skin will go through
the paper and collect on the lens surfaces. Inspect the lenses again and,
if they are clean, assemble the eyepiece without delay.
Micrometer eyepieces. Unscrew also in the middle to get access to the
micrometer plate. When unscrewing, hold the eyepiece right side up so
that the micrometer plate will not be dropped. Often it will suffice to clean
the upper surface of the plate with a camel's-hair brush without removing
32 Use of Optical Aids Expt.l

it from the mounting. It may be taken out, however, and inspected and
cleaned as described for lenses.
Cross hairs are either real hairs mounted on a ring or engravings on
a glass plate. In the former instance, cleaning is rarely necessary, but
may be done by gently blowing against the hairs which must not be touched,
for they are easily broken. In the latter instance, proceed as with eyepiece
micrometer scales.
9. Objective. Inspection and cleaning of lenses has been outlined under
the caption "Eyepiece". Standard objectives must not be taken apart,
and cleaning that requires the separation of lenses should be done by the
manufacturer. Thus restrict cleaning to the lower surface of the front lens
and the upper surface of the top lens. For cleaning the latter, use a swab
of cotton on a toothpick and finish with a strong blast of clean air.
The exposed surface of the front lens needs special and continuous
attention. Clean it immediately if it should have had contact with a prepara-
tion or been exposed to a preparation giving off vapors of any kind. As
the first step in cleaning, the surface of the lens and the surrounding part
of the mounting may be carefully rinsed with tap water and wiped dry
with lens paper. Alcohol and other organic solvents are best avoided
since they may act upon a cement used by the manufacturer.
An inexpensive microscope may have an objective of the separable
type, which provides two or three magnifications. In such instances,
inspect and clean each section. When using it, keep in mind that such an
objective has as many front lenses as magnifications and that the perfection
of each of these front lenses must be continuously guarded.
10. Coarse Adjustment. Test for excess play and lost motion. The
body tube carrying the full equipment should remain in every position
into which it is brought. Check the directions of the manufacturer concerning
tightening and loosening the rack-and-pinion motion.
Microscopes may still be found, that lack a stop preventing the pinion
from overriding the rack, and consequently their coarse adjustment may
become damaged by inexperience, haste, or carelessness in meshing the
gears. For inspection, raise the body tube to its highest position with
the pinion head. Note that the gears are disengaged, and then grasp the
body tube and pull it up to remove it from the stand for inspection of the
rack. Rotate the pinion head while inspecting the teeth of the pinion.
Any imperfections of the teeth of either part of the mechanism must be
corrected immediately to prevent its complete destruction.
For the meshing of rack and pinion, insert the rack of the body tube
into the pinion slot and push the tube down until the rack just touches
the pinion. Then impart short turns to the pinion head and a very gentle
downward pressure to the tube until proper meshing occurs so that the
gear works smoothly . Avoid haste and the use of force. Follow the procedure
Expt.l The Microscope 33

recommended for meshing whenever it happens that the pinion overrides

the rack.
11. Oondenser. With some microscopes, the substage is moved by a
rack-and-piniQn arrangement and may be removed from the stand like
the body tube. Others have a quick-acting screw which permits swinging
the condenser out of the optic axis. The condenser proper is usually held
in a ring clamp which may be provided with a set screw; usually, the
uppermost lens of the condenser may be removed by unscrewing the
mounting. In the instance of petrographic stands, provision may be made
allowing the uppermost lens of the condenser to be swung aside by lever
action. An iris diaphragm and polarizing equipment may be attached
below the condenser lenses. The mirror is attached to either substage
or stand and may be removed by pulling the carrying rod out of its socket.
Usually the parts of the substage are more conveniently accessible after
tilting the microscope into a horizontal position.
Swing out the condenser or remove it completely from the microscope.
Dust and dirt collect mostly on the top surfaces of the front lenses. Inspect
and clean the accessible lens surfaces as directed above. Dust will also
collect on the top surface of a diaphragm; remove it with a brush (working
on the inverted apparatus) or a blast of clean air. Close and open an iris
diaphragm, and inspect the blades for absence of corrosion, kinks, and bends.
If the microscope is equipped with objectives of magnification 20 and
higher, the centration of the corresponding condensers should be checked.
Do this, after focusing objective and condenser upon a preparation, by
removing the eyepiece and looking down the tube at the back aperture
of the objective. When the substage diaphragm is nearly closed, its small
opening should be seen at the center of the back aperture. A thoroughly
satisfactory substage will permit adjustment. If none is provided, however,
the centering may be done at the objective or the objective changing device;
this assumes that a rotating stage is also equipped with centering screws.
12. Stage. Inspect for corrosion and damage by heat. Sockets for
attaching clips or other apparatus should not be clogged. A central opening
of 25 mm or more in diameter will permit the use of all kinds of condensers.
A rotating stage should be provided with graduations around the
circumference and a vernier; a locking device is desirable. The stage
should rotate smoothly without any indication of lateral motion. Inspection
should show whether or not it is readily removable for lubrication and
adjustment. Center the stage (Expt. 6) and make certain that all necessary
accessories are supplied.
13. Final Inspection. Assemble the microscope after cleaning and
repeat the tests given under (7). If necessary, repeat the cleaning and
adjustment. Permanent imperfections of lenses can be removed by the
manufacturer.
Benedettl·Pichler, Identification 3
34 Use of Optical Aids

Illumination of the Specimen

Whether or not the eye is aided by a magnifying lens or a microscope,
the basic requirement for the observation of the structure or color of a
specimen is that the latter is able to properly act upon light of the right
kind. Light travelling through the specimen will be affected differently
from light reflected from the surface of the specimen. Light transmitted
through the specimen will show the color resulting from selective absorption
(body color) and will reveal the shape by means of refraction rather than
reflection. Light reflected from the surface of the specimen will show the
color resulting from selective reflection (surface color) and reveal the
structure by reflection rather than refraction. Light passing in such a
direction through the specimen that it cannot reach the eye but for gross
deflection may reveal heterogeneity and fluorescence. In addition and
quite generally, no matter what the effect of the specimen, it will be the
more readily observed the more it differs from the surroundings.
It follows for all visual observation, with and without use of magnifying
devices, that illumination (kind of light and direction of incidence upon
the specimen in relation to the location of the eye) and background will
decide what may be seen. Magnifying devices only aid in resolving the
detailed information already contained in the light coming from the specimen.
As a rule, white light similar in composition to sunlight is preferred for
illumination so that colors appear as in the natural light of the day. There
is never any harm in trying ultraviolet which may reveal tell-tale fluorescence.
Monochromatic light is useful in photomicrography and in the determination
of optical constants. Polychromatic illumination gives special darkfield
effects.
The angle of incidence may be varied through 180 degrees in an attempt
of finding all available information. By all means, one should try transmitted
light as well as reflected light with backgrounds of various colors. The
detection of heterogeneity (turbidity, opalescence) and fluorescence calls
for Tyndall illumination (incidence vertical to the line of vision and black
background) or darkfield illumination. One might keep in mind that
appearances revealed by reflected light are more readily interpreted since
one is accustomed to see by reflected light. The wide use of transmitted
light in microscopy is made possible by the fact that many small objects
are transparent.
The methods of illumination used in microscopy are briefly outlined
in Table II. The variety is explained by the desire of seeing small detail.
Very slight gradations of optical density must be brought out by the use of
special diaphragms and other devices in the image-forming system and in
the illuminating apparatus: Schlieren microscope and phase contrast
microscopy (88).
Expt.2 Illumination of the Specimen 35

Table II. Type8 of Illumination U8ed in Vi8ual Micr08copy

Transmitted Light Reflected Light

Axial: a) Parallel: beam of light Axial: beam or pencil of light

parallel to the optic axis of the micro- parallel to the optic axis of the micro-
scope; 80urce of light at a great di8tance scope; tran8parent reflector (cover 8lip
and plane mirror in the optic axi8, inclined at 45 degree8) between objective
Fig.2a. and 8pecimen 80 that a horizontal beam
b) Oonvergent: pencil of rays along of light i8 reflected along the optic axi8
the optic axis; concave mirror, Fig. 2b, upon the surface under examination;
or condenser, Fig. 3. vertical illuminator for medium- and
Oblique: a) Parallel: beam of light high-power objective8 (metallography).
inclined to the optic axis; plane mirror Inclined: beam or pencil of rays
swung to one 8ide and no condenser. coming from above the stage and
b) Oonvergent: pencil ofraysinclined inclined to the optic axis; direct
to the optic axis; concave mirror 8wung illumination from light 80urce located
to one 8ide and no condenser; condenser above the 8tage.
u8ed and cardboard inserted to cover all Annular Oblique: Silverman illu-
but one 8ide; condenser with 8mall minator for low magnifications; Epi-
opening of its diaphragm off center. Oondenser W (Zeiss) or Ultropak (Leitz)
c) Annular: dark-field condenser; tor medium and high magnification8.
two-color illumination of Wright and
Rheinberg used in Zei88 , M ikropoly-
chromar (88).

Experiment 2
Illumination of M icr08copical Specimen8
Slide with ground-glass surface; squares of white, black, and colored paper,
about 3 cm X 3 cm; cedarwood oil.
To obtain a suitable specimen, draw a line with a silver coin through
the center of the rough surface of the ground-glass slide. Parallel to this
line and about 1 mm away from it, make another streak with a penny.
With a soft black pencil draw a line vertically across both streaks.
Place the slide on the stage of the microscope so that the intersections
of the lines appear at the center of the opening. Use the weakest objective
and remove the condenser. Adjust the illumination and focus upon the
rough surface of the slide as directed in Expt. 1 (5). Try sunlight or the
light from a distant lamp with the plane mirror, and also try the microscope
lamp with plane and with concave mirror to obtain even illumination of
satisfactory brightness over the whole field of vision. Note that the choice
between plane and concave mirror is determined by the size of the field
of vision, the intensity and distance of the source of light, and the needed
brightness of the image, Fig. 2. Observe that the rough surface of the slide
interferes greatly with the recognition of the particles left behind by coins
and pencil.
36 Use of Optical Aids Expt.2

Cut off the light coming from the mirror, slip a square of white paper
underneath the slide, and direct the illuminating beam of light down
upon the preparation at an angle of about 45 degrees. To this end, either
raise the microscope lamp, or use a special arc lamp or sunlight which may
be guided with the use of a mirror (plane or concave). Note the appearance
of the rough glass surface and the visibility of graphite and the metallic
particles.
Place a fragment of a cover slip upon the part of the slide where the
streaks are located, and (with capillary or glass rod) add to the edge of
the cover slip a droplet of cedarwood oil so that the latter is drawn, by
capillarity into the space between slide and cover. Since the cedarwood oil
has nearly the same refractive index as the glass, the particles o~ the streak
now appear imbedded in a sheet of clear glass. Note the visibility and
color of the particles by reflected light when using white, black, and colored
paper as background. Finally tum back to transmitted light il.,nd again
note the color of the particles. When doing this, be certain to exclude
reflected light by cupping the hands around preparation and objective
or enclosing the stage by a black curtain. Try the plane mirror as well
as the convex one. Note that transmitted light cannot show anything
beyond the silhouette of an opaque object.
Change to an objective of magnification 20 or 10, focus with the standard
procedure, Expt. 1 (5), and try to obtain satisfactory illumination by
transmitted light with the use of plane and concave mirror. Finally insert
the condenser below the stage and try for even illumination of the whole
field and satisfactory brightness by using the condenser once with the top
lens in place and once with the top lens removed. In each instance, have
the substage diaphragm wide open and move the condenser up and down
until the whole field is brightly illuminated; then reduce the intensity by
closing the substage diaphragm until the image can be viewed with comfort.
Since the opening of the field diaphragm of the eyepiece is practically
constant, the field of vision (the actual area of the object which is reproduced
inside this diaphragm and consequently in the final image) must decrease
.at the rate at which the square of the magnification of the objective grows.
Since the final image on the retina, screen, or photographic plate retains its
size and always needs the same amount of light, it becomes necessary to
,concentrate the needed light into a rapidly contracting area of the object
when proceeding to higher objective magnifications. This is done by
-collecting the light by means of the substage condenser into the field
.of vision, and the smaller the latter becomes, the more powerful must be
the condenser..The focal length of the condenser should be properly related
to that. of the objective, and for the most efficient use of the available
light, both should have the same aperture, Fig. 3a. Removal of the upper-
most lens reduces the focal length and the aperture of the condenser so
Expt.3 Illumination of the Specimen 37

that it becomes better suited for use with objectives of moderate magnifying
power. In spite of this, because of the difference of aperture, only a small
portion (shaded in Fig. 3b) of the light may be used for image formation.
Under these conditions, unconventional use of the concave side of the
mirror and empirical readjustment of the position of the condenser may
improve the illumination.
Again cut off the light coming from the mirror and try observation by
reflected light with backgrounds of different color. Note the effectiveness

Fig. 3. Use of Condensers. The shaded area indicates the pencil of light actually
used for image formation

of the various kinds of illumination in the differentiation of the three

types of particles and in the observation of the structure of the rough
glass surface. If available, try also other types of illumination.

Experiment 3
Calibration ot Eyepiece Micrometer; Working Distances and Fields ot Vision
Stage micrometer; square of millimeter graph paper.
The center of the stage micrometer carries an accurately divided scale
produced by a photographic process or by engraving. The value of the
divisions is indicated by the manufacturer. If necessary, the stage micro-
meter may be cleaned with lens paper, but care should be taken to avoid
damage to the cover slip protecting the micrometer scale.
First decide upon the tube length to be used and adjust it accordingly.
Then carefully focus the eyepiece micrometer scale with the eye lens.
Place the square of graph paper upon the stage, attach the objective
of lowest magnification, and by either transmitted or reflected light focus
38 Use of Optical Aids Expt.3

upon the ruling. Obtain the diameter of the field of vision in millimeters
by counting the number of millimeter spaces included. Get the working
distance of the objective (i. e., the distance from the preparation to the
front lens of the objective focused upon the preparation) with a millimeter
rule held parallel to the optic axis of the microscope.
Comparison 01 Scales. Test for absence of parallax by slightly moving
the eye in front of the eyepiece. If the rulings seem to change their relative
positions when the eye is moved, repeat the focusing of the eyepiece micro-
meter scale and of the ruling on the graph paper until a movement of the
eye will no longer give a relative displacement of the images of the two
scales. This will happen when the two scales are truly located in conjugated

Fig. 4. Calibration of Eyepiece Micrometer. Using the coincidences at 4 and 14 of

the eyepiece scale and the right edges of these lines, one finds 10 div. = 0.17 mm
and 1 div. = 17 Il-m

image planes. Obviously, a micrometer eyepiece in conjunction with an

image forming system (objective of microscope or telescope, lens of refractom-
eter) may be generally used for the exclusion of parallactic errors by
the projection of the image of the object into the plane of the rulings.
Rotate the eyepiece or the stage so that the lines of both scales become
parallel. Then select two points, not too far from the center of the field
and approximately one-third of the diameter of the field apart, where
lines of the eyepiece scale coincide with lines of the graph paper. It may
be necessary to move the graph paper to obtain coincidence. The lines of
the graph paper may appear so wide that it becomes necessary to decide
whether the center or an edge of the lines shall be used.
Determine and record the distance between the two points of coincidence
in units of both scales. Compute and record the value of a division of the
eyepiece micrometer scale for the particular objective and tube length.
If a stage micrometer scale is not available, repeat with all other objectives
and collect all data in a table.
 The Immersion Method for the Determination of Refraction 39

It a stage micrometer scale is available, use it for repeating the comparison

<>f scales. The direct focusing of any small object such as a stage micrometer
,scale is rather trying, especially when a strong objective is being used.
It is difficult to place the slide so that the small object gets within the
minute field of vision, the location of which may only be guessed. In such
instances, focus first upon some coarse structure located in approximately
the same level as the object: the edge of a cover slip or label; or the upper
edge of the slide, etc. Focus by the standard procedure, Expt. 1 (5),
and then move the slide until the object (micrometer sClile) appears in
the field of vision.
Starting with the test for absence of parallax, repeat the comparison
,of scales as directed above. From the known value of the division of the
.stage micrometer scale compute the value of the division of the eyepiece
micrometer scale for the particular objective and tube length in absolute
measure (mm or ,um). If the tenths of the divisions have been estimated,
three to four figures may be significant. Since the eyepiece scale is now
.accurately calibrated, compute the distance between the lines of the graph
paper in millimeters.
Repeat the whole experiment with all objectives supplied and collect
.all data (eyepiece and tube length; objective: focal length, working distance,
magnification, diameter of field of vision, value of division of micrometer
.scale, and possibly distance determined for one interval of the graph paper)
in form of a table which may be posted inside the microscope case.

'The Immersion Method for the Determination of Refraction

The outlines of an object become visible if the object and its surroundings
behave differently to light. They may differ in absorption or rate of
-transmission (refractive index) or both. Selective absorption in the visible
range imparts color; complete absorption, opacity of a homogeneous object.
Differences in the rate of transmission lead to refraction, reflection, and
·diffraction phenomena at the boundaries, which render them visible.
When the outlines of a colorless transparent solid are invisible after
immersion into a colorless clear liquid, one may assume that their refractive
indices are identical. If the refraction of either one of them may be found,
that of the other is already determined. Series of liquids as well as solids
have been collected to serve as standards of known refractive index and
may be obtained from laboratory supply houses. A series of solid standards,
,supplied in powder form, has been developed by KOFLER (98, 163); it
;starts with cryolite (n = 1.3400), fluorite (1.4339), and vitreous silica
(1.4584) and continues with the use of twenty different glasses up to
'n = 1.6718. R. FISCHER adds sodium fluoride (1.3255) and lithium fluoride
.(1.3918). A list of liquid standards is assembled in Table III. Since even
40 Use of Optical Aids

the most stable substances may slowly undergo changes, the refractive
index of liquid standards should be occasionally checked with the refractom-
eter. Gaps in refraction may be bridged by mixing adjacent liquids until
the outlines of the solid disappear, whereupon the index of the mixture
is determined with the ABBE refractometer; to this end, it is necessary
that the liquids are miscible, and it is desirable that they have about
the same vapor tension (boiling point) to prevent fractionation during use.
Since the solid must be insoluble in the liquid medium, it is necessary to
have alternatives, and aqueous solutions of NaCI, KI, and K 2HgI 4 will
provide standards for the range from 1.34 (water) to 1.72.

Table III. Liquids for Use as Standards in Refractive Index Determination

n Ib.Pt·1 n lb. pt_

I °0 °0

A B
Methanol 1.329 65 Mixtures of SPANGEN- I
Water 1.333 100 BERG (1225)
Acetone 1.359 56 1.47 to 1.62
Ethanol 1.37 78 Glycerol and 1.47 290
n-Hexane 1.37 69 Quinoline 1.624 237
Ethyl acetate 1.373 77 1.54 to 1.62
n-Heptane 1.388 98 Diethylaniline and 1.542 218
n-Butanol 1.399 ll8 Quinoline 1.624 237
n-Pentanol 1.40 138 1.59 to 1.74
n-Butyl chloride 1.402 79 Bromoform and 1.596 150'
Methylene iodide 1.740 180·
p-Dioxane 1.422 101
Ethylene glycol 1.431 197 1.62 to 1.66
Ethyl citrate 1.442 Quinoline and 1.624 237
iX-Bromonaphthalene 1.658 281
Chloroform 1.45 61
Cyclohexanone 1.450 156 C
Cyclohexanol 1.466 161 Aqueous Solutions
p-Cymene 1.490 177 Sodium chloride 1.33 to 1.37
Benzene 1.501 80 Potassium iodide 1.33 to 1.50
Ethyl iodide 1.513 72 Potassium tetra-
Anisole 1.5171154 iodomercurate 1.50 to 1.72
Chlorobenzene 1.525 132

The dependence of the refractive index upon the frequency of the light
and the temperature should be considered. Both factors affect the refractive
index of liquids much more than that of solids. In the instance of liquids.
the refractive index decreases by about 0.0015 for an increase of 10 nm
(100 A) in the wave length and by about 0.0004 per degree Celsius (centigrade)
rise of temperature. For solids, the dispersion due to difference in wave
Expt. 4 The Immersion Method for the Determination of Refraction 41

length is about half that given for liquids, and the temperature effect is
about one-tenth of that for liquids.
It follows that the boundaries cannot be made to vanish when white
light is being used since the refractive indices of solid and liquid may be
matched for only one color of light at a time; when this happens, the
boundary still remains outlined in the colors of those parts of the spectrum
for which balance has not been achieved. Use of monochromatic light
overcomes this difficulty and permits estimation of the index to one unit
in the third decimal place (for a given wave length and temperature);
if the monochromatic light is chosen so that it is not noticeably absorbed
by the two media, the immersion method may be extended to objects
with body color. In addition, small differences in refraction may be bridged
by varying either the wave length by means of an adjustable monochromator
or the temperature with a heating stage (98, 163); both are adjustable
with EMMONS' double variation apparatus consisting of arc lamp, mono-
chromator, polarizing microscope equipped with special universal stage,
ABBE refractometer, and system for circulating water of controlled temper-
ature (113).
The effect of differences in refractive index is already known from
the experiment (2) with the ground-glass surface. It may be imagined
that the difference of refractive index may be estimated to the nearest
tenth of the unit by mere comparison with a scale of specimens exhibiting
differences of 0.0, 0.1, 0.2, and 0.3. A systematic procedure will start by
comparing with a standard of n = 1.5 and use a test that indicates whether
the unknown has the higher or lower refractive index. The next comparison
will accordingly be made with a standard of about 1.4 or 1.6. Two more
trials should reduce the gap to about 0.02, whereafter it should be possible
to decide upon the procedure for the final matching. Tests for determining
the direction of the refraction difference are described in Expt. 4.
For lack of suitable immersion liquids, the immersion method becomes
impractical for solids with refractive indices above 1.75.

Experiment 4
Phenomena Oau..~ed by Differences ~n Refraction
Ethanol, ethylene bromide, quinoline, and cedarwood oil.
The following experiments suggest also types of illumination for getting
a good image of objects immersed in a medium of similar refractive index.
A. Clean three slides, and place upon the center of each a drop of about
0.5,u1 of 1 % NaCI solution. Put the slides aside until the drops have
evaporated completely, and use the time for starting part B of the experi-
ment (see below).
42 Use of Optical Aids Expt.4

When evaporation is complete, drop on each residue a clean cover slip.

Transfer one of the preparations to the stage of the microscope and focus
with an objective of magnification 10 or somewhat stronger. If the residue
is small or difficult to see, start by focusing the edge of the cover glass.
Most of the residue may form irregular clusters of crystals, but usually
it will be possible to find some well developed single crystals with square
or rectangular outline. Use such crystals for the following observations.
With a stirring rod, transfer a drop of ethanol to the edge of the cover
slip so that the liquid is drawn into the space between slide and cover glass.
Focus upon a simple crystal and note that the outlines are now less
pronounced than before adding the ethanol. Vary the intensity of the
illumination by opening and closing the substage diaphragm, raising and
lowering the condenser, or changing the distance between mirror and light
source. Note that the outlines become weaker when the intensity ot the
illumination is increased. The difference of refractive indices is now
1.54 - 1.36 = 0.18. The direction of the difference is found by the
following two tests of about equal sensitivity, ± 0.001 with monochromatic
and ± 0.005 with white light. Frequently, it will be wise to try both tests
to find which gives the more conclusive evidence.
The Becke Test is most suitable for thin objects bounded by surfaces
about parallel to the optical axis of the microscope. It may not be reliable
with rounded grains.
Move the slide to bring the test crystal into the center of the field.
Use weak axial illumination, either just the plane mirror with the source
at a distance or the condenser with the diaphragm only slightly open.
Focus the crystal to get it bounded by a single dark line. Continuously
looking at the crystal, slowly raise the body tube a short distance. Note
that a fine bright line appears parallel to the dark outline. This "halo",
Fig. 5, is the Becke line. Slowly lower the body tube and note that the
Becke line moves toward the black outline and disappears when the crystal
is sharply in focus again. Lower the body tube some more and observe
the appearance of the Becke line on the other side of the black outline.
If the phenomenon is not distinctly visible, repeat the experiment with a
thinner (smaller) crystal.
For interpreting the Becke test, remember the rule: on lowering the
body tube, the Becke line moves into the medium of lower refractive index;
on raising the tube, the Becke line moves in the direction in which the
refraction rises.
Oblique Illumination. Retain in the center of the field the crystal
which gave a good Becke test. If a substage condenser is being used,
bring it into the highest position that gives a· good illumination; this is
necessary to avoid an otherwise possible reversing of the shadow effect.
Open the substage diaphragm wide and obtain oblique illumination by
Expt. 4 The Immersion Method for the Determination of Refraction 43

covering first all but the extreme left side of the condenser opening with
a strip of black cardboard, about 4-cm wide, and then moving it to the
left until only the extreme right side of the condenser opening admits light.
Move the strip back-and-forth while watching the crystals through the
microscope for changes in the distribution of shadow and light.
If the substage condenser is mounted to permit lateral movements,
repeat the experiment by using a narrow opening of the substage diaphragm
and moving the condenser-diaphragm combination back-and-forth between
the extreme positions on the left and on the right.

Fig. 5 Fig. 6
Fig. 5. Sodium Chloride Crystals Immersed in Alcohol. Microscope is raised above
the position of sharp focus
Fig. 6. Glass Thread Immersed in Water. Mirror of the microscope is swung to
the left·hand side

Finally remove the condenser, open wide a remammg diaphragm,

and use the plane side of the mirror. Start with the mirror in the optic
axis of the microscope, and adjust it so that the best illumination is obtained.
Then, while observing the crystals through the microscope, swing the
mirror to the left side until a shadow enters the field of vision and its
edge almost arrives at the opposite boundary of the field, Fig. 6. While
watching for changes in the distribution of shadow and light in the images
of the crystals, swing the mirror through the center to the right side-and
then back again, etc.
This test is best suited for rounded objects and may be disappointing
with the sodium chloride crystals. The phenomenon may be improved by
using an objective of lower aperture, and a special objective equipped
with iris diaphragm is recommended for the investigation of very small
particles (88). The interpretation of the observed phenomenon must
consider the method of illumination in use.
44 Use of Optical Aids

With the condenser in high position so that it focuses above the prepara-
tion, the heavy shadow in the image of the object will move with the source
of the light (opening in the diaphragm) from the left to the right and will
face the shaded side of the field if the object has the higher refractive index
than the surrounding medium. (If the condenser is lowered to focus below
the preparation, the phenomenon is reversed and also less distinct.)
If the test is made with the mirror and without use of a condenser, the
bright spot or line in the image of the object will follow the motions of the
mirror from left to right if the object has the higher refractive index than
the surrounding medium.
Repeat the Becke test with the two other sodium chloride residues
after treating one with ethylene bromide (n = 1.54) and the other with
quinoline (n = 1.6) and applying cover slips.
The effect of ethylene bromide depends upon the purity of the reagent
and upon the temperature. As a rule, the outlines of the crystals will
still be visible when the intensity of illumination is reduced to a practical
minimum. If color fringes appear, identity of refraction may be assumed
for that part of the spectrum that does not show up in the fringes. If the
proper monochromatic light were used, the outlines would completely
disappear.
B. Use glass wool or obtain glass thread by drawing out a rod to a
diameter of first I mm and then 0.1 to 0.2 mm in the manner used for the
preparation of capillaries and capillary pipets from tubing, Expt. 19.
Mount a l-cm piece of fine glass thread in water between slide and
cover slip. Place the preparation on the stage so that the image of the
thread crosses the center of the field going from 6 to 12 o'clock. (Treating
the field of vision like the face of a clock, one finds it easy to specify locations
such as "two-thirds of the way from the center to eight thirty".)
Apply the Becke test and oblique illumination, and repeat this with
cedarwood oil (n = 1.515) for immersion liquid. Note that both tests
are satisfactory with this latter medium.
C. Aerate saliva in the mouth; transfer a droplet of the foaming liquid
to a slide, and cover it with a glass slip at once. The bubbles of air (n = 1.00)
retain their spherical shape long enough to permit use of the Becke test
and of oblique illumination.

Observation of Schlieren
TOEPLER defined schlieren (pr. shleeren) as regions of changing refraction
and devised a highly sensitive method for their detection by an arrangement
of diaphragms which eliminate the participation in the image formation
of those rays which pass through the carefully corrected central portions
of the lenses (170, 1014). Rather simple combinations of lenses and
diaphragms (606, 760, 1014, 1220, 1221) suffice to demonstrate the hot
Expt. I) Observation of Schlieren 45

air rising from a human hand, to detect inhomogeneity in a transparent

plastic sheet (1055), or to photograph the compression waves surrounding
a projectile flying through air. EMICH (1015, 1017, 1018, 1157) adopted
the technique for the study of chemical problems and demonstrated that
it offers a very sensitive means for detecting small differences in refraction,
density, or both and consequently for recognizing hidden circumstances
that bring such differences about: different composition, different concentra-
tion, temperature differences, diffusion or (and) chemical action affecting
concentration or temperature, etc.
Testing identity and purity are foremost among the analytical applica-
tions. To these ends, the least sensitive methods of schlieren observation
will usually suffice. Because of the effects of the particular impurities,
they will often allow to identify a certain batch of some "pure" solvent.
For such comparison, one permits a slow, fine stream of the unknown
liquid to enter a small sample of the standard which must have the same
temperature. If a suitable method of observation is used and the two
liquids differ sufficiently in refraction or dispersion, the stream of the
liquid becomes visible and the distribution of light and shade indicates
the medium of higher refractive index. At the same time, the direction
of flow shows which medium has the higher density. Miscible liquids of
the same optical density may give schlieren nevertheless as a consequence
of phenomena occurring in the zone of contact (chemical reaction or diffusion
leading to the establishment of concentration gradients), but this cannot
happen with substances which are closely similar such as isomers.
The visual method for the observation of schlieren (1155) used in the
following experiment permits detecting differences of 0.0001 to 0.0002 in
the refractive indices and in the specific gravities of two liquids. The
magnitude of the difference in refraction may be crudely estimated by
comparison with schlieren given by known refraction differences and
identical optical conditions (position of schlieren, eye, etc.). Observed
should be the width and the intensity of the shadow portion of the schlieren.
The differences may be expressed in empirical numbers by using some
simple device permitting measurement of that displacement of some part
of the optical system, which represents a sufficient deviation from the
most sensitive adjustment to render the schlieren invisible.

Experiment 5
Visual Observation of Schlieren
Cells. The cell, Fig. 7, formed at one end of a standard microscope slide,
may be less than 1 mm thick and may be given the outline of a V or of
an amphora to reduce the capacity to 0.1 ml and less. The glass parts may
be formed by cutting or grinding microscope slides. For use with aqueous
46 Use of Optical Aids Expt.5

solutions, they may be cemented together with Canada balsam or any

suitable plastic cement. For use with organic liquids, the parts may be
polished so that they form a tight cell when pressed together in a suitable
metallic frame. It is also possible to buy cells made entirely of glass.
Capillary Pipets. For the preparation see
Expts. 19 and 22. Determine the diameter of the
opening, 0.1 to 0.2 mm, as directed in Expt. 22.
The wide part should be 1 mm or more in
diameter.
Background. A suitable background is ob-
tained by mounting a frosted 60-watt bulb behind
a ground-glass plate, approximately 25 cm by
25 cm. The left half of the glass plate is made
opaque by covering with a metal sheet, a black
cardboard, or painting with black lacquer. It is
important that the dark part of the field is set
off by a sharp perpendicular boundary against
the bright half.
Procedure. Fast emergence from a wide tip
does not only waste sample, it also gives "false"
schlieren because of the friction which raises
the temperature of the flowing liquid. Thus,
select a pipet with a bore of the tip not exceeding
0.2 mm. Insert the tip into 1 % NH 4CI solution
and apply gentle suction with the mouth until
about 4 cm of the length of the wide capillary
are filled.
Stretch a rubber band around the upper part
of a clean glass cell as shown in Fig. 7, and with
a pipet add water to fill three fourth of the cell.
Mount the capillary pipet on the slide of the cell as
shown by the illustration, but do not insert the
tip of the pipet into the cell liquid. Place the
cell assembly for a few minutes upon a metal
Fig. 7. Observation of
Schlieren; n a t . size object (stand of a microscope) to make certain
that the liquids acquire the same temperature.
Grasp the cell at the upper left-hand corner of the slide and push the
capillary pipet down so that its tip enters the liquid in the cell and the
outflow starts. Because of the narrow bore of the tip of the pipet, the
flow of the ammonium chloride solution continues for several minutes
giving ample time for the observation of the phenomenon.
Hold the cell about 120 cm in front of the ground-glass plate so that
the perpendicular boundary of the dark half of the field is seen near to
 Use of Polarized Light 47

the opening of the pipet, Fig. 7. Note that the NH,Cl solution flows to
the bottom of the cell and that the shaded half of the schlieren faces away
from the dark half of the background. The phenomenon corresponds to
differences of 0.001 in refraction and 0.003 in density, both being higher
for the solution flowing from the pipet.
Clean and dry the cell as directed for centrifuge cones (p. 124) by washing
it with distilled water and then drawing air through it. Then repeat the
schlieren observation by using 1 % NH 4CI in the cell and water in the
capillary pipet. This time insert the opening of the pipet deeper into the
cell liquid than before. Note that the schlieren ascends with the shaded
part of it facing the dark half of the background.
Again clean the cell and repeat the experiment. This time use water
in both, the cell and the pipet; no schlieren should be observed.
Empirical data on schlieren strength may be obtained by mounting the
apparatus on a board of about 1.5-m length so that the relative positions
of lamp, background, cell, and eye are fixed (ll55). Mount the ground-glass
plate in a vertical position so that it may be moved sideways and its position
may be read off a scale. Locate the cell 120 em in front of the background,
and mount 25 em in front of the cell an eyepiece consisting of a blackened
tube with a vertical slit (I to 2 mm wide) at the far end. First adjust the
set-up to obtain maximum sensitivity for the detection of refraction
differences. To obtain estimates of schlieren strength, move the ground-
glass plate to the left until the schlieren disappears (dark shadow close
to the opening of the pipet vanishes) and then without delay to the right
until the schlieren again becomes invisibl~ and record the distance between
the two points. Absolute data may be obtained by preparing a calibration
curve with liquids of known refractive indices.

Use of Polarized Light

All use of polarized light is based upon the fact that properties of
anisotropic matter (all crystallized matter except cubic crystals; cubic
crystals and glasses under strain; and liquids having the molecules more
or less uniformly oriented) are a function of the direction of action with
regard to the internal structure. When the interdependence of kind of
substance, internal structure, and variability of properties is once known,
a study of the latter may be used to get information on the internal structure
and on the chemical nature.
The action of anisotropic matter upon light may be briefly summarized
as follows. In most directions, only plane polarized light is transmitted,
and only two certain planes of vibration which always are vertical to one
another are available. Both planes of vibration may be used simultaneously,
and light entering the anisotropic material is resolved into the two com-
48 Use of Optical Aids

ponents using the two available planes of vibration in that ratio which
makes the most efficient use of the nature of the incoming light; ordinary
light and plane polarized light vibrating at an angle of 45 degrees to both
offered planes of vibration are distributed with equal intensity; plane
polarized light which has one of the possible planes of vibration continues
through the anisotropic material with that plane of vibration, and the
other component will not form; plane polarized light with the plane of
vibration inclined at an angle different from zero and 45 to the offered
planes of vibration is very unequally distributed so as to favor that plane
which is closer to the plane of vibration of the incoming light. As a rule,
the anisotropic material acts more or less differently upon the two com-
ponents of light, vibrating in planes perpendicular to one another, even
when they travel along the same path. As a rule, the two components
travel with different speed, which implies different refractive indices,
different wave lengths, and acquisition of a phase difference on the path
through the anisotropic matter; the two components may also experience
different selective absorption and acquire different "color".
Any anisotropic material produces a mixture of equal parts of two
components of plane polarized light when illuminated with ordinary light.
Polarizing equipment, regardless whether it is used for the investigation
of large objects or whether it is combined with a microscope for the study
of small objects, always has the purpose of getting plane polarized light
of just one plane of vibration, and this may be utilized in two ways. Ob-
viously it may serve for isolating either one component of plane polarized
light emerging from the object so that the properties of the object (refraction,
absorption) may be studied in the light of one component at a time
(determination of refractive index, observation of pleochroism). Most
frequently it is used for combining the two components of plane polarized
light emerging from the object so that they assume identical planes of
vibration and consequently interfere. The phenomena resulting from
interference permit conclusions concerning type and orientation of the
structure of the object.
In principle, the polarizing apparatus proper is extremely simple. Its
two parts, the polarizer and the analyz~r, are identical and interchangeable
in essence. Both are device~ for obtaining plane polarized light of only
one plane of vibration and usually consist of some anisotropic medium
(crystal or aggregate of crystals having identical orientation) that unavoid-
ably converts ordinary light into two components of plane polarized light,
but has the exceptional property of removing one of the components by
either turning it away on a different path (Nicol prism or nicol for short)
or by absorbing it (various types of Polaroid). One of these devices (polars)
is called polarizer and is mounted below the stage to supply to the prepara-
tion plane polarized light with one definite plane of vibration. The other
 Use of Polarized Light 49

is called analyzer and is mounted above the preparation to get all light
emerging from the preparation to vibrate in one prescribed plane.
Any conventional microscope may be quite readily converted to a
polarizing microscope with the use of accessories having Polaroid films
and retardation plates (mica or selenite) in circular mountings between
glass plates. The polarizer may be inserted into the stage aperture or into
a slot above or beneath the substage diaphragm. The (cap) analyzer is
usually placed upon the top of the eyepiece. Retardation plates may be
obtained in mountings that fit on top of the polarizer mount. A substitute
for a rotating stage may be obtained by fitting a circular (graduated) disk
around a short piece of metal tubing that slips into the opening of the
stationary. microscope stage.
A polarizing microscope (petrographic microscope) may use Nicol prisms
for polarizer and analyzer. The polarizer is usually mounted below the
diaphragm of the substage so that it may be rotated around the optical
axis of the microscope. The direction of the plane of vibration (or that
perpendicular to it) is usually indicated upon the mounting, and the 6 to
12 o'clock position of that direction may be indicated by a stud falling
into a notch of the mounting. The analyzer is frequently mounted above
the objective so that it may be quickly slid in and out of the tube of the
microscope. Below it, a slot is provided in the tube for the insertion of
retardation plates and quartz wedge. The rotating stage is graduated
and provision is made for aligning axis of rotation of the stage with the
optic axis determined by the objective. The centering may be done with
the stage, with the objective changing device, or with individual objectives,
and an objective changing device is used, wmch assures that objectives are
always returned into the same position.
Essential for a polarizing microscope is furthermore that the cross-hair
eyepieces carry a stud to engage in a notch of the draw-tube so that the
hairs indicate the directions 6 to 12 and 3 to 9 o'clock, which are also
the directions of vibration in polarizer and analyzer in the positions parallel
and crossed. Obviously, the draw tube must be keyed to the body tube
so that it cannot be rotated in the latter. Customarily, a divisible Abbe
condenser with iris diaphragm is provided so that it may be focused
mechanically or completely swung aside together with the polarizer.
A Bertrand lens, an auxiliary lens for the observation of axial figures,
is frequently found mounted in a metal frame so that it may be slid into
the body tube above the analyzer.
The interpretation of the phenomena observed with a polarizing micro-
scope easily leads to wrong conclusions and will give a meager yield if not
based upon a rather thorough study of crystal optics (89, 93, 96, 98,
107, 110).
Be~I.Plchler. Identification
50 Use of Optical Aids Expt. fj.

Experiment 6
Teating and Adjusting a Polarizing Microscope
The inspection of the stand and the microscope proper has been treated
in Expt. l. The following additional tests apply to the polarizing equipment.
and accessories.
l. Approximate Vibration Directions 01 Polarizer and Analyzer. The
following procedure is frequently needed just for checking the meaning'
of marks on the mountings.
Take either the analyzer or the polarizer out of the microscope, hold
it close to the eye, and observe through it the light reflected from some·
plane horizontal surface of dark color such as a black glass plate, table
top, or some surface coated with black enamel or lacquer. Most of the
reflected light is plane polarized and vibrates horizontal, i. e., in a direction.
parallel to the reflecting surface. Rotate the polarizing device to find
the position in which it transmits least of the reflected light. In this.
position, the diameter produced by a perpendicular cut through the·
mounting represents the vibration direction of the transmitted polarized
light. (The horizontal diameter represents the vibration direction of the-
component swallowed by the device.)
2. Action 01 Polarizer and Analyzer. Place upon the stage of the micro-·
scope a slide with a cover slip on it. Focus upon the edge of the covE,lr slip'
and adjust the illumination to get bright illumination of the whole field
when both, analyzer and polarizer, are removed. Insert the polarizer below
the stage so that its vibration direction goes from 6 to 12 o'clock (in micro-
projection, it may be preferable to rotate the polarizer into the position
in which it will transmit any plane polarized light already obtained by
reflection on the surface of a mirror or prism). As a rule, it is impossible
to perceive that the intensity of the illumination is cut into half by the
insertion of the polarizer.
Insert the analyzer and rotate it (or if this cannot be done, rotate·
the polarizer) until no more light is transmitted and the field appears.
dark when looking into the eyepiece. This happens when the light.
transmitted by one polarizing device has that plane of vibration which
makes it the component that is swallowed up by the other. The planes.
of vibrations of the components that are transmitted by the devices are
crossed at right angles.
The vibration direction of one polar is already known; that of the other
must now be perpendicular to it. If the adjustments are properly carried
out, the vibration directions should be parallel to the cross hairs in the
eyepiece (6 to 12 and 3 to 9 o'clock). A more precise test follows.
under (6).
Expt.6 Use of Polarized Light 51

The position parallel is obtained by rotating the analyzer or, if this

cannot be done, the polarizer through 90 degrees so that the vibration
directions of their transmitted components arrive in the same plane.
Rotate one of the polars a few times through 360 degrees and observe
that the positions of minimum and maximum brightness corresponding
to crossed and parallel occur twice during a full tum and that the blackening
out occurs quite suddenly and at a precisely adjustable position, whereai
the position of maximum brightness is only vaguely defined. If the rotating
device has a graduated scale, record the average position of minimum
brightness and its standard error computed from several settings. If the
analyzer is in fixed position and the polarizer is being rotated, minimum
brightness should occur when the stud clicks into the notch (readings zero
and 180). If the polarizer is set to zero and the analyzer is rotated, darkness
should occur at 90 and 270 if polarizer and analyzer circles have the zero
at 12 o'clock.
Test all objectives by focusing on the edge of the cover slip in a bright
field and then crossing the polars to the position of minimum brightness.
Even with very strong illumination, the field should become reasonably
and evenly dark. If fringes or spots of color or brightness are noticed,
move the preparation, rotate the objective (and move the lenses of the
condenser if this is possible) in an attempt to find the source of the imperfec-
tion. Slides and lenses may become anisotropic due to mechanical streslil
or heating and rapid cooling. A small amount of light is always derived
from the partial depolarization on refraction and is most noticeable with
high-power objectives having several strongly curved lenses (88).
3. Centering. Add some glyceroi to the cover slip to hold it in place.
Remove the analyzer. With a cross-hair eyepiece in place, focus a comer
of the cover slip with the objective having the lowest magnification. Move
the slide to get the point of the comer to coincide with the intersection
of the cross hairs and, if necessary, use a stage clip to hold the slide in place.
While looking into the microscope, rotate the stage. To this end, cup the
hands around the stage so that they make contact with its outer edge
but do not touch the top surface of the stage. Estimate the center of the
circle which is described by the image of the point of the cover slip and
mark it by moving the slide until the point of the slip is located there.
Using only the centering screws, bring the point to the intersection of
the cross hairs. Again test by rotating the stage and repeat the process
of adjusting and testing until the mark represented by the comer of the cover
glass remains at the intersection of the cross hairs when the stage is rotated.
If the centering is done at the objective, repeat it for each objective
belonging to the microscope. If the centering is done at the stage or at
the tube of the microscope, repeat it with the strongest objective and convince
yourself that the setting is satisfactory for the objectives of lower power also.
,.
52 Use of Optical Aids Expt.7

Finally check the centration of the condenser with the top lens in
place and the strongest objective as directed in Expt. 1.
4. Checking the Graduation of the Stage. Align one edge of the cover slip
with one of the cross hairs. Use a stage clip if necessary. Read the position
of the stage, and then rotate the stage to turn the image upside down so
that the edge is again lined up with the cross hair. Again read the position
of the stage, which should have changed by 180 degrees. If it appears
advisable, repeat the test by using another part of the scale (turn the slide).
5. Checking the Angle of the Cross Hairs. Align the edge of the cover
glass first with one cross hair, and read the position of the stage. Rotate
the stage until the edge is lined up with the other cross hair, and again
read the position. The readings should differ by 90 degrees.
6. Checking the Position of the Cross Hairs. The cross hairs are supposed
to indicate the vibration directions in the polarizing devices in the positions
crossed and parallel.
Place a large drop of nitrobenzene upon the center of a microscope slide.
Add a small amount of anthraquinone which must be crystallized as fine
orthorhombic needles. Stir and place a cover slip upon the mixture. Use
an objective of magnification lO and focus first upon the edge of the cover
slip; then select a perfect, long needle, and move the slide until one of its
long edges goes through the intersection of the cross hairs.
Secure strong, even illumination of the field. Insert the analyzer, and
carefully cross analyzer and polarizer so that their vibration directions
presumably take the marked positions (click) 6 to 12 and 3 to 9 o'clock,
while the field acquires minimum brightness. Rotate the stage while cupping
the hands around it until the needle is in the position of minimum brightness
(appears as dark as the field). Then turn or remove the analyzer and
determine the angle formed by edge of the needle and the cross hair.
Return the analyzer into the crossed position, turn the stage through
90 degrees, and again bring the needle in the position of minimum brightness
with the hands cupped around the stage. Turn or remove the analyzer
and observe the inclination of the edge of the needle to the second cross hair.
Both times, the cross hairs should be closely parallel to the edge of the
crystal. Otherwise they do not correctly indicate the vibration directions
of the polarizing devices and must be aligned by either rotating the cross
hairs or both nicols.
Reserve the preparation for Expt. 10.

Experiment 7
Isotropic and Anisotropic Substances in Polarized Light (96, 1l0)
Red first order selenite (gypsum) plate, quartz wedge as supplied with
polarizing microscopes. Substitutes may be made of thin cellophane sheet:
cut a ribbon, 5 to 10 mm wide, and cut this into squares; mount between slides
Expt.7 Use of Polarized Light 53

and cover slips (a) a sandwich of three or four squares in the same orientation
and slightly displaced so that one side forms a stair with steps of less than I-mm
width, (b) two squares, one rotated against the other through 90 degrees, side
by side, but slightly overlapping to give a double layer of less than I-mm width_-
Crystals of Na2 S20 a • 5 H 2 0.
Using the objective of lowest power and a cross-hair eyepiece, remove
the analyzer and focus upon an edge of the gypsum (monoclinic, for D line:
ex = 1.5205, (J = 1.5226, ?' = 1.5296) plate or the cellophane square~
Note the relation of this edge to the vibration direction of the slower
component of light usually indicated upon the mounting of the selenite
plate (if the mounting does not permit focusing upon an edge of the selenite
plate, use India ink or some marking pen to dra'Y a parallel to the direction
indicated upon the glass cover cemented over the selenite plate and focus
upon this line).
Insert the analyzer and cross the polars. Rotate the stage and confirm
that the anisotropic gypsum plate is always the same hue of red excepting
when in the four positions of extinction ocurring during a full tum of
360 degrees. Note that the parts of the field that are not taken up by
the gypsum plate and contain only isotropic material (glass, immersion
liquid, clear cement, air) always remain dark. Confirm that, in the positions
of complete extinction, the cross hairs indicating the vibration directions
in polarizer and analyzer also indicate the vibration directions in the specimen,
the gypsum plate (the one marked upon the mounting and the other
vertical to it). Try to explain why darkness must prevail with crossed
polarizers when the specimen is isotropic or when an anisotropic specimen
has the vibration directions coinciding with those of the polars; why is
there brightness and sometimes color when an anisotropic specimen is not
in the extinction position 1 Why is the hue of the color always the same
when the stage is rotated 1
Make the polars parallel and again rotate the stage. Observe that the
phenomena are complementary to those observed with crossed polars:
the isotropic media remain always bright and appear in the same color
as in ordinary light; the gypsum plate is colorless bright in the position
of extinction and green in the 45 degree position (its vibration directions at
45 degrees to those of the polars). Note that the green color fades rapidly
as soon as the specimen is rotated out of the 45 degree position and that
the plate is white most of the time when the stage is rotated. Try to explain
this as well as the green color in the 45-degree position.
Substitute the quartz wedge (stair) for the gypsum plate and, using
crossed polars, select a portion that shows bands of different color. Repeat.
the procedure followed with the selenite plate as directed above. Realize
that the differences in color must be due to differences in thickness since
all other conditions are equal. Note that all colors retain their hue during
54 Use of Optical Aids Expt.7

the rotation of the stage and that all of them change to complementary
colors when proceeding from crossed to parallel polars. Also observe that
the colors are in general more brilliant with crossed than with parallel polars.
Cross the polars and bring the quartz wedge into a 45 degree position.
Keeping it in this position, push it through the field of vision while observing
through the microscope. Pay attention to the succession of colors. Note
that (a) the thin edge appears black; (b) the appearance changes to gray,
lavender, yellow, orange, red, violet, blue, green, yellow, orange, red, etc.
as the thickness increases; (c) that the series of colors is repeated, but that
the reds, blues, etc. of the successive series (orders) differ in hue so that
a certain color and hue corresponds to a definite thickness; (d) that the
colors lose purity and brilliance and assume pastel hues as order and
thickness increase; (e) that it is possible to distinguish and recognize the
colors of the various orders either by memory or by comparison with the
color series of a wedge.
Make the polars parallel and repeat the observation of the succession
of colors (it may be easier to keep the wedge in the 45 degree position by
using the slot in the body tube). Note that the series starts at the thin
edge with white, yellow, orange, and red of first order and then continues
as with crossed polars.
Bring the band of the green of second order, i. e., the first green from
the thin end, into the center of the field and then change to crossed polars.
Confirm that the red obtained as complementary color is the red of first
order by slowly moving the thin edge of the wedge into the field and
observing the succession of colors. Note that crossed polars require a
thicker specimen (longer path through the specimen) than parallel polarizers
to produce the same interference color, and that the sequence of color bands
moves closer to the thin edge when changing from crossed to parallel.
Repeat all the experiments with the quartz wedge by using mono-
chromatic light (sodium lamp or color filter) for illumination, and try to
explain the phenomena. Observe that the dark bands move toward the
thin edge when changing from crossed to parallel. Change again to white
light for the last part of the experiment, which follows.
Place upon the center of a slide a small clear crystal of Na 2 S 20 3 • 5 H 20
and fuse it by placing the slide upon a steam bath. Without delay, place
a cover slip on the drop of melt. The melt should fill the entire space
between the glass surfaces. If too much thiosulfate was used, remove the
excess by touching the edge of filter paper to the edge of the cover slip
until only a very thin film of liquid is left. Start crystallization by touching
the melt at the edge of the cover glass with a needle that has been in contact
with a crystal of the pentahydrate.
Remove the analyzer and use an objective of power 10 for focusing
the preparation. Note that the outlines of the thin monoclinic plates are
Expt.8 Use of Polarized Light 55

not very distinct and are much better seen with polarized light. Use
<lrossed polars and rotate the stage. Note that: (a) air bubbles enclosed
by the crystals remain always black; (b) each plate retains its color during
the revolution of the stage; (c) not all plates have the same color in spite
of the fact that they consist of the same materials and must have
the same thickness; explanation? (d) that extinction does not occur
simultaneously for all crystals in the field since they are differently oriented.
Observe the inclination of the long edges of the plates to the cross hairs
when they are in the position of extinction, and repeat the observation
with different sets of plates. Search for a center (point of seeding 1) from
which crystals radiate in all directions, and bring it into the center of the
field. Rotate the stage. Observe and explain the more or less stationary
black cross. Prepare a crude pencil sketch of some part of the preparation,
and indicate by arrows the vibration directions in the various crystals.
Select a very colorful portion of the preparation, and then rotate the
analyzer and observe the change to complementary colors when going
from crossed to parallel and back. Finally place the analyzer into position
parallel, rotate the stage, and observe that the color display is far less
dazzling than with crossed polarizers.
Save the preparation for Expts. 8 and 13.
If a spectroscopic eyepiece is available, use it to analyze the compositions
of colors of increasing order as exhibited by the quartz wedge and also of
crystals showing the white of higher order. For an alternative, the image
of the crystals may be projected onto the slit of a standard spectroscope (1283).

Experiment 8
Determination of Vibration Directions in Relation to Profile (96, nO)
Three I-inch watch glasses, filter paper; saturated solutions of Na 2 S20 S '
. 5 H 20, BaCI2 • 2 H 20, and CuS0 4 • 5 H 20.
Of each of the three reagent solutions take one drop of 10 #1 = 0.01 mI,
transfer it to the center of a microscope slide, and spread it to a circle
of not more than l-cm diameter. Put the slides aside, but inspect them
from time to time. When crystals begin to separate, start observing them
with the lowest magnification available. When well developed crystals
of 0.1- to I-mm length or diameter have been obtained, stop the evaporation
by inverting small watch glasses over the drops or by drawing off the mother
liquor by touching filter paper to it. The latter alternative has the dis-
advantage that annoying crusts of tiny crystals may form on evaporation
of residual liquid. These may be dissolved by breathing upon the slide
and quickly covering the residue with a small watch glass.
The outcome of crystallization experiments is quite unpredictable,
but one may hope that some suitably developed crystals will be obtained.
56 Use of Optical Aids Expt.8

If crystallization is too copious or too rapid, dilute some of each solution

with an equal volume of water, and repeat the crystallization with drops of
the diluted solutions.
Perform the following procedure with all preparations which appear
promising and try it also with the thiosulfate crystals of Expt. 7.
Remove the analyzer and use the lowest available magnification to
start the search for suitably developed crystals. Large thin plates with
clearly defined edges are most desirable.
When a crystal has been selected, move the slide to bring it into. the
center of the field. If necessary, use a stronger objective so that the out-
standing dimension of the crystal becomes more than one-fourth of the
diameter of the field. Then use the polarizing equipment in position crossed,
and rotate the stage to bring the crystal in the position of complete extinc-
tion. Remove the analyzer and observe the position of the outstanding
dimension of the crystal (long edge, outstanding diagonal) in relation to
the cross hairs which give the vibration directions for the given view of
the crystal.
The extinction is called parallel (symmetric), if the outstanding dimension
(edge or diagonal) is parallel and vertical to the cross hairs. It is oblique,
if the outstanding dimension is inclined to the cross hairs. The extinction
angle is the angle enclosed by the outstanding direction and the nearest
cross hair.
Try to record all pertinent facts without worrying about the correct
crystallographic interpretation. Prepare a sketch of the crystal and indicate
the vibration directions. Record all angle measurements in the sketch.
To measure a silhouette or profile angle, move the slide to bring the
point of the angle to the intersection of the cross hairs. Rotate the stage
to align one side of the angle with one of the cross hairs and read the
position of the stage with the use of the vernier; it is preferable to have the
point of the angle slightly off the center of revolution so that the lines
become parallel when they are very close to one another without actually
coinciding. Record the position and then rotate the stage so that the
aligned cross hair sweeps through the angle. Align it with the other side
of the angle in like manner as before, and again record the position of the
stage. Repeat the process and compute averages for both readings. Their
difference represents the angle.
Determine the angle of extinction as follows. Move the slide so that
the outstanding direction of the crystal passes through the intersection
of the cross hairs and is parallel and very close to one of the cross hairs.
Read and record the position of the stage. Cross polarizer and analyzer
and rotate the stage with the hands cupped around it until the crystal
is in the position of extinction that is closest to the starting position.
Again read the position of the stage. Remove the analyzer, return to the
Expt.9 Use of Polarized Light 57

starting position, and repeat the determination. The difference of the

averages of the readings represents the extinction angle which is recorded
in the sketch.
If possible, repeat the angle measurements with several crystals of each
substance and record the results with the use of drawings. Occasional large
discrepancies should not surprise. Profile angles depend upon the orientation
of the crystal, and the same holds for extinction angles. The true out-
standing directions may be obscured by irregular development of crystals.
If a rotating stage is not available, angle measurements may be performed
with a goniometer eyepiece, or the image may be projected or photographed
and a protractor used on the screen image, tracings, or a (possibly enlarged)
photomicrograph.
Save all preparations containing well-shaped large crystals. Na 2 S 20 a ·
. 5 H 20 and BaCI 2 • 2 H 2 0 are monoclinic; CuS0 4 • 5 H 20 is triclinic.

Experiment 9
Use of Compensators (96, 1l0)
First-order red gypsum plate; quarter wave length mica plate; quartz wedge.-
Saturated solutions of NH H 2PO and NH CIO ; the latter may be pr-!pared
by bringing together solutions of equivalent amounts of Ba(CIO )2 and (NH )2S0~.
Proceed as in Experiment 8 to get well-developed crystals of NH 4 H 2P0 4
and NH 4CI0 4 •
The first part of the experiment requires superimposing the quartz
wedge upon a retardation (gypsum or mica) plate. If the microscope tube
has a slot, use it for the quartz wedge and place the retardation plate
upon the stage; if not, try to insert the retardation plate into the substage
and use the wedge upon the stage; if this too cannot be done, clamp the
retardation plate to the stage and use the wedge on top of it.
Cross the polars and focus on the top surface of the device upon the
stage. Take the quartz wedge out of the path of light and rotate the
gypsum plate into that 45-degree position of maximum brightness for
which the slower component (arrow) vibrates in the direction 7: 30 to
1: 30 o'clock. Superimpose the quartz wedge with the same vibration
direction of the slower component; with the thin edge leading, advance
it slowly into the field and observe the Rhift of the color bands. If it. is
possible, rearrange the position of the gypsum plate so that it occupies
only the upper half of the field and the other half of the field shows
the color bands as they occur in the wedge without the addition of the
gypsum plate. Note that the adding of the gypsum plate, which by itself
produces the red of first order, advances the colors of the wedge by one
order. The gypsum plate adds to the thickness of the wedge so that its
very edge shows already the red of first order.
58 Use of Optical Aids Expt.9

Rotate the gypsum plate through 90 degrees so that the vibration

direction of its slower component is parallel to the vibration direction of
the faster component in the wedge. If possible, again move the gypsum
plate so that it appears only in the upper half of the field. Note
that the action of the gypsum plate subtracts from the thickness of the
wedge. Where the red of first order is located in the wedge, the combination
gives a black band indicating zero thickness or complete compensation.
The optical thickness increases on both sides of the black band. Going
toward the thick end of the wedge, the colors of first, second, third order
appear displaced by the space of one order; toward the thin edge, the
optical effect of the gypsum plate increases from the gray to the red of
first order, which latter appears at the edge where the effect of the quartz
has become zero.
Obviously, the quartz wedge may be used to determine the order of
an interference color by superimposing it upon the specimen so that it
compensates and then advancing it until the compensation becomes complete
and the specimen blacks out. This will happen at that point where the
optical effect (thickness) of the wedge is equal to that of the specimen and,
consequently, where the wedge exhibits the same color as the specimen.
It then remains to identify the order of the color by counting the number
of orders (red bands) between the color giving complete compensation
and the thin edge of the compensator. The experiment also shows that
the gypsum plate exhibits the red of first order and may serve as a correspond-
ing standard of retardation or optical thickness.
The quartz wedge as well as any other anisotropic plate (gypsum, mica)
for which the vibration direction of the slower component is known may
also serve to identify the vibration direction of the slower component in
any specimen by finding whether superimposition with vibration directions
parallel gives compensation or increases the optical thickness. The wedge
is generally applicable; the usefulness of other compensators depends
upon the thickness (interference color) of the specimen.
If necessary, restore the combination of wedge and gypsum plate
showing the black band of complete compensation. Then rotate the analyzer
into the position parallel and study the effect.
Repeat the experiment by using the combination of quarter wave length
mica plate and quartz wedge; USe first crossed, then parallel polars. Note
the shift of colors in the wedge when the mica plate is added in positions
of addition and subtraction (compensation). Confirm that the mica plate
gives a gray of first order.
Using the analyzer in position crossed, superimpose the gypsum and
mica plates. If possible, use one upon the stage and the other in the slot
of the tube or in the substage; also try to arrange that the plate upon the
stage occupies only half of the field of vision. Focus upon the preparation
Expt.9 Use of Polarized Light 59

upon the stage, and rotate the stage so that the (standard) retardation
plate upon it obtains the 45-degree position of maximum brightness.
Adjust its position so that it occupies half the field, and then superimpose
the other plate so that the vibration directions of the slower components
are parallel. Observe that the combination produces blue (of second
order = first order red + first order gray). Rotate the analyzer to parallel,
note the effect, and return it to crossed. Rotate the stage through 90 degrees
to obtain compensation. Observe that the combination gives yellow (first
order = first order red - first order gray). Rotate the analyzer to
parallel.
From the experiment recognize the truth of the following statements.
If the specimen is thin, i. e., gives by itself between crossed polars a gray
of first order, bring it into the 45-degree position of maximum brightness
and superimpose a first order red plate (gypsum, also called selenite).
It will, because of its small optical effect, change the interference color
of the thick plate only slightly, from first order red to either second order
blue (addition) or to first order yellow-orange (compensation). There
cannot be any doubt concerning the meaning of the color change, and
the directions of the slow and fast components in the specimen follow
from the known directions in the first order red plate.
If the specimen is thick, i. e., gives by itself between crossed nicols a
color, use a quarter wave length (mica) plate. The very thin (standard)
retardation plate can change the interference color of the specimen only
little, just to the adjacent colors, and the decision whether addition or
compensation has taken place will offer no difficulty.
Obviously, use of a compensating wedge of widely varying thickness
is most promising if the specimen is quite thick and exhibits the white
of higher order.
Practice the identification of interference colors and the establishment
of the vibration directions of slow and fast components with crystals
showing between crossed nicols the gray of first order, such showing colors,
and possibly some showing the white of higher order. Use crystals of
NH 4CI0 4 (orthorhombic plates, for D line: IX = 1.4818, {3 = 1.4833,
Y = 1.4881), NH 4H 2P0 4 (tetragonal combinations of prism and short
bipyramid which latter may show the color fringes of a wedge; for D line:
8 = 1.4792, w = 1.5246), Na 2 S 20 3 • 5 H 20 (monoclinic prisms, for D line:

IX = 1.4886, {3 = 1.5079, Y = 1.5360), BaCI 2 • 2 H 2 0 (monoclinic octagonal

tablets, IX = 1.63, {3 = 1.65, Y = 1.66), and CuS0 4 ' 5 H 20 (triclinic, short

prismatic, IX = 1.51, {3 = 1.537, Y = 1.543). Prepare drawings and indicate
the directions of fast and slow components. Save all preparations containing
well developed crystals for Expt. 13.
Place the quartz wedge upon the stage of the microscope and inspect
the band of first order red with crossed polars and the wedge in the 45-degree
60 Use of Optical Aids Expt.l0

position. Observe that the characteristic purple band is quite narrow so

that a slight variation in optical thickness suffices to change the color to
violet-blue or orange-yellow. For this reason, the red of first order is also
known as the sensitive tint, and the first order red plate is used for the
confirmation of birefringency (anisotropism) in instances where the faint
brightening of the object might be due to stray light.
Focus a preparation (NH 4C10 4 ) that contains crystals which appear
very dark gray between crossed polars. Cup the hands around the stage
or use a black tent around it, and move a crystal into the center of the
field, that barely brightens up when the stage is rotated. Insert the gypsum
plate in the 45-degree position, and again rotate the stage.
Obviously, the first order red plate may also be used in demonstrations
by microprojection to enliven the appearance of specimens showing mostly
grays of first order or to obtain light for observation of manipulations in
the isotropic parts of specimens.

Experiment 10
On the Determination of the Refractive Indices of Anisotropic
Materials (96, 110)
Use the slide prepared in Expt. 6: orthorhombic needles of anthraquinone
in nitrobenzene, nn at 20° C = 1.5520.
Remove analyzer and polarizer, and focus upon the edge of the cover
slip with an objective of magnification 10 or higher. Select a long needle
with light outlines, and move the slide so that one of its long edges goes
through the intersection of the cross hairs. Rotate the stage and observe
that the outlines of the needle do not change their distinctness. Try the
Becke test as outlined in Expt. 4.
Insert the polarizer or the analyzer (or both in parallel orientation)
so that the vibration direction of the transmitted light is 6 to 12 o'clock.
Illuminate the field brightly, rotate the stage, and observe that the outlines
of the needle nearly disappear when its long edge is parallel to the cross
hair 6 to 12 o'clock and become strongest in a position vertical to this.
Explain the phenomenon and use crossed polarizers to confirm that these
two positions of the needle correspond to extinction positions. Why is it
sufficient to use either the polarizer or the analyzer 1
Return to the use of one polar giving light of the vibration direction
6 to 12 o'clock. Apply the Becke test in both positions of extinction;
use monochromatic light if color fringes should be observed. Make a
drawing of the needle, and provide the vibration directions with estimates
of the corresponding indices of refraction. Try to confirm the finding
with the use of a suitable compensator (preceding Expt.).
Expt. 11, 12 Use of Polarized Light 61

If a preparation with well developed crystals of NH 4 H 2P0 4 is available,

use it to repeat the experiment after applying benzene as immersion liquid
and a cover slip.

Experiment 11
Observation of Pleochroism (96, 110)
Benzoquinone, solid; 5% solution of hydroquinone in ethanol; 5% solution
of resorcinol in eth:mol.
Take two slides and deposit upon the center of each about 0.5 mg
(0.5,u1) benzoquinone in the form of several small particles. Treat one
deposit with 1 drop (10,ul) of the hydro quinone solution and the other
with a like volume of the resorcinol solution. Allow the solutions to spread
and evaporate.
First investigate the residue containing resorcinol quinhydrone. Use
one polar to give white light of vibration direction 6 to 12 o'clock. Use a
bright field and low magnification to find crystals and aggregates (feathery
or reminiscent of oriental rug designs) that exhibit a striking color change
when the stage is rotated.
For study, select a well developed single crystal. Bring it into the center
of the field and change the objective according to need for higher magnifica-
tion. Be certain to obtain a brightly illuminated field (top lens of condenser
swung in). Prepare a drawing of the crystal; (using crossed polars) determine
the vibration directions and then (with one polar) the colors belonging to
them, and enter the information into the drawing. If a spectroscopic
eyepiece is available, use it to get the spectra of the two colors. For an
alternative, the image of the crystal may be projected onto the slit of a
standard spectroscope (1283).
Repeat the experiment with the hydro quinone quinhydrone. The
preparations evaporate within a few days.

Experiment 12
Observation of Axial Figures (96,110)
Objective, 4 mm; 2.5 cm X 2.5 cm square of mica (muscovite) sheet, more
than 0.5 mm thick; iodoform, solid; xylene.
In a test tube, prepare about 2 ml of a saturated solution of iodoform
in xylene. Place a large drop, 0.5 ml, of the solution on a clean 3-inch
watch glass and put aside for slow evaporation. If necessary, retard the
evaporation by partly covering with a second watch glass. Inspect from
time to time without disturbing the solution. Try to get the hexagonal
plates as large as possible, but remove the mother liquor by taking it up
with filter paper before parts of the plates project above the liquid. Leave
the crystals upon the watch glass.
62 Use of Optical Aids Expt.12

Place the mica sheet (monoclinic, for D line: (X = 1.561; f3 = 1.590,

'Y = 1.594) upon the stage, remove the analyzer, and focus upon the top
surface with an objective of magnification 20. Use white light, the plane
mirror, and focus with the condenser (top lens in place) to fill the aperture
of the objective. Cross the polars and rotate the stage to a (45-degree)
position of maximum brightness. Try the three methods for the observation
of the axial figure:
a) LASSAULX. Remove the eyepiece and look down through the body
tube at the back aperture of the objective. If the aperture is not filled with
light, try opening the substage diaphragm, using the concave mirror,
focusing the condenser, and using the condenser without top lens. When
best conditions are obtained, rotate the stage and observe that a black
cross opens to form two hyperbolas.
b) BERTRAND. Replace the eyepiece, insert the Bertrand lens into the
tube, and focus upon the back focal plane of the objective by m.oving the
draw tube up and down. Rotate the stage.
c) KLEIN. Remove the Bertrand lens. Examine the Ramsden disk
with unaided eye or (better) with a magnifying lens. Rotate the stage.
Repeat the experiment first using a magnification 40 and then a
magnification 10 objective. Each time, adjust the illumination accordingly.
Repeat the experiment with the microscope focused upon the surfaoe
of a hexagonal tablet of iodoform (hexagonal; for D line: e = 1.750,
co = 1.800), which should be large enough to occupy nearly the whole
field of vision, better all of it. Naturally, this also depends upon the
objective magnification. Observe that one is viewing along the principal
axis (and axis of isotropism) of the crystal which remains black during
the rotation of the stage when crossed nicols are used. Note that the black
cross of the axial figure does not open up when the stage is rotated. Determine
the sign of double refraction by inserting a first order red plate. Note
the appearance of yellow and blue patches in the four quadrants close
to the center of the figure (near the intersection of the purple cross).
Imagine a line drawn through the two yellow patches. If this line is
vertical to the vibration direction of the slower component in the compen-
sator, the specimen is positively birefringent (+); if the directions are
parallel, negatively birefringent.
Flash figures of a uniaxial substance may be observed with NH 4 H 2P0 4
if large crystals have been obtained. If desired, further experiments may
be found described in the literature (88). The interpretation of biaxial
interference figures requires much practice and study (96, llO). The
observation of interference figures on small crystals or grains requires
careful centering of the whole optical equipment and of the specimen as
well as additional field diaphragms to narrow the field of vision so that
it is completely occupied by the specimen (88).
Expt.13 Use of Polarized Light 63

Experiment 13
Transition Phenomena in Polarized Light (159, 160, 163)
Narrow slides which may be obtained by lengthwise cutting a standard
microscope slide to get two or three strips of 8- to 12-mm width. - Solids:
NH4NOs, Na 2 S20 s ' 5 H 20, thymol, ethyl ester of p-azoxybenzoic acid (or some
other substance that gives a crystalline liquid).
Narrow slides are preferable when strong heating is required since
they are less liable to crack. Put into the center of such a slide about
0.5 mg (0.5,ttl) of thymol (e = 1.609, w = 1.525 for line D; m. pt. 49.6 0 C),
and place a half of a cover slip over it. Heat by moving the slide back-and-
forth about 5 mm above a small microflame or pilotflame of a Bunsen burner
until the thymol melts. The melt fills the space between slide and cover
slip and, as a rule, remains supercooled and does not crystallize. Remove
any excess of melt by taking it up with the edge of absorbent filter paper;
to obtain a very thin film of liquid, place the slide on paper and press down
upon the cover slip while touching the edge of the absorbent paper to the
edge of the cover glass.
Proceed in the same manner to obtain preparations with NH 4NO s and
the benzoic ester. These substances crystallize when the slide cools. Depend-
ing upon the volatility of the substance, preparations of this type will
keep for months or years and may be used for many repetitions of the
experiments.
A. Place the melt of thymol upon the stage, remove the analyzer,
and with objective magnification 10 focus upon a gas bubble in the melt
or on the edge of the cover slip. Move into the field a spot where the melt
seems accessible at the edge of the cover slip. Make the field medium bright
and use somewhat oblique illumination. Touch the point of a needle to
a crystal of thymol, and then-while observing through the microscope-touch
the melt with the needle. Observe that crystal plates begin to form and
to grow slowly. Move the slide and observe the front of the crystal mass
as it advances through the melt until it reaches the opposite edge of the
cover glass. The plates are colorless like the melt and are not too distinctly
visible if the illumination is not cleverly selected. Test this by using strong
axial illumination.
Heat the preparation over the microflame to melt the crystals, and
then return it to the stage for a repetition of the experiment. This time,
use axial illumination with strong light and crossed polars. Observe that
polarized light provides a sensitive test for the formation of an anisotropic
phase. When crystallization is complete, rotate the stage and observe
the positions of extinction for groups of crystal plates. Air bubbles remain
dark when the stage is rotated. Large plates or areas of like color may
be used for the observation of axial figures (preceding Expt.); plates that
64 Use of Optical Aids Expt.13

remain dark during the whole revolution of the stage should be tried first;
in addition, one might try to make a new preparation with a much thicker
layer of thymol.
Repeat the experiment once more, especially if many of the plates
show the gray of first order. This time use crossed polars and the first
order red plate in the 45-degree position. Note that this renders the field
bright so that the seeding is easily observed without hindering the ready
recognition of the separating crystals. The colors will be more brilliant
than without the gypsum plate if the preparation is very thin; they may
be less impressive if the preparation is thick. Rotate the stage, and observe
that the plates change their color when they enter a new quadrant (provided
that the gypsum plate is stationary). Try the polars in the position parallel.
B. Use the Na 2 S 20 3 • 5 H 20 preparation made for Expt. 7. First focus
the microscope upon the crystals. Place the preparation upon the steam
bath to liquify it. Allow it to cool, and then place it on the microscope
stage so that the edge of the cover slip appears in the field. Do not change
the focus of the instrument. Cross the polars. While looking into the eyepiece,
seed at the edge of the cover glass with a needle that has been touched to
a crystal of the salt. Move the slide to keep the advancing crystal front in
the field until the crystallization is complete. The phenomenon proceeds
rapidly through the melt. Finally, rotate the stage, change the polars
to parallel, etc., melt the salt again and repeat the crystallization with
crossed polars and the gypsum plate in place, etc.
C. Focus the NH 4 N0 3 preparation with polars parallel. To prevent
damage, adjust the condenser so that its front lens is at least 5 mm below
the slide. Without changing the focus, remove the preparation and heat
it above a microflame until it is completely liquified. Without delay, place
the hot preparation upon the stage so that the axis of the microscope passes
through its center. Watch through the microscope while the preparation
cools. Gas bubbles, recognizable by their round outlines, may be used to
correct the focus. When the temperature has dropped to 170 0 C, cubic
NH 4N0 3 separates, and the melt solidifies. Rotate the polar to position
crossed and note that the field is dark even when the stage is rotated.
Keep watching through the eyepiece. Below 125 0 C, anisotropic
rhombohedral crystals form, and the field becomes bright. Continue
watching the preparation. Further transitions occur at 84 C (rhombic I),
0

at 32 0 C (rhombic II), and - 17 0 C (tetragonal); the first two may be

observed and are recognized by a motion in the preparation.
Repeat the experiment with crossed polars and the gypsum plate
inserted in position of maximum brightness. Try it also with a 5 x objective.
D. Focus the crystals of the benzoic ester without the analyzer. Use
medium brightness and oblique illumination to obtain good contrast.
Make certain that the front lens of the condenser remains 5 mm below the
Expt.14 Use of Polarized Light 65

slide. Then heat the preparation over a microflame until the cloudy melt,
which formes first, becomes clear. Without delay, place the hot preparation
upon the stage and watch it through the eyepiece. Note that striations
appear in the clear melt at a certain temperature and persist until crystalliza-
tion takes place.
Repeat the experiment with strong axial illumination and crossed polars
(without and with addition of gypsum plate). Note that the first anisotropic
phase is liquid; this is indicated by the motion in the preparation and the
round outlines of gas bubbles, and it may be demonstrated by the flow
produced by tapping the cover glass with a rod.
Save the preparations used.

Experiment 14
Preserving Microscopical Preparations
Round cover slips; rings, 2 to 3 mm wide, cut of strong writing paper to
fit the cover slips; small brushes. - Canada balsam, benzene, xylene, chloroform,
gelatin, glycerol, adhesive (Ad-A-Grip glue), Duco cement, Glyptal 1201, red.
Whether the preparations are intended for evidence or for demonstrations,
one will have the desire to preserve confirmatory tests without change just
as they were obtained. This is reasonably simple if only solids are involved.
Frequently, however, the preparation consists of some solid that has
separated from a solution which will evaporate and leave a copious residue
that may completely hide the objects of interest_ There are two ways
to avoid this. Either evaporation is prevented, or the liquid is removed
before evaporation is able to mar the preparation.
Evaporation is safely prevented by sealing in glass, and this is easily
done if the test or preparation has been obtained in a narrow tubing or
capillary. The tubing is sealed at both ends and kept together with a
description inside a stoppered test tube in a dark place (locker) with little
variation of temperature. In this way, the specimen may be centrifuged,
warmed, or cooled each time before mounting it for microscopic inspection.
A liquid preparation on a slide may be preserved for a limited time
by covering it and sealing the edge of the cover. A small drop may be
saved, without disturbing it at all, by inverting over it a small watch glass
so that it does not touch the drop and sealing the edge by applying melted
wax, Duco cement, or Glyptal with a brush.
Putting a cover slip upon a liquid preparation has very often the effect
of completely destroying its characteristic features. Solids that do not
adhere to the slide, will shift their positions and frequently float to the
edges of the cover slip so that they get lost or can no longer be observed
because of the refraction phenomenon at the edge. Before applying this
procedure to an important preparation, it will be wise to practice it with
Benedetti-Pichler, Identification 5
66 Use of Optical Aids Expt.14

duplicates obtained for this end. The size of the solid particles must be
considered since it may determine the thickness of the liquid layer. The
area of the square or round cover slip should be adjusted to the volume
of the liquid (a layer of 0.2-mm thickness below a cover of I-cm edge
obviously has a volume of 20,tt1 = 0.02 ml). In general, it may be preferable
to reduce the volume of the liquid by means of touching it with filter
paper (Expt.40) or a capillary pipet (Expt.43) or to use a cover slip of
such size that the liquid is unable to fill the whole space. When the cover
slip is in place, its edge is sealed by applying melted wax, Duco cement,
or Glyptal with a brush. Instead of sealing in this manner, one may place
a drop of Canada balsam upon the cover slip and then superimpose a second,
larger cover glass; this may be quite practical if the first cover slip is small,
possibly only a fragment of a cover slip. Enough Canada balsam must
be taken to fill the whole space between the large cover and the slide.
As an alternative, one may consider first painting around the drop
on the slide a barrier of repellent wax or cement and then dropping upon
this ring the cover slip, the surface of which may have been treated to
make it repellent (Dri-Film, Desicote).
The removal of the liquid will give more permanent mounts. One will
use the techniques described in Expts. 40 and 43. As a rule, washing will
be necessary to prevent the marring of the preparation by residues from
small amounts of liquid left behind. Obviously, these operations will
change the appearance of the preparation if the solid particles to not cling
to the slide, and again it will be neces~ary to practice on expendable
duplicates before proceeding with a specimen that shall be preserved as
evidence. The resulting dry preparation is then treated like a solid.
The mounting of solids requires careful consideration of the nature of
the solid. Preliminary experiments may be needed to make certain that
the adding of a mounting medium will not destroy the appearance of the
preparation and that the mounting medium has no effect upon the specimen.
The color and the refractive index of the mounting medium should be
considered. The shape of transparent objects is best visible if the refractive
index of the surrounding medium differs by about 0.1; the color of the
object and structures in its interior are best seen when it is immersed in
a medium of the same refractive index. The mounting medium must not
be anisotropic if observation in polarized light is intended.
Mounting in Oanada Balsam or equivalent plastic preparations requires
that the specimen is dry. Drying may be accomplished by heating or by
washing with alcohol.
To exclude air bubbles, first wet the dry preparation with a small
volume of benzene or other suitable solvent. Without delay, add a drop
of the balsam not larger than just sufficient for filling the space below
the cover slip. Place the latter upon the balsam drop, and allow it to
 Basic Rules 67

settle by gravity into the intended position. If it seems desirable, finally

press down upon the cover slip with the eraser end of a pencil to extrude
excess mounting medium which is removed with filter paper that may
be moistened with solvent.
If permissible, heat the preparation several hours at 70 to 80° C.
Otherwise allow it to stand for several days at room temperature; cover
it with a watch glass or bell jar so that solvent vapors may dissipate.
When the preparation has hardened, remove any excess mounting medium
showing at the edge of the cover slip by scraping and finally wiping
with filter paper moistened with solvent. Finally seal by painting the edge
of the cover glass with Duco varnish or Glyptal.
Mounting in glycerol jelly does not require dry preparations and may
also be found suitable for organic crystals which are attacked or dissolved
by Canada balsam.
Proceed in principle as with Canada balsam. A solution with 10% gelatine,
30 % glycerol, and 60% water must be applied quite hot. If this is not permiss-
ible, add more water; a solution with 1 % to 2 % gelatine remains liquid down
to room temperature and may require some evaporation for setting to a gel.
Dry mounting has been recommended by EMICH for its wide applicability
and for the fact that it does not affect the arrangement of objects in the
preparation. To-day, it appears especially attractive because of its simplicity
and the availability of excellent waterproof adhesives. The disadvantage
that many transparent objects give poor images by transmitted light when
surrounded by air may often be overcome by changing to observation by
reflected light.
Rings of writing paper, the outer diameter equal to that of the round
cover slips, may be obtained with the use of dies. Make certain that the
surface of the slide around the 'specimen is clean. Holding the paper ring
with forceps, paint both sides of it with adhesive (Ad-A-Grip or similar).
Place the ring upon the slide so that it surrounds the specimens, and place
the clean cover slip on top of it. Place a 200-g weight upon the cover slip
and allow to stand until the adhesive has set. After 24 hours, paint the
edge of the cover slip with Duco varnish or Glyptal.
It is obvious that all permanent mounts should be labelled with special
care. This may be done with special ink for writing on glass, or by attaching
a label to the top of the slide.

Technique of Experimentation and Observation

Basic Rules
To have a good probability of success beware of haste and make it a
habit to let reasoning precede action. Before starting any experiment,
make certain: (1) that you know its purpose, (2) that you understand the
5*
68 Technique of Experimentation

principles involved in its performance, (3) that you know how to perform
it without danger to yourself and your neighbors and without risk of losing
the material under investigation, (4) that you know what phenomena may
be expected and how they are best and most sensitively observed.
To begin with, realize that a difficult task is not made easier by increasing
its complexity. Contaminating the material under investigation by the
use of apparatus which is not perfectly clean or reagents that contain
impurities renders the task of identification more complex and may make
futile the interpretation of the phenomena which may be caused by the
introduced foreign matter and not by anything contained in the original
object. Observation is made difficult and sometimes impossible by dirt
on the outside of transparent containers.
These considerations lead to the conclusion that cleanliness must be
practiced to the utmost. Not only all apparatus containing the material
under investigation, but also everything that gets into contact with it,
including the hands, must be clean. Bench tops and shelves should be
clean at all times; this will prevent that utensils and hands become soiled.
For the same reason do never permit dirty apparatus to accumulate, but
clean it immediately after use. The success of cleaning is also better assured
when it is still known what the apparatus contained and the proper solvent
for the removal of adhering matter is obvious.
Use water and reagents of good quality, and maintain their quality
most conscientiously as follows: (1) Before removing a stopper make certain
that the outside of the container is clean. If necessary, remove dust or
crusts with a cloth (if the contents are hygroscopic or react with water)
or by rinsing under the tap and then carefully drying with a cloth. (2) Clean
spoons, spatulas, pipets before and after use; do not use utensils that are
attacked by the reagent. (3) Handle the stopper so that it cannot become
contaminated; if this should happen by accident, wash and dry the stopper.
(4) Do not pour from a container directly into apparatus from which gases
(NHa, HCI, N0 2 , C1 2 , Br2 ) or vapors (H 2 0, I 2 , solvents) may be taken up.
These would be drawn into the reagent container. (5) Close the reagent
container after removal of the needed amount. (6) Take care that a reagent
which does not appear reliable for any reason is tested and, if necessary,
discarded.

Work on the Gram Scale

A triple-beam balance responding to 0.01 g is suited for the weighing
of samples; a completely enclosed torsion balance with a capacity of 500 g
to 1 Kg is recommended for the weighing of reagents. Suitable measuring
spoons may be used for adding approximately known amounts of powdered
or finely granular reagents. Graduated cylinders of 10-ml and 50-ml
capacity suffice for the measuring of solutions.
 Work on the Gram Scale 69

Most of the chemical treatment is performed in low-form beakers of

lQQ-ml to 4QQ-ml capacity and in standard test tubes,
Solids are handled with a spoon made of horn, plastic, porcelain, or glass.
Reagents are often added directly from the reagent bottle, but this should
not be done if vapors or gases are contained or may form in the receiving
vessel since these would be drawn with the air into the bottle and taken
up by the reagent. In such instances, pour a small amount of the reagent
into a clean test tube, beaker, dish, etc. and add it from there to the reaction
mixture; discard whatever is left and not needed of this contaminated
portion.
Whenever this is required, add the reagent dropwise by cautious pouring.
The task is facilitated by touching the rim of the reagent container to
a stirring rod that guides the liquid into the receiving vessel; the use of
medicine droppers is not recommended since it increases the danger of
introducing contamination.
It is understood that mixing is required after addition of a reagent or
each portion of a reagent. Do this by stirring or swirling; to mix the contents
of a test tube, grasp it with three fingers just below the opening and impart
jerky motions through the wrist so that the closed end moves sideways
through short arcs.
To adjust the pH of a solution, add small portions, finally drops, of
the required reagent. Mix after each addition and, using a stirring rod,
transfer a small amount of the mixture to the indicator paper. If the
desired pH is overstepped, go back by adding still smaller portions of the
suitable neutralizing agent.
For saturating with hydrogen sulfide (or any other gas), transfer the
solution to an Erlenmeyer flask of suitable capacity, 250 m!. Close the
flask with a one-hole rubber stopper provided with a glass tubing, fire-
polished at both ends, tha,t reaches to the bottom of the flask. Use a clean
length of flexible rubber tubing for connecting to the gas line. The latter
should be supplied with a pressure head of about 50 cm .water column
(4 cm mercury); if the line contains a simple gas washing bottle, this
will also serve for a convenient bubble counter. To start with, lift the
stopper in the neck of the Erlenmeyer and use a rapid stream of gas to
displace the air above the liquid. Then, without shutting off the gas
supply, tighten the stopper and shake the flask to get intimate mixing
of the liquid and the gas in it. Continue until no more bubbles rise in the
washing bottle (Erlenmeyer flask), which indicates that saturation must
have been reached since no more gas is consumed. Turn off the gas supply
and lift the stopper before disconnecting. Remember, there is pressure
in the flask and its contents will be forced out through the line if the latter is
broken or disconnected. To safeguard against such loss due to some accident
(washing bottle pops open or is broken, rubber tubing slipping off glass
70 Technique of Experimentation

connector, failure of generator) raise the inlet tubing in the Erlenmeyer

so that it ends in the gas space above the liquid; of course, this makes it
desirable to have a bubble counter in the line.
For stirring rods, cut clean Pyrex rod of 3- to 4-mm diameter into lengths
of 8, 12, and 16 cm and carefully firepolish both ends.
For heating, place beaker and flasks upon a wire gauze. Heat in test
tubes by letting the gas flame play directly upon the glass . A steam bath
for test tubes, watch glass, or evaporating
• dishes is easily improvised by using a
Pyrex beaker with boiling water; a test
tube is simply placed into the water; a
watch glass or dish is placed upon the
opening of the beaker.
Evaporation of solutions perform pro-
perly by heating on the steam bath in
open, flat, thin-walled dishes of suitable
material (Pyrex, porcelain, vitreous silica,
platinum). The process may be consider-
ably hastened by blowing a stream of
filtered air from a capillary upon the
surface of the liquid. This also aids in
keeping dust away. Use a dish of 3- to
4-inch (7 to lO cm) diameter, and add
the solution in several portions, but never
fill the dish close to the rim. (Rapidly
J • boiling in an open container causes loss by
Fig. 8. Filtration; about 1/3 nat.
spray and spattering; if it is done in
size covered containers, flasks, or test tubes,
condensate dripping or flowing back to
the' hot part of the apparatus may cause breakage and loss.)
Fusions perform in"crucibles of 5- to lO-ml capacity of suitable material
(porcelain, vitreous silica, iron, nickel, platinum). Rest the crucible on
a clean clay or silica triangle supported by a ring stand or tripod. Ashing
and some fusions may be advantageously performed in a small electric
muffle furnace.
For bead tests use a straight platinum wire, 6 cm long and 0.3 to 0.5 mm
in diameter, one end of which is sealed into a length of glass tubing that
serves for handle. Use the technique described in Expt. 59 and P.39.
Mount the handle in a cork which fits a standard test tube, and keep the
wire suspended in the stoppered test tube when it it not in use.
When transferring a liquid from one container to another, render the
outside of the pouring edge repellent. With the finger, get some oil from
the scalp or the side of the nose and rub it on the glass surface. Either
 Work on the Gram Scale 71

rest the pouring edge on the rim of the receiving vessel, which may be held
inclined, and allow the liquid to flow down along the inside wall of it,
or guide the stream of the liquid with the stirring rod, Fig. 8.
Use filtration for the separation of solid from liquid. To this end, keep
ready at least two porosities of filter paper, a very slow paper for solids
consisting of minute particles (precipitate of BaS04 , etc.) and a very fast
paper for gelatinous precipitates (hydrous oxides, sulfides). Use fluted
filters when only the filtrate is wanted.
Fold the disk of paper exactly in half and once again to the outline
of the quadrant. If the amount of solid is very small, now trim the size
of the filter with shears so that the solid will fill about one-fourth of the
capacity (cone of paper). Open the cone and note that one side is formed
by a triple layer of paper. Tear off the corner of the outside double layer,
and then tear off some more of the outermost sheet. Fit the dry cone into
the glass funnel, best a 58-degree funnel with long stem. The rim of the
cone must fit snugly to the glass, but the apex of the cone need not touch
the funnel. Moisten the filter and, with clean fingers, press its edge to
the glass to obtain a tight fit all around. Fill the filter with water and keep
on repairing the fit of the edge until the stem remains completely filled
with water even after the filter has been drained empty. Leave this water
in the stem; it will provide a slight amount of suction and speed the
filtration.
Place the funnel into the filter stand and put the receiving vessel
underneath so that the stem touches its side and the filtrate will flow down
without splashing. With the stirring rod, guide the liquid into the filter,
Fig. 8. Fill about three-fourth of the filter and allow it to drain completely
before replenishing. If there is much liquid and little solid, first decant
most of the clear liquid through the filter and save this portion of the
filtrate before starting the transfer of the solid. For the transfer of the solid,
obtain a slurry by mixing with the stirring rod or swirling, and guide
the slurry into the filter by means of the stirring rod. Save the filtrate
before starting with the washing of the solid. The washings usually may
be rejected since they would only dilute the filtrate.
Washing may be combined with the finishing of the transfer of the
solid. The wash liquid supplied from a wash bottle (preferably a plastic
squeeze bottle) may be used to rinse solid into the filter. For washing,
fill the filter nearly full with the wash liquid and allow it to drain completely
before adding the next portion of wash liquid. The latter may be added
from a beaker by guiding the stream with a stirring rod. To avoid spattering,
use caution when adding wash liquid from a wash bottle, especially when
using the force of the jet for stirring up the solid in the paper cone or
washing the solid down into the tip of the cone. In many instances, it will
suffice to wash a grainy or crystalline precipitate twice or a gelatinous
72 Technique of Experimentation

precipitate four or five times. If the substances contained in the filtrate

interfere with subsequent operations, however, it may become necessary
to continue washing to the practically complete elimination of the objection-
able substances.
Separation by Decantation is preferable and saves time if the solid
phase is quite insoluble and settles reasonably fast to occupy a space less
than one tenth of the capacity of the container. In such instances, prepare
a filter as usual. Allow the precipitate to settle in the container which is
already inclined nearly to pouring position (by suitably supporting it in
a beaker, cardboard box, mortar, etc.) and then pour off the nearly clear
solution into the filter. Thoroughly mix the remaining thick slurry with
ten times its volume of wash liquid, allow to settle, and again decant
through the filter. If the washing is repeated twice in this manner, only
0.0001 of the solutes contained in the "original" liquid phase will remain
with the "final" slurry. One may decide to remove most of them by
transferring the slurry to the filter, or one may prefer to treat the slurry
as it is; the reagent may be added by percolating it first through the filter
that has collected part of the solid.
Removing the Solid from the Filter. Simply use a spatula or spoon on
the moist cake if much solid has been collected and not all of it is needed.
Moderate and small amounts of collected solid, which are needed in their
entirety, treat depending upon their solubility behavior.
If the solid is readily soluble and is to be dissolved for further treatment,
dissolve it on the filter cone by adding the solvent drop by drop. As an
alternative, close the stem of the funnel by applying a rubber or plastic
cap or stopper to its opening and then fill the filter cone with the solvent.
In either instance, allow the filter to drain completely when dissolution
has become complete. If the solution should be turbid or opalescent, you
may try to clear it by passing it repeatedly through the filter, possibly
after stirring into it a small amount of paper pulp ("filter aid"). If the
volume of solution is small, it may become desirable to improve the yield
by finally folding the filter and pressing it into the funnel (finger cots may
be used). If diluting is permissible, the yield may be improved by washing
the filter previous to pressing it out; diluting may be avoided by using
to this end a reagent solution which would have to be added anyway.
If the solid is only partly soluble and requires thorough digesting,
its treatment in the filter cone is not recommended. Either punch a hole
into the apex of the filter cone with the stirring rod and use the jet of the
wash bottle to wash the solid through the stem of the funnel into a suitable
receiving vessel, or remove the filter from the funnel, unfold it, tear off
the parts holding none of the solid, and place the rest of the filter with
the solid adhering to it into the beaker or dish selected for treatment..
Undesirable diluting may be avoided in the first procedure by using,
 Work on the Gram Scale 73

instead of water, a measured volume of reagent solution from a small

washbottle that may be made of a test tube. The second procedure must
be avoided if the digestion with the reagent puts products of the breakdown
of cellulose into the solution, which could interfere in the subsequent steps.
Glass fiber paper could be used in such instances. Obviously, the material
of the broken-up filter will be collected together with the insoluble part
of the solid; this may be inconvenient at some times and desirable at
others when the disintegrated tissue acts as filter aid.
Small amounts of insoluble solid may be isolated by ashing the filter
following the procedure of quantitative analysis (13). This may cause
changes and losses because of reduction, oxidation, and volatilization.
Extraction 01 Solids. Use the technique for the washing of solids by
filtration, decantation, or a combination of the two.
Extraction 01 Liquids. Perform the extraction in a Squibb type separatory
funnel of pear shape, in a test tube with the use of a siphon as described
below for the centigram scale, or in a syringe (45) of 20- to 100-ml capacity.
When using solvents with high vapor tension, release the pressure in the
separatory funnel from time to time by opening the stopcock for a brief
time while holding the funnel in an inverted position (thumb upon the
stopper) so that the liquid does not reach the outlet to the stopcock (13).
Beware of the formation of more or less stable emulsions. Of course,
vigorous shaking with intimate intermingling of the two liquids quickly
establishes equilibrium, but do not use this expedient unless you have
assured yourself by a side test with a small amount that the emulsion
will break up readily even after vigorous shaking or stirring. If such
assurance has not been obtained, reach equilibrium by gently rocking
the container or by slowly stirring to keep the liquids in motion without
mixing them.
If a stable emulsion has once formed, use small portions of it for finding
a means to break it and then treat the bulk accordingly. Various ways
are suggested in the literature (13, 58). Exposure to vibration of very high
pitch, but below the ultrasonic range, may be tried. For small apparatus,
it will suffice to fasten down the interrupter of an electromagnetic door
buzzer, operating via transformer on 110 v a. c., so that it cannot make
and break and vibrates with an almost imperceptible motion (413). The
apparatus is touched to the oscillating arm.
Adding a small amount of alcohol or a large excess of organic solvent
may break up an emulsion, and one may try to change the pH or to
precipitate responsible emulsifying agents. The emulsion may be removed
and washed by filtration. As a last resort, evaporate the whole mixture
with the proper precautions until the emulsion disappears. Small amounts
of solid often act as stabilizer; thus be certain to obtain a perfectly clear
solution before making another attempt at extraction.
74 Technique of Experimentation

For methods of continuous extraction, which may avoid the formation

of emulsions, see the literature (13,45, 147).
Distillation. A distilling flask of 25- to 250-ml capacity with the side
arm placed at the center of the neck is generally useful. If the side arm
is made 30 cm long, it may serve as condenser tube, and it will accomodate
a short cooling jacket. If a ground glass stopper is not provided, corrosive
liquids may be distilled by placing a very short cold-finger condenser
(short test tube with cooling liquid) into the opening of the neck and
regulating the rate of distillation so that the ring of condensate does not
climb much above the opening of the side arm.

Fig. 9. Refluxing and Distillation; about 1/3 nat. size

If loss of volatile components must be carefully avoided, the distilling

flask shown in Fig. 9b is recommended, which is designed to avoid dead
spaces and contact of the vapors with joints of any kind. The condenser
tube c may receive a water jacket; the inlet tube i may be closed during
distillation or connected to a supply of suitable inert gas. The residue
in the flask may be siphoned off through i (861).
Sublimation. Place the substance into a boat of porcelain, vitreous
silica, or platinum and introduce the boat into a glass tube of 8- to 12-mm
bore and 25- to 40-cm length, which has been clamped in a horizontal
position. At the location of the boat, place a roll of wire gauze around the
tube to obtain uniform heating. Connect one side of the tube to a supply
of suitable gas and the other side to absorption apparatus that may seem
indicated. Heat the material in a very slow stream of gas. The sublimate
will collect downstream from the boat which will finally hold the non-
volatile matter.
The sublimate may often be very conveniently recovered by mounting
the tube, Fig. 9a, in the neck of a flask by means of a rubber stopper,
Expt.15 Work on the Gram Scale 75

cork, or ring of asbestos paper and using it as a reflux condenser with a

suitable solvent boiling in the flask.
Oonfirmatory Tests. Crystal precipitations upon the microscope slide
and spot tests are recommended for use, Expts. 24 to 42. The traditional
confirmatory tests of the gram scale are performed in test tubes. From
a few drops to about 2 ml of the solution to be tested are placed into a
test tube and treated with the reagent. Efforvescence, the separation
of a precipitate, and the appearance of a strong coloration in the liquid
are readily perceived. A small amount of precipitate, turbidity, opalescence,
and fluorescence are best seen before a black background and with a strong
beam of light more or less vertical to the line of viewing. Opalescence and
turbidity indicate the formation of a new phase, provided that the reagents
have been clear. Weak colorations may become visible when looking
axially through the test tube which is held over a well illuminated, white
paper.
Efforvescence indicates the liberation of a gas. Especially gases that
are not lighter than air will collect above the liquid contents of the test tube
and may be identified by means of reagents (test paper, reagent drop in
platinum loop or hanging on stirring rod) inserted into the gas space.
These tests are far more sensitive when using special apparatus recommended
for the centigram-and milligram scales, see below.
Storing Material under Investigation. Solids that have been collected
upon filter paper are best immediately transferred to the apparatus in
which they will undergo the next treatment. Beakers or dishes containing
material under investigation are covered with watch glasses; the contents
of flasks or test tubes are best protected by inverting over the opening
a small beaker or wide vial; stoppers are not recommended.

Experiment 15
Phenomena Observed Upon Heating in an Inert Atmosphere
Tubing of 3- to 6-mm bore of glass with high softening point, 21-cm lengths. -
Brass, filings or turnings; barium chloride dihydrate or clear cleavage plates
of gypsum; sucrose; benzoic acid of tested purity; pyrite; ammonium sulfate;
cinnabar, HgS; anhydrous sodium carbonate; pH test paper; Nessler reagent
(dissolve 11.5 g HgI 2 and 8 g KI in water to give 50 ml, add a like volume 6-F
NaOH, mix, and decant from any precipitate that may form on standing);
mercurous test solution (1.4 g Hg 2 (N0 3 )2' 2 H 2 0 in 100 m11-F HN0 3 ); 1-F BaCI2 ;
0.02-F KMn0 4 •
The material is heated at the bottom of a narrow test tube so that
air has little access to it and oxidation will not take place to an extent
that would obscure other phenomena.
Starting with lengths of 21 em, prepare test tubes of 3- to 6-mm bore
and lO-cm length. Using an oxygen blast flame, obtain two test tubes
76 Technique of Experimentation Expt.15

from each length by first drawing a capillary at the middle of the tubing,
and then sealing at the tapers. The sealed ends may be rounded off by
reheating and blowing to obtain hemispherical shape; if this is done, place
the finished tubing into a drying oven of 130 0 C for the removal of moisture
that may have collected in the tubing.
To perform an experiment, bring about 2 to 10 mg (10 Jll, i. e. a cube
of about 2-mm edge) of the solid to the bottom of the tube. Grasp the
tube near the opening with the fingers or with a narrow strip of paper
that has been folded two to three times (form the strip into a V and hold
the tube in the apex of the V). Holding the tube at an angle of 45 degrees
to the horizontal, slowly heat the bottom with the non-luminous flame
of the Tirrill burner. Whenever necessary, interrupt heating fol' the
observation of the phenomena: change in appearance, melting, sublimation,
distillation, liberation of vapors or gases. If this promises additional
information, finally raise the temperature to the softening point of the
glass, about 800 to 900 C (faint red incandescence visible only in a dark
0

place: 550 C; dark red glow: 700 C; bright red glow: 950 C). Try the
0 0 0

experiment with the substances listed below.

a) Brass filings or turning. Heat to a high temperature. If necessary,
use a magnifying glass to inspect the color changes on the residue and the
distillate of zinc and zinc oxide.
b) Clear crystals of BaC1 2 • 2 H 20 or clear cleavage piece of gypsum.
Note the change in the appearance of the residue. Test the condensate
with pH test paper.
c) Sugar or filter paper or sliver of wood. Note the odor and test the
pH of a distillate.
d) Small amount of pure benzoic acid. Finally try to obtain pyrolysis
by heating the tube above the sublimate.
e) Pyrite, FeS 2 • Observe the changes on the residue and the appearance
and volatility of the distillate. Note the odor.
f) Ammonium sulfate. Observe odor and pH of the gases (liquids)
obtained at different stages of the decomposition. Test the fumes with (1) a
solution being 0.5 F in both BaC1 2 and HCl, (2) a solution 0.01 F in KMn0 4
and 1 F in H 2 S0 4 ; take up a drop of the reagent with a platinum or glass
loop, P. 36-or with the end of a stirring rod-and hold it in this manner
in the path of the fumes.
Repeat the experiment, but this time mix the ammonium sulfate with
an equal volume of anhydrous sodium carbonate. Test the liberated gas
with pH paper, a droplet of 12 F HCl at the end of a stirring rod, and a
narrow strip of filter paper which has been treated near the end with a
tiny droplet of Nessler reagent (or strong mercurous nitrate solution and
then bathed in a larger volume of NaCl solution).
Expt.16 Work on the Gram Scale 77

g) Cinnabar, HgS. Scratch the sublimate with a glass needle and

observe a color change which may take place on standing at room
temperature. Repeat the experiment, but mix the sulfide with an equal
volume of· anhydrous Na 2CO a•
It is obvious that the sublimates, distillates, and residues obtained in
these experiments could be tested and identified by the microtechniques
described later.

Experiment 16
Phenomena Observed Upon Heating in a Stream of Air
Supply of glass tubing, 3- to 6-mm bore, of high softening point. - Cinnabar
(HgS); galena (PbS); sulfur; selenium; tellurium; graphite or carbon; pH test
paper; 0.1% fuchsine solution; 0.02-F KMn0 4; lime water.
Cut the tubing into lengths of 12 cm. Using a spatula (aluminum,
nickel, or copper wire hammered flat at one end), deposit 2 to 10 mg
(10 fll, i. e., cube of about 2-mm edge) of the substance to be tested inside
the tube and approximately 3 cm from the opening. Grasp the tube near
the empty end as directed in Expt. 15 and hold it inclined at an angle
of 30 degrees to the horizontal so that the sample is near the lower end.
Use a small (2-cm height), just non-luminous Bunsen flame to heat the
part of the tube containing the sample. The hot air rises in the tube to
escape at the upper end, while fresh air is drawn in below the heated sample.
Gradually raise the temperature while observing the sample. From
time to time interrupt the heating and test the escaping gases at the upper
opening of the tube. Try the experiment with the substances listed below.
a) Cinnabar. Heat very slowly. Test the escaping gas with pH test
paper, litmus paper, a narrow strip of filter paper which has been treated
near the end with a tiny droplet of fuchsine solution. Inspect the condensate
with a magnifying glass, and wipe it together with a glass thread or a steel
needle.
b) Galena, PbS. Note the changes on the solid. Test the escaping gas
with the fuchsine spot on filter paper, with litmus paper, with a droplet
of lime water, and with a droplet of solution 0.01 F in KMn04 and 1 F in
H 2 S04, Observe the odor.
c) Sulfur. Test the escaping gas as under (b). Observe any condensate
with the magnifying glass.
d) Selenium. Observe the odor and the phenomena in the tube.
e) Tellurium. As under (d).
f) Graphite. Observe the sample during heating. Test the escaping
gas with pH paper, odor, fuchsin spot, and a droplet of lime water.
78 Technique of Experimentation Expt.17

Experiment 17
H eating Upon the Charcoal Block
Small rectangular block of dense charcoal as obtainable from laboratory
supply houses or block of graphite; 20-cm length of 4- to 6-mm bore glass tubing
of high softening point; small mortar; strong permanent magnet; supply of
hydrogen gas at low pressure. - Brass, filings or turnings; CdCOs ; KNO s ;
SrS04 ; Fe20 S ; As 20 S ; anhydrous Na2COS '
Reductions take place when material is heated in contact with carbon.
Many metals separate as such either as a powder or collected into a bead
depending upon whether or not the melting point is reached. Volatile
metals may evaporate completely or partially and combine with the
oxygen of the air surrounding the reaction mixture to deposit as oxides
upon the charcoal around the heated material. In presence of sodium
carbonate, compounds of halogens, sulfur, and phosphorus are converted
to halides, sulfide, and phosphide.
BELCHER, HARRISON, and STEPHEN (942) recommend the following
procedure without specifying the nature of the gas. It is obvious that
the needed high temperatures may be reached only if the temperature
of the flame is raised as its size diminishes and the amount of reaction
mixture is increased. A small hydrogen flame gives the required temperature
and provides a reducing atmosphere. As an alternative, a micro oxygen
torch (142, 170, 888) might be used and could be adjusted to give an
oxidizing flame.
Start with a 20-cm length of glass tubing of 6-mm bore and high
softening point, and use an oxygen blast lamp. Heat the tubing 5 cm from
one end; slowly rotate in the flame without exerting a pull so that glass
collects and the tubing contracts to a diameter of about 3 mm. Then
remove from the flame, allow to cool somewhat while continuously rotating
the tube, and then slowly draw a short, thick-walled capillary of about
0.5-mm bore. Cut the capillary near the short end of the tubing, and
remove the latter. Then heat the capillary in a Tirrill flame about 2 cm
from the wide tubing, and draw it out to a fine point. Cut the taper to
get an orifice of 0.1- to 0.2-mm diameter. Finally make a right-angle bend
in the middle of the remaining short piece of capillary, and connect the
wide end of the tube by means of a long piece of flexible rubber tubing
to the supply of hydrogen.
With some suitable tool (cork borer, drill, end of file) cut into the surface
of the charcoal block a very shallow circular depression (1 mm deep,
10 mm in diameter), the rim of which is just high enough to keep beads
in place without preventing full access of the flame to the material. Place
the block upon the bench and prop up one end of it so that the surface of
the block is inclined at about 30 degrees to the horizontal. Place 5 mg t()
Expt.18 Work on the Gram Scale 79

50 mg of the material to be tested into the depression so that it is piled

together at the lower half of the rim.
Regulate the pressure to get a soft hydrogen flame of 2- to 5-mm length
and, by hand, direct this flame so that it plays upon the sample from above.
According to need, raise the temperature by increasing the pressure which
will give the flame a length of 10 mm to 30 mm. Do not direct a stiff
hydrogen flame on powdery material before it has sintered, for it would
blow the powder out of the depression. Judge the temperature by the
color of the light emitted by the incandescent material. Try the experiment
with the substances listed below.
a) Brass. Note the color of the flame at different times. Observe
the changes in color and shape, which the metal undergoes. Raise the
temperature to about 900 0 and observe the deposit of 7,nO around the
0

depression, which is yellow when hot and white when cold. The deposit
does not volatilize but may show a green luminescence when touched
with the oxidizing seam of the flame. Moisten the copper bead with I2-F HOI
and observe the color of the flame when it is again heated.
b) Cadmium Carbonate. Heat to about 800 0 O. The brown deposit
may be surrounded by a thin film of the oxide showing blue iridescence
similar to the "eyes" in the plumage of peacocks.
c) Potassium Nitrate. Note the deflagration upon the coal and the
flame coloration.
d) Strontium Sulfate. Observe the strong incandescence of the heated
sample. After cooling, transfer the residue to a bright silver coin and
moisten it there with a small droplet of water. After a few minutes, rinse
and observe the brown spot of AgS on the surface of the coin, Hepar test.
Mix SrS04 with an equal volume of Na 200 s and repeat the test with the
mixture.
e) Red Ferric Oxide. Mix with a like volume of Na 200 S and heat on
the charcoal to the highest attainable temperature. Allow to cool, scrape
the residue into a mortar, and grind it to a powder. Transfer the latter
to a sheet of paper. Move a magnet along the underside of the paper
to pull the magnetic particles out of the mixture and to collect them in a pile.
f) Arsenic Trioxide. Heat with a very small flame and note the odor
and the coloration of the flame. White fumes may be seen and a white
deposit may be obtained at some distance from the flame.

Experiment 18
Performance of Flame Tests
Nickel wire, 0.5- to I-mm diameter, 20 em long; copper wire of like dimensions;
I5-cm length of glass tubing, 2 to 6 mm in diameter; small square of blue cobalt
glass, 5 cmx5 cm; test tube filled with powdered K 2Cr 2 0 7 • - NH 4CI; KCI;
LiCl; SrSO4; CaCOs ; MnCl 2 • 4 H 20; TlNOs ; CHCIs ; CHIs.
80 Technique of Experimentation Expt.18

The emission of colored light by an otherwise non-luminous flame is

caused by the excitation of certain atoms or molecules which then radiate
with frequencies characteristic of the species and also determined by the
temperature of the flame. Volatilization is the prerequisite, and differences
in volatility may be used to resolve interferences. All alkali salts are volatile
and may be driven off at relatively low temperature before converting
refractory compounds of other metals to the volatile halides.
Flame tests are very useful for the detection of lithium, strontium,
barium, thallium, and the halogens. The value of the sodium test is greatly
diminished by its extraordinary sensitivity and the wide distribution of
the element. The coloration given by calcium is difficult to identify without
the use of a spectroscope. Violet, pale blue, and green colorations must
be interpreted with careful consideration of the conditions of the test and
the treatment of the sample. Obviously, spectroscopic observation of
the colorations will, under all conditions, enhance the reliability of the
conclusions.
The traditional use of a platinum wire for introducing the sample into
the flame is convenient if the absence of heavy metals which could be
reduced and alloy the wire is assured. If this assurance is lacking, a fiber
of vitreous silica, drawn out from the end of a rod or cemented into the
end of a glass handle, is preferable; the used end may be broken off and
discarded whenever cleaning proves difficult.
Nickel and steel wire are practical substitutes. Use thick-walled glass
tubing of 2-mm bore and I5-cm length. If the bore of the tubing is larger,
heat the ends of the tubing so that they contract to give orifices of little
more than I-mm diameter. Draw a 20-cm length of wire through the
tubing, which serves for handle, so that 5 cm of the wire is exposed at one
end.
Pour some 6-F HOI into a watch glass, and clean a nickel or steel wire
(just like.a platinum wire or a silica fiber) by dipping the end into the acid
and then igniting in the hottest part of the roaring Bunsen flame. Repeat
this until there is no more coloration obtained in the flame before or after
the wire becomes incandescent. Use only very small amounts of sample
for each test.
a) Sodium. Touch the end of the clean wire with the fingers, and then
insert it into the cool outer mantle of the roaring flame about I cm above
the barrel of the burner. Observe how long the yellow coloration persists.
Repeat the experiment, but look at the flame through the cobalt glass.
Repeat the experiment in a dark room and observe that the dichromate
in the test tube looks yellow when illuminated by the sodium flame.
b) Place some KOI upon a watch glass. Dip the end of the clean wire
in distilled water and then pick up with it a small kernel of KOI and bring
it into the seam of the roaring flame just above the barrel of the burner.
 Work on the Centigram Scale 81

Note that the coloration is visible through the cobalt glass. Repeat the
test with Liel and with TINO a. Finally clean the wire by treating it with
Hel and igniting.
c) Take up with the clean end of the wire some SrS0 4 and heat first
in the coolest part of the flame (seam above the barrel) and finally in the
hottest part above the tip of the inner blue cone. Note the strong
incandescence of the sample. Try to reduce sulfate to sulfide by slowly
moving the sample at the end of the wire through the
tip of the inner cone down to the center of the orifice lJuler
of the barrel. Allow to cool in the stream of emerging --.+.-"'-~ flzidl~/111
sum
gas, and then quickly withdraw through the seam
of the flame. Hold the end of the wire into the gas
space of a bottle containing 12-F HCI to convert
sulfide into chloride, and then again heat the end of
the wire as at the beginning. ..
liSO
' .
Repeat the experiment with CaC0 3 • There is no 4,-=::"""'- lumllltlUs
~iluc'nllip
need for reducing the sample, but it should be treated b
with HCI fumes.
d) For the following Beilstein tests use the just
non-luminous flame; the copper wire melts in the
roaring flame of the Bunsen burner. Clean the wire efltJ

after each experiment by just heating it in the flame

until the green coloration does no longer appear.
Spread the various solids on a watch glass, and pick
up small kernels with the hot end of the wire. Observe
the blue to green flame obtained with NH 4 CI,
MnCI 2 • 4 H 20 , and CHIao In all instances, quickly Fig .lO. Non-Luminous
Bunsen Flame. Tem-
introduce the substance into the hot part of the p eratures are given in
flame to avoid the possibility of its evaporation d egrees Celsius; the
r educing tip b ecomes
before it can react with the copper oxide.
luminous when the air
Try chloroform by touching the liquid with the ports are n early closed;
hot wire and then inserting it into the hot part about 1/2nat. size
of the flame. - As an alternative, heat the clean
end of the copper wire in the flame and introduce chloroform vapor
at the air port of the burner; to this end, take into the lower end of a
capillary about 1 mg (1,u1) of chloroform and insert this end into the air
port of the burner.
Try the chlorine test with kernels of KCl.

Work on the Centigram Scale

On the centigram scale it begins to become practical supplying the
individual work bench with a complete set of reagents. Short, wide screwcap
vials are recommended for solids; liquid reagents may be kept in bottles
Benedetti-Pichler, Identification 6
82 Technique of Experimentation

of 15- to 30-ml capacity and, if this seems desirable, equipped with medicine
dropper and rubber bulb. Plastic vials should be used for fl}lorides, and
plastic bottles are recommended for distilled water and alkaline solutions.
The Weighing of Samples may be performed with a pulp balance,
a Sauter balance, a "Quartz Spiral Spring Balance" (Microchemical
Specialties Co., Berkeley, Cal.) , or some simplified Salvioni balance (142).
The Estimation of Volume requires measuring pipets of 0.1- to 0.2-ml
capacity, calibrated to indicate 0.01 ml = 10,ul. Transfer pipets of 0.5-,
1-, and 2-ml capacity may be obtained by crudely calibrating a dropper
obtained by drawing a capillary of 0.5-mm bore and 8-cm length at one
end of a glass tube of 6-mm bore and 12-cm length; the capillary tip of
such a pipet may be calibrated as is done with capillary pipets (Expt. 22)
and then used for measuring volumes from 1 to 20,ul.
Centrifuge Tubes. The most useful type of apparatus for general work
is a centrifuge tube of 16-mm bore in the upper cylindrical part of 2- to
3-cm length and a conical taper of 4- to 5-cm length, which should be
formed to end with a hemispherical cup of 2- to 3-mm radius at the apex.
Reagents are Added by transferring them with microspatula, small
measuring cup (for finely powdered solids), pipet, or dropper to the inside
surface just below the opening of the centrifuge cone or test tube (micro-
beaker). Brief swirling in the centrifuge is sufficient to collect all material
in the tip of the tube.
Adding portions of l-,ul volume is approximately equivalent to the
dropwise addition of reagents on the gram scale. For the adjustment of
pH (neutralizing), it will thus be necessary to add acid and base in portions
of l,ul. Ammonia and hydrogen chloride may be added by blowing upon
the surface of the solution a current of air which has been passed through
a gas washing bottle containing solutions of suitable strength of these
reagents. In many instances, it may be preferable to evaporate the solution
to dryness for the removal of ammonia or volatile acids and to dissolve
the residue in a solution of the desired pH.
Mixing. To obtain a stirring rod, take a 7-cm length of glass rod of
3-mm diameter and draw out one end to a thread of I-mm diameter and
8-cm length. Firepolish both ends. Fuse a bead of 2-mm diameter at the
end of the thread; hold the rod at an angle of 45 degrees to the horizontal
while touching the end of the thread to the seam of the Bunsen flame.
This will place the bead at an angle to the axis of the rod and give better
stirring action when twirling the thick end of the rod between the fingers.
Either rinse the stirring rod back into the tube when withdrawing it or,
and this does not dilute the contents, touch it repeatedly to the inside
surface of the tube to strip it of liquid adhering to it.
Mixing is also obtained by holding a centrifuge tube or microbeaker
below the rim between thumb and index finger of the left hand so that it
 Work on the Centigram Scale 83

hangs vertically and striking the bottom end with the fingers of the
right.
Treating with Gas. The gas may be bubbled through the liquid as
described on p. 107. As a convenient alternative, provide the centrifuge
tube containing the liquid with a one-hole rubber stopper carrying a delivery
tube of about 15-cm length (2- to 5-mm bore), one end of which is drawn
to a quick taper and cut to give an orifice of about 0.5-mm diameter.
Obtain a slight constriction 5 cm from the other end by allowing the glass
to gather and collapse while being rotated in the flame. Cut the tube at
a distance of 1 cm from the constriction, and firepolish both ends. After
cooling, insert above the constriction a loose wad of cotton which will
hold back any dust particles or spray carried along by the gas. Insert
the tubing into the stopper so that the orifice will remain at least 3 to 4 cm
above the surface of the liquid. Use flexible rubber tubing of about 30-cm
length for connecting to the tap which should supply the gas at a reasonably
constant, low pressure of 10- to 80-cm water column above the atmo-
spheric pressure. Insertion of a bubble counter into the supply line is
desirable.
For saturating a liquid, first hold the stopper in the opening of the
centrifuge tube to leave a narrow gap for the escape and open wide the
gas tap. When the air has been displaced from the tube after a few seconds,
apply the stopper tightly. To promote rapid absorption of the gas, agitate
the contents by striking the tip of the tube as suggested in the preceding
section on mixing or incline the tube at an angle of twenty degrees to the
horizontal and rotate it around its axis so that the liquid is spread over
the inside surface of the tube. Shaking the contents is not recommended
since this would get them into contact with the stopper and the delivery
tube. For heating, the centrifuge tube may be inserted into a block, water
bath, or steam bath without interrupting the connection to the gas supply.
Saturation is indicated by the absence of bubbles in the counter.
Heating. For heating, immerse centrifuge tubes in water of the desired
temperature. A steam bath may be readily improvised by placing the
tube into a beaker with boiling water. The escape of gases or vapors may
be prevented by closing the tube with a suitable stopper and tying it down
with copper or aluminum wire (13). Elaborate heating blocks may be
obtained from laboratory supply houses (155, 162, 904).
Fusions may be carried out in crucibles of 1- to 2-ml capacity and
made of porcelain, vitreous silica, nickel, silver, or platinum. In the place
of a platinum microcrucible, the cover of a large platinum crucible or a
square of platinum foil may be used. If the mass of sample and flux does
not exceed 150 mg, a bead may be fused at the end of a suitable wire of
0.5- to I-mm diameter, P. 39. A small test tube, microbeaker, or centrifuge
tube of vitreous silica is well suited for pyrosul£ate fusions if care is taken
6*
84 Technique of Experimentation

to spread the melt in a thin layer over the inside surface by inclining and
rotating the tube while solidification is taking place.
In all instances, the material should be finely ground and intimately
mixed with the powdered flux before it is placed into the crucible or taken
up with a wire. The apparatus is best supported by a small triangle made
of 2- to 3-mm tubing of clay, ceramic, or vitreous silica (Vitreosil) which
need not be clear, or of platinum wire or silica rod of 1- to 2-mm diameter.
A micro burner that gives a small, blue, oxidizing flame should be used for
heating and may be shielded with a Pyrex tube of 20-mm height and
25-mm o. d. (625); a small electric muffle oven may prove convenient.
Centrifuges. The scale of work is already small enough to make efficient
use of centrifuges for the separation of phases and the collecting of matter
in the point of tubes. To-day, electric centrifuges are widely used and
generally available. Hand driven centrifuges (155, 162), however, may
be obtained in good quality from abroad. They are recommended for
field work on the centigram scale, since it is felt that centrifuging and
decantation surpass all methods of filtration in convenience and efficiency
and should not be abandoned merely for lack of suitable current. On the
milligram scale and smaller scales of work, hand centrifuges are convenient
even in the well-equipped laboratory and save time since one turn of the
crank frequently suffices to obtain the desired effect.
With the customary, gear-driven hand centrifuges, up to 2000 revolutions
per minute may be obtained; with friction couplings, Gorbach obtains up
to 40000 revolutions per minute for the centrifuging of capillaries. Comparable
speed is obtained by KIRK (157) with the use of compressed air for the
centrifuging of small tubes of up to O.I-ml capacity. Electric centrifuges
of standard design for tubes up to 15-ml capacity give from 1700 to 6000
revolutions per minute; higher speeds than these are very rarely required,
but may be obtained for work in small tubes by using "multispeed"
attachments that give up to 18000 r. p. m.
Sturdy electric centrifuges are preferable for work on the centigram
scale. Hand-driven centrifuges are recommended for work on the milligram-
and somewhat smaller scales. Their life will be lengthened if care is taken
to start them gradually. The use of shear pins of suitably soft material
in the crankshaft will save the crank which usually breaks at the notch
engaging the pin.
The relative centrifugal force is the ratio of the centrifugal force over
the gravitational force acting on the same mass. It may be estimated
with the use of a nomogram (406). It is,
relative centrifugal force = 1.2· 10-5 r n 2
when r is the radius of the orbit of revolution in centimeters and n is the
number of revolutions per minute.
 Work on the Centigram Scale 85

The centrifuge should be provided with a shield having a large diameter

in the plane of revolution. Exchangeable heads for different types of
centrifuge tubes are very useful. The microhead suggested by EMICH,
Fig. 11 b, has long slots in the arm of aluminum sheet, which facilitate
the centrifuging of long capillaries. The relatively long arms increase the
centrifugal force. The makeshift arrangement of Fig. 11 a, which uses the
aluminum shell of a clinical hand centrifuge, is less satisfactory because of
the short radius r. Heads that hold the whirling tubes in an inclined
position are not recommended since difficulties may arise when long tubes
(capillaries) have to be centrifuged.
The metal shells for microcones of O.7-ml capacity may be either open
or closed at the bottom as shown in Figs. 11 band c. The open shells are

/rubbe/'
b"mf

1
" Col/OI7
b
Fig. II. Centrifuging of Microcones; about 1/2 nat. size

very practical since they allow placing the tip of tubes at a maximum
distance from the center of revolution so that the length of the glass
apparatus is mainly limited by the radius of the shield of the centrifuge. The
apparatus may be supported by the rim of the shell or at the bottom
open'ng of it; rubber rings may serve for cushions in either instance and,
to this end, are cut from rubber tubing of suitable dimensions. Closed
shells may be used for floating the glass cones in water during swirling to
definitely exclude the possibility of breakage. Microcones of heavy Pyrex
tubing break rarely, however, and the danger may be further reduced
by the use of cushions below the st~rdy flanged rim and (or) underneath
the tip.
Centrifuge tubes of 3-ml and larger capacity are usually centrifuged
in closed shells. If they are of soft glass or thin glass, it is best to fill the
space between tube and shell with water so that they float during whirling.
Should a tube collapse, the shell should be immediately and thoroughly
86 Technique of Experimentation

cleaned to rinse out corrosive substances and to remove glass splinters

that might otherwise cause the destruction of some more centrifuge tubes.
Use of Centrifuges. The momenta acting in diametrically opposite
directions of the centrifuge head must be approximately equal to prevent
the appearance of a destructive force. The requirement is most conveniently
met by a similar distribution of equivalent masses on opposite arms.
When working with centrifuge tubes of 3-ml or larger capacity, place
the shells containing the tubes on opposite pans of a trip scale and add
water to the lighter combination until equilibrium is established; depending
upon the requirements of the task, place the water into the centrifuge
tube or into the space between tube and metal shell. Less care is needed
with the light microcones of O.7-ml capacity and smaller apparatus; as
a rule, it is sufficient to place tubes of equal size into the opposing shells.
Close the cover of the shield before starting the centrifuge and do not open
it before the centrifuge has come to a stop. Start and stop the centrifuge
gradually; sudden changes of speed are not only destructive but also
prevent the clean separation of phases, especially in apparatus of large bore.
Immediately stop the centrifuge if unusual vibration indicates lack of
balance.
Separation of Solid and Liquid Phase. The separation is best accomplished
by centrifuging and decanting. As a rule, it will suffice to centrifuge for
about one minute. The solid frequently becomes so tightly packed that
decantation may be accomplished by simply pouring off the supernatant
liquid. If some solid floats upon the surface of the liquid, it may be combined
with the bulk by adding a few drops of alcohol and whirling again. The
alcohol is added at the rim of the tube so that it rinses down the inner
surface of it.
If clean separation of the phases is needed and cannot be obtained
by pouring, lift off the liquid by means of a capillary siphon, Fig. 31, a
suction-operated siphon, Fig. 32, a suitable pipet, or the capillary siphon
pipet, Fig. 12. In all instances, solid floating on the surface of the liquid
usually collects upon the interior surface of the centrifuge tube and upon
the outside of the inserted tube. If difficulties should arise, one may try
to wet the tip of the siphon or pipet with alcohol before inserting it into
an aqueous liquid contained in the centrifuge tube. The change in surface
tension when the tip of the pipet makes contact with the liquid in the
centrifuge tube may sink some of the solid particles floating on the surface
and drive the rest to the wall of the tube.
The Capillary Siphon Pipet (868), Fig. 12, is shown in A in its simplest
form which may be readily made in the laboratory; B may be further
modified by placing a filter mat into the somewhat widened orifice of the
intake tube and doubling the height of the heavy-walled arm of the siphon
to obtain stronger suction. The apparatus is best clamped to a stand that
 Work on the Centigram Scale 87

permits raising and lowering by mechanical means, and also the centrifuge
tube may be 1.leld with a clamp below the rim during decantation. Suction
with the mouth is only needed to start the operation of the siphon; the
collected liquid is finally transferred by either expelling it through the
siphon or draining it off the bulb, whereupon the device is ready for the
collection of washings.
Pipets are convenient for the removal of aliquot parts of a liquid above
a precipitate, but they are not the most practical tool if a neat separation
by decantation is required. For the latter purpose, the pipet should have
a capacity sufficient to take the whole of the liquid at one time. Sucking

Fig. 12. Capillary Siphon Pipet; 1/3 nat. size. CLARKE, B . L., and H. W. HERMANCE,
Mikrochemie 10, 289 (1935)

with the mouth is not practical, but the pipet could be connected to an
aspirator bottle. Pipet controls of the syringe type and made of glass
or metal with capacities up to 10 ml may be obtained from supply houses
and may permit more convenient regulation than control bulbs of rubber.
Pipets may be prepared from tubing of 6-mm to 12-mm bore, which is
simply drawn out to thick-walled capillaries to furnish the tip and the
stem on both sides of the bulb consisting of a suitable length of the original
tubing.
Clean separation requires that the solid is thoroughly compacted so
that it may be closely approached with the orifice of the siphon or pipet.
When the bulk of the liquid has been removed, the centrifuge tube is spun
again to collect liquid adhering to the walls which is then also drawn off
and combined with the bulk.
Washing the Solid. Thoroughly mix the solid with the wash liquid,
centrifuge, and decant.
88 Technique of Experimentation

Transfer of Solids. Precipitates and residues are usually treated in

the centrifuge tube in which they have been collected. If transfer to another
apparatus is required, stir to a slurry with a liquid (reagent to be added
next or solvent that does not interfere or may be readily removed) and
transfer the slurry with a rapidly operating pipet or siphon to prevent
the settling out of the solid. At times, it may be found possible and
convenient to get rid of most of the transfer liquid by allowing the solid
to settle into the tip of a pipet which may then deliver all the solid with
the first drop(s) of discharge.
Obviously, if a large amount of solid has been collected, portions of
the moist cake may be transferred with a small spatula.
Extraction of Solids may be performed in a centrifuge tube with the
procedure for washing solids. One may heat the solid with each portion

Fig. 13. Stork's Bill Designed by G. GORBACH; about 1/3 nat. size

of solvent and decant the hot solution. Apparatus for the automatic
extraction of solids may be readily improvised with the use of either the
percolating or the siphoning (Soxhlet) principle. Many such devices have
been described (147, 155, 162), but special attention should be given to
the comparative study of BATT and ALBER (428).
Liquid-Liquid Extraction may be performed in the centrifuge tube as
described for the milligram scale, p. 116. Separatory funnels have been
designed by ALBER (431) and CHERONIS (147). KIRK (157, 443) describes
extractors for heavier-than-water and lighter-than-water solvents; a
syringe extractor (451) may be adapted to this scale and would serve for
both types of solvent. The separation obtained with the simple storkbill
of GORBACH (155), which may be blown from tubing of 6-mm to 8-mm bore,
will often suffice. The solution and the solvent are taken up with the aid
of an aspirator and the apparatus in position of Fig. 13a. In the same
position, the lighter liquid may be decanted by suitable tilting of the tube.
Better control is had when draining the heavier liquid with the apparatus
in position Fig. 13b. While mixing the phases, the tip should be closed
with the finger or a cap. Suitable apparatus for continuous extraction
have been described (45, 147), but will rarely be needed.
 Work on the Centigram Scale 89

The simple apparat.us of KIRK and CHARLOTTE L. BROWN (937, 952)

for solvents heavier-than-water gives a double transfer in one operation.
A sealed tube st with iron core is placed into the bottle of the apparatus
shown in Fig. 14a, and 15 to 40 ml of heavier-than-water solvent or solvent
mixture is added. About 22 ml of the aqueous solution to be extracted
is introduced into the outer space 0 through the side arm, and 5 ml aqueous
solution is placed into the inner tube i, which opens 6 to 7 mm above the
bottom of the bottle. Obviously, the solution in the inner compartment

o i o

(J I!)cm

Fig. 14. a Extractor for Heavier-than-Water Solvents; KIRK, P. L. , and CHARLOTTE

L. BROWN, Mikrochimica Acta 1957, 715. - b Distilla tion in the Centrifuge Tube

must contain a reagent which should be insoluble in the organic solvent

and which strips the solvent of the extracted matter. The apparatus is
placed upon the top surface of a "magnetic stirrer" and the stirring bar
on the bottom of the bottle is kept ~n rotating motion for 3 hours. If the
distribution ratios are suitable, the substance of interest will collect in
the aqueous phase of the inner compartment.
The apparatus was used to isolate small amounts of basic or acidic
compounds in a state of reasonable purity from complex biological materials.
The pH of the test mixture was adjusted to liberate the acids or bases so
that they could be extracted by chloroform, carbon tetrachloride, or
mixtures of the two. The inner compartment held I-F ammonia or 0.1- to
I-F hydrochloric acid. Thus, barbiturates and alkaloids could be isolated
with yields of usually better than 50 %.
90 Technique of Experimentation

Evaporation may be performed in a centrifuge tube of chemically

resistant glass by heating in a steam bath or block and blowing a stream
of filtered air from a capillary on the surface of the liquid, Fig. 28. Also
suitable are crucibles and small dishes of porcelain, vitreous silica, and
platinum, which may be heated on some bath, block, or hot plate or by
means of an infrared lamp (13, 147). A jet of clean air blown across the
surface of the liquid will in all instances materially speed evaporation and
keep away dust.
Distillation. A variety of distilling apparatus have been described (146,
147). For analytical tasks, the simplest designs are preferable.
If the volatile matter of interest is carried over by a relatively large
amount of other vapors, apparatus of the type shown in Fig. 9b will be
suitable if made in the proper dimensions. Distillation in the centrifuge
tube as described by Boos (141, 1251) allows simple and complete recovery
of distillate and residue. Place the solution to be distilled into the small
tube a, Fig. 14b, which may be graduated and has a handle attached to
its rim. By means of a hook at the end of a glass rod, lower it into the
centrifuge tube b so that the rim comes to rest upon the three indentations
projecting into the interior of the large tube. Insert the internal heater c
into the solution in the small tube. The illustration shows an enlarged
view of its essential part, a platinum wire of 0.6-mm diameter which is
filed down to 0.2 mm at the point of the V. Connect through a 1- to 5-volt
variable step-down transformer to the 1l0-volt a-c line. Immerse the
large centrifuge tube into a suitable cooling bath (ice water, ice and NaCI,
etc.) up to the cork, and give enough current to produce slow boiling.
Always keep the point of the heating wire immersed in the liquid. To
finish the distillation, shut off the current and leave the apparatus for
two minutes in the cooling bath. Then remove the heater and centrifuge
the large tube with the small one in place; finally lift out the latter by
means of the hooked rod.
Apparatus shown in Fig. 15 may be used when the residue of the
distillation contains solutes that must be collected for investigation.
Ground glass stoppers may replace the indicated corks. For good separation,
it is essential to heat so that the ring of condensate rises very slowly through
the vertical or nearly vertical tubing and finally remains stationary at
the level of the side arm (Fig. 15a and b) or at the level where the fraction
is collected with a capillary pipet (c). A tiny microflame may be used for
heating and may be shielded with a small chimney made of glass tubing
and partly closed at the bottom with a disk of cardboard or metal foil.
A boiling capillary, d, is placed into the liquid to obtain even boiling,
or a platinum wire of 0.05- to O.I-mm diameter is inserted through the
bottom, Fig. 15c.
 Work on the Centigram Scale 91

The simple apparatus (1015) a may be held in hand during distillation;

the ring of condensate is kept at the height of the side arm until the desired
amount of condensate has collected in the knee of the latter, from where
it is taken up into a capillary pipet. Several fractions may be obtained.
The residue of the distillation is also collected with a pipet. Apparatus b (872)

em
2

o
o

o s IOem

Fig. 15. Di tillation on the Centigram Scalo

uses an indented column which is easily made in the laboratory and gives
less hold-up than packed columns. Column b as well as tube c may be
surrounded by a glass mantle.
The fractionating tube c may be used like EMICH'S (below) and whirled
between fractions if a suitable centrifuge is available. Vuring distillation,
it is best clamped to a stand. The fine tips of the capillary pipets may be
bent, and the fractions should be collected from the topmost part of the
ring of condensate when the latter ceases rising.
92 Technique of Experimentation

A thermocouple might be used in the top of column b. As a rule, it

seems more practical to determine the boiling points after collection of
the fractions, Expt.21. All apparatus of Fig. 15 could be connected to
a vacuum line for distillation under reduced pressure, but the simple
apparatus of BABCOCK (146, 147, 445) may be preferable for occasional
analytical use.
Emich's fractionating tube, Expt. 20, is recommended for the separation
of a mixture of liquids which does not leave a non-volatile residue that
requires investigation. The technique is readily modified for the isolation
of traces of very volatile liquids (861, 1154).
Sublimation is frequently performed to obtain well developed crY8tal8
for crystallographic study. Crystals developed in three dimensions are
obtained by very slow sublimation at temperatures far below those usually
considered as sublimation temperatures (sulfur, arsenic trioxide, vanillin,
and alizarin at 50 C; mercuric chloride, benzoic acid, and caffein at 20 C;
0 0

indigo, morphine, quinine, and strychnine at 100 0 C, etc.). Even when

the condensing surface has nearly the same temperature as the heated
material, very small quantities (10 flg and down to 0.001 flg) of substances
may be isolated in distinctly crystalline form if the condensing surface is
very close to the heated material.
R. KEMPF (1142) described an electric heating block with elaborate
temperature control, whic"i} can be used also in the vacuum under a bell jar.
A simple metal heating block will serve, however, for sublimation under
atmospheric pressure. A very small amount of the substance (drug, textile
fibers, paper) is ground up with a drop of water or some other suitable
liquid and transferred as a thin slurry to the top surface of the heating
block to cover an area of 5- to 10-mm diameter; as an alternative, a solution
of the material may be applied. After evaporation of the liquid, a cover
slip or a microscope slide is placed upon the residue to receive the sublimate.
The temperature of the block is adjusted so that the sublimation requires
from 4 to 48 hours. Cooling is required only when the sublimation is
performed below 100 C. To this end, the flat bottom of a wide metal
0

tube with circulating cooling fluid is placed on top of the slide or cover slip.
The sublimation of indigo at 180 to 200 C from a few fibers of dyed
0 0

cloth is recommended for practicing the technique.

To obtain very thin tablet8 for the measurement of profile angles,.
SHEAD (633, 958) uses a sublimation cell consisting of two standard micro-
scope slides separated by a glass ring, 4 mm high and 16 mm in inner
diameter, which is heated on a metal bar to a temperature giving a medium
rate of sublimation (l0° to 50 below the melting point of odorous solids
0

and 80 0 to 150 0 C below the melting point of non-odorous solids). The

surface of the receiving slide is slightly greased with oil from the fingers
or face and then wiped to leave only an invisible film of it. When the cell
 Work on the Centigram Scale 93

and the receiving slide (which may be placed on the bar for heating it up)
have attained the sublimation temperature, a small quantity of the material
is placed upon the bottom of the cell which is then quickly covered with the
warm receiving slide. Relatively large, thin crystal plates are obtained
within two to ten minutes.
Various apparatus have been described (147), which may be used for
the purpose of analytical separation. To gain time, vaporization at relatively
high temperatures is used, and the collecting surface is usually cooled to
obtain good recovery of the volatile substances. The temperature gradient
becomes especially steep in those instances where the distance between
charge and condensing surface is small. CHERONIS and RONZIO (147)

r lr
CoMn! w~!~r
r
o IOcm

Fig. 16. Sublimation According to SOLTYS, Mikrochemie, Emich·Festschrift 275 (1930)

recommend a distance of 7 mm to 25 mm for sublimation under reduced

pressure of 1 to 20 mm mercury column. The charge should be introduced
as a fine powder to obtain a maximum of surface; if a non-volatile film
forms on the surface and sublimation stops, the pulverizing may have
to be repeated (or a drop of suitable solvent may be added and the sublima-
tion resumed after the solvent has evaporated to leave the volatile matter
behind, spread as a thin film over the heated area). Sublimation should
be carried out at least 10° C below the melting point if the latter is between
40° and 100°; or 50° to 80° below, if the melting point is between 100°
and 200 °; and 100° to 150 ° below, if the melting point is above 200°.
High temperatures and high rate of sublimation may, however, give poor
separation of fractions and may have to be adjusted according to the needs
of a given task.
Sublimation with a current of carrier gas through a straight tube of
suitable dimensions, p. 74, satisfies the requirements of analytical separa-
tion. The apparatus of SOLTYS will prove quite efficient for the complete
separation of one volatile fraction from a non-volatile residue. The more
elaborate procedure of GETTLER, UMBERGER, and GOLDBAUM provides for
the convenient separation of several volatile fractions.
94 Technique of Experimentation

SOLTYS (855) places the finely powdered substance upon a disk of

coarse fritted glass, sealed into the tube, and holds it in position by pressing
upon it a disk of asbestos or glass-fiber paper, which fits snugly into the
tube. The tip of the cold-finger condenser is brought close to tlte other side
of the fritted-glass disk. Air or inert gas is admitted through a capillary
which is so fine that a rapid stream of bubbles of I-mm diameter is obtained
when blowing air through it into water. The tube is heated in a metal
block with thermometer, and a vacuum of 20 mm is applied. The capillary
may be replaced by a glass tubing with cotton filter for sublimation in
a gas stream of 1 to 2 ml per second. Sublimation by atmospheric pressure
requires more time and efficient cooling and gives fluffy sublimates of
crystal aggregates, whereas a dense crust is obtained on the condenser
when applying a vacuum.
A relatively simple procedure which should lend itself to application
to much smaller quantities has been described by FLASCHENTRAGER
et al. (930). Quantitative separations are obtained by high vacuum
sublimation in a tube of 8-mm inner diameter and 17- to 20-cm length,
which is sealed at one end. After a preliminary heating with application
of high vacuum, 10 to 20 mg of the weighed material are brought to the
sealed end by means of a charging tube (13). Separation is then accomplished
by suitable heating in a metal block of 17-cm length, whereby more and
more of the tube is withdrawn from the block as the temperature is raised.
The fractions are isolated by cutting the tube into sections.
GETTLER, UMBERGER, and GOLDBAUM (447) obtain simple recovery
of sublimate fractions and residue by using a 14/20 ground joint for
separating the lower part of the sublimation tube, which contains the
charge, from the upper part and by lining the inside surface of the latter
with a transparent sheet that catches the sublimates and may be easily
withdrawn from the tube.
The lining is obtained by cutting a strip of 4-cm width and 15-cm length
from a roll of cellulose sausage casing obtainable from the Visking Corpora-
tion (Chicago, Ill.). The strip is rinsed in ethyl ether, dried, rolled around
a glass rod, and thus inserted through the end of the tube, which carries
the male part of the ground joint, so that about 6 mm of the strip remain
outside the tube. The rod is then removed, and the strip is grasped by the
end projecting from the ground taper and rotated clockwise until it for.ms
a helix inside the tube. This is followed by counter clockwise twisting so
that the coil partially unwinds, expands, and wedges tightly against the
wall of the tube. The sheet is then trimmed flush with the lower end of
the tube.
The upper end of the sublimation tube is connected to a vacuum line
able to give 1 mm pressure and less. After introducing the charge into
the lower part, the ground glass joint is connected while the vacuum is
 Work on the Centigram Scale 95

turned on. Heat is applied so that the temperature of the block rises about
two degrees per minute. When a sublimate appears, the temperature
is held constant as long as the sublimate increases. Then the temperature
is again gradually raised until another fraction sublimes, etc. In this
manner, the components of a mixture may be obtained at different levels,
especially when a preceding trial sublimation has shown the best rate of
heating, etc.
When the sublimation is finished, the upper tube is separated from
the lower, and the film is removed by hooking it with a dental probe.
em
!(I

I
~

I
o

Fig. 17. Sublimation According to GETTLER, UMBERGER, and GOLDBAUllI, Analyt.

Chemistry 22, 600 (1950)

After withdrawal from the tube, it is unwound and fastened with thumbtacks
to a cork pad for a few hours to make it lie flat. The fractions may be
separated by cutting the film. Crystals may be studied on the film, but
must be transferred to glass slides for inspection in polarized light because
the film is anisotropic. Also chemical treatment is possible on the film
which may be heated to 300 C and is resistant to most solvents, alkalies,
0

and even cold concentrated acids. It dissolves in cellulose solvents such as

acetone and ethyl acetate.
SCHMIDT (955) uses a film of polyterephthalic ester (Hostaphan of Kalle
in Wiesbaden, Germany) rolled into a simple cylinder and a heating device
(commercially available, Willi Gunther, Nurnberg) producing a temperature
gradient along the tube. It is pointed out that the identification of the
sublimates is facilitated by the oriented growth of the crystals upon the
96 Technique of Experimentation

anisotropic film. Sublimation in the horizontal tube has also the advantage
that crystals of the sublimate, which separate from the supporting film
as a consequence of recrystallization, cannot drop back to the sample.
Liberation of Gases for identification may be performed in one of the
several simple apparatus developed by FEIGL (121) and his coworkers.
Of these, the one shown in Fig. ISc is readily assembled in the laboratory.
The sample is treated with the reagent at the bottom of the small beaker;
and the small pieces of moist test paper are placed upon the outer surface
of the spherical bulb. As an alternative, the latter may be moistened with

u u u

• c:> I b
I
IDem

Fig. 18. Libera t.ion and T esting of Gases ; e is enlarged five times as compared with
t.he other apparatus

reagent solution or absorbent solution which is later rinsed into the depression
of a spot plate or taken up by spot test paper. The closed apparatus is
either allowed to stand at room temperature or gently warmed.
The gas reaction cell, Fig. ISb, may be assembled from a dish for the cup
grease test (Corning 3210) and a rectangular glass plate. The transfer of
the gas is left to spontaneous action as in the apparatus of FEIGL, but
spray may be prevented from getting to the reagent. The bottom of the
dish is treated so that drops will not spread; ruling squares with paraffin
lines on clean glass is practical, but making the surface repellent renders
difficult the deposition of drops. The sample and a drop of reagent for
the liberation of gas are placed upon the bottom of the dish so that they
will not mix. Into the reagent drop is placed a piece of sealed capillary
 Work on the Centigram Scale 97

containing a 6-mm length of thin iron wire, shown magnified in Fig. 18e.
Small pieces of reagent papers cut into triangles, squares, and rectangles
of 1- to 3-mm2 area are moistened and pasted to the clean underside of
the glass plate forming the cover of the cell. Small reagent drops are
deposited upon this surface. After the cell has been closed, a strong magnet
(Alnico) is applied to its underside, and the reagent drop and sample are
wiped together by suitably moving the capillary containing the wire.
A blank test is readily obtained by allowing the closed cell to stand
for some time before combining the reagent with the test drop. The change
taking place after adding the reagent gives convincing proof that the active
agent comes from the test drop. The transfer of spray is prevented by
placing into the cell a partition made from a strip of filter paper or glass
fiber paper, which is as long as the diameter of the cell and nearly as wide
as the cell is high. The strip is folded in the middle to give a right angle
and then placed into the cell so that the legs of the angle take the positions
of the hands of a clock at 3 hours. Test drop and reagent drop are deposited
on one side of the partition; the reagent drops and the test papers, on the
other. The strip of paper may also be treated to absorb interfering gases.
The more elaborate apparatus, Fig. 18a is recommended if spray must
be kept away from the reagents or the gas is only slowly given off by the
reaction mixture. A capillary is used to place the reagent which liberates
the gas into the tip of the tube, and the short piece of capillary E with the
sealed-in iron wire is added. A droplet of the solution to be tested is
deposited upon the inner surface of the tube, about 15 mm above the reagent
solution. If the sample is a solid, a droplet of water is deposited first,
and then a particle of the solid is transferred to this droplet. Capillary
pipets and fine glass rods with the tips bent at an angle, Fig. 18d, are
suited for these tasks; the interior of the tube may be made liquid repellent.
Scrubbed air or inert gas is introduced through a fine capillary of
0.05- to 0.2-mm bore, which is broken off to the proper length so that it
just reaches the bottom of the conical end of the tube. A very satisfactory
supply ot air is obtained by securing with wire a 2-hole rubber stopper
carrying two inlet tubes in the opening of a 10- to 20-liter bottle which
may contain about 100 ml of I-F NaOH. A short piece of rubber tubing
connects a rubber bulb (bellows) to the inlet, and the outlet is provided
with a long flexible rubber tubing carrying a screw clamp. A supply of
air pressure built up with the rubber bulb is sufficient to maintain a fine
stream of air bubbles for a long time.
Into the bottom of the wide funnel of the apparatus is dropped a disk
of filter paper or glass fiber paper for catching spray. A strip of paper,
cotton, or glass wool treated with reagent for the removal of interfering
gases may be placed below the paper disk. Small pieces of test papers
may be attached moist to the inner surface of the funnel. Reagent droplets
Benedetti-Pichler, Identification 7
98 Technique of Experimentation

are more advantageously deposited upon the underside of a cover slip

placed upon the opening of the funnel.
A blank test may be performed by bubbling gas through the apparatus
for some time before using the glass-enclosed iron wire for wiping together
the sample and the reagent solution. The stirrer is operated by means of
an Alnico magnet applied to the outside of the tube. The apparatus may
be made in the laboratory, whereas the more elaborate one of MALISSA (162,
894) requires the skill of a glass blower.
Confirmatory Tests may be carried out in small test tubes and centrifuge
cones. As a rule, one prefers to use a test plate or spotting on filter paper.
These procedures have the merits of simplicity and speed, but the techniques
described for use on the milligram scale are preferable when reliability
must be the first consideration.
The test plates are made of white or black, glazed porcelain or of chemically
resistant glass. In the latter instance, the background may be readily
changed for observation by laying the plate upon paper or tile of suitable
contrasting color. The customary depressions in the top surface of the
plates are convenient but not essential.
It is important to clean the test plates immediately after use. After
drying, the top surface should be made liquid repellent to prevent the
spreading of the drops. This may be accomplished by wiping it with a
cloth or wad of cotton that has been moistened with a mixture of ethanol
and diethyl ether.
Drops of the test solution and of the reagents are usually transferred
to the plate by means of stirring rods or medicine droppers. To avoid
contamination of these tools and to render unnecessary their cleaning after
use, the drops should be deposited side by side without merging, which
is easily accomplished if the drops do not spread. Use of a rod for mixing
may be avoided since combining of the drops and mixing may be accomplished
by blowing the air from a capillary tangentially onto the side of the drops.
Difficulties may arise when organic solvents are used and the drops
spread. Thus precipitates and colorations may be carried up to and over
the rim of the depression so that it becomes necessary to interpret the
outcome of the test by comparison with blanks and controls (tests with
known amounts of sought substance).
For the performance of spot tests on paper, it is recommended to place
a strip or square of it upon the opening of a small beaker, porcelain crucible,
etc. and to deposit drops of test solution or reagents upon the center of the
horizontally supported sheet so that they may spread in the paper without
hindrance. The drops are transferred by means of stirring rods, medicine
droppers, or pipets. It is obvious that spot test paper must be handled
and stored in a manner that excludes contamination.
Expt.19 Work on the Centigram Scale 99

Infrared heat lamps are recommended for the warming of spot tests
on test plates (432) and paper, for the drying of test papers, etc.
Storing Material Under Investigation is simple. The centrifuge tube
containing the liquid, solid, or mixture of both is closed with a suitable
stopper, or a small beaker or wide vial is inverted and placed over the
opening of the tube. The centrifuge tube is placed into a suitable block
or heaker to keep it in upright position, possibly under a bell jar. Precipitates
which oxidize in air or age and change their solubility should not be stored,
but dissolved without delay. Solutions in volatile solvents may be sealed
into glass containers; the tube containing the solution may be slid into a
test tube which is then fused shut.
Cleaning Apparatus. Feathers and pipe cleaners will be found useful
for the cleaning of centrifuge tubes and similar apparatus. A suitable
suction device (p. 124) may serve for rinsing. For quick drying, apparatus
may be inverted over a vertical glass tubing from which emerges hot air
coming from a heated aluminum box or coil.

Experiment 19
Preparation of Oapillaries, Oapillary Pipets, and Microburners
Supply of glass tubing of 6- to 8-mm bore. Tubing of 12- to 18-mm bore
may be used to advantage if a strong blast flame is available. The tubing should
be tested by drawing capillaries of about 0.2-mm bore and bending them to get
an estimate of the elasticity. For most work, one will prefer a glass that is not
brittle like Pyrex. Certain kinds of soft glass are well suited.
Cut the tubing into suitable lengths, and then clean it outside and inside
with soap solution and a wad of cotton tied into the middle of an aluminum
wire (or suitable brush). Rinse with tap water and distilled water. Finally
wrap bundles of the tubing into paper towels and store them in an upright
position for draining and drying.
For drawing capillaries, grasp a piece of tubing at the first and third
quarter of its length. While rotating it, heat its center just above the inner
blue cone of the large roaring flame of a Tirrill burner (Meker or Fisher
burner). If the heating is uniform, the flame will become luminous at the
same time along the whole length of the heated part of the tubing. When
the glass has become soft, remove the tubing from the flame; continue
rotating it and leisurely draw it out to a capillary of 0.5- to I-mm bore.
The rate of drawing determines bore and length of the resulting capillary.
Watch closely while drawing and regulate the rate according to requirement.
Do not release the pull until the glass has hardened since otherwise bent
capillaries will result.
With the sharp edge of broken porcelain, cut the capillary so that
6 cm of it remain with each piece of the original tubing. These short pieces
7·
100 Technique of Experimentation Expt. 19

serve as handles when drawing the next length of capillary fro~ the portion
of the tubing adjoining the taper. The procedure gives lengths of capillary
separated by small bulbs. Cut the capillaries so that 6-cm lengths of
capillary remain on either side of the bulbs which are saved for use as
pipets. Seal the capillaries at both ends and keep them in a box or tube
until needed. - The wide ends of the original tubing, each with a 6-cm
capillary attached, may be fused together to get a handle on each side
and then drawn into capillaries.
Several meters of capillary may be obtained by one drawing from tubing
of 12- to 18-mm bore. Heat the tubing in a large blue blast flame until soft.
Then remove the tubing from the flame and hold one end steady while
a second person walks away with the other end.
Preparation of Oapillary Pipets. Select a piece of capillary of 0.5- to
I-mm bore and 20-cm length. Heat the center of it in the seam of the

a, lJ
I I
~-----.
I
o
Fig. 19. Capillary Pipet; approx. nat. size. The tip is shown three times enlarged

lowest part of the Bunsen flame, just above the barrel, until the glass
just softens. Then remove it from the flame and pull with a quick motion
so that a fine capillary of 0.05- to 0.2-mm bore and about 10- to 20-cm
length results. If the fine capillary can be bent into a loop without
breaking, it probably has the desired diameter.
Break the fine capillary about 2 cm from the tapered portions to get
two pipets, Fig. 19. Finally, for a crude estimation of the bore of the fine
capillary, lay the pipet upon a slide and inspect under the microscope with
a total magnification of about 30 diameters. Measure the bore at the
fine orifice with the eyepiece micrometer.
Microburner. Draw out one end of Pyrex tubing, 6 to 8 mm in diameter
and about 15 cm long, to a capillary of about I-mm outer diameter. Cut
this capillary at a point 2 to 3 cm from the taper and bend it as desired.
The wide tube may be held in a clamp, or two glass prongs may be fused
on near the taper to serve for legs when the tube is placed upon the bench.
Flexible rubber tubing is used to connect the wide tube to the gas tap.
For adjusting the size of the flame, place a loose plug of cotton inside the
rubber tubing and apply a screw clamp at the location of the cotton plug.
The size of the microflame is determined by the orifice of the capillary.
To obtain a very small flame, proceed as follows: light the microflame
Expt.20 Work on the Centigram Scale 101

and open wide the tap to get the maximum height of the flame. Then
touch the orifice of the capillary to the edge of a roaring Bunsen flame
(oxygen blast flame if the glass has a high softening point) until the height
of the microflame is reduced to 5 mm. Regulation with cotton plug and
screw clamp will then permit to obtain a flame the size of a pinhead.
Concerning a simple shield for the microflame, made from a 20-mm
length of Pyrex tubing of 25-mm outer diameter, see STOCK and FILL (625).

Experiment 20
Emich'8 Method of Fractional Di8tillation (1011)
Fractionating Tube and Handle. - The fractionating tube, Fig. 20, is made
of Pyrex tubing of 4-mm bore. A slight constriction separates the two chambers,
each of which is 3 em long. The lower chamber is half filled with asbestos which
is briefly ignited before it is introduced into the tube and pressed together.
From the bottom of the tube is drawn a solid spike of glass, which snugly fits
into the handle, a piece of Pyrex tubing, 15 em long and 3 mm in bore.
Mixture of equal volumes of ethanol and water or of equal volumes of acetone,
carbon tetrachloride, and benzene or toluene.
Drying the Fractionating Tube. Place the tube in its handle and, holding
and rotating it as indicated in Fig. 20, heat it in a just non-luminous large
Bunsen flame to 200 to 300 C. When dry, place the handle into the
0 0

opening of a block to hold it and tube in perpendicular position for cooling.

Di8tillation. Place a pencil on a sheet of paper to rest upon it the tips
of about ten capillary pipets intended for the collection of the fractions.
To avoid confusion, write the numbers of the pipets upon the paper where
they rest on it.
Take up a 4-cm length (0.03 ml = 30,ul) of the liquid to be distilled
into a capillary of I-mm bore and touch the end containing the liquid
to the asbestos in the cold fractionating tube. When the liquid has been
taken up by the asbestos, withdraw the capillary and centrifuge any
droplets adhering to the walls of the tube into the asbestos; see Fig. II a and c.
To perform the distillation, insert the spike of the fractionating tube
into the handle. Rotate the latter in the left hand and heat above a just
non-luminous Bunsen flame of 2-cm height as shown in Fig. 20. Incline
the tube at 45 degrees and hold it so that the top surface of the asbestos
is about 4 cm above the tip of the flame. Watch the portion of the tube
above the asbestos for the appearance of a ring of condensate. If the
heating is properly done, this ring will rise slowly up the lower chamber.
Remove the tube from the flame when the condensate reaches the constric-
tion, and wait until it has passed into the upper chamber. Stop rotating
the tube and hold it nearly horizontal. Insert the tip of capillary pipet No. I
into the uppermost portion of the condensate, and allow the liquid to rise
until its meniscus arrives about I mm above the taper of the capillary.
102 Technique of Experimentation Expt. 21

Storing the Fraction. Hold the capillary pipet containing the fraction
slightly inclined to the horizontal so that the tip points upward, and touch
the very end of the latter to the lower part of the seam of the Bunsen flame.
If the tip is very fine, the liquid quietly recedes and the end of the capillary
fuses to a small sphere. If the tip is too wide, there may be some sputtering
before the sealing takes place. To make certain, place the pipet upon a
slide and inspect the sealed tip under the microscope. There usually is
a gas bubble or several of them in the fine capillary, which is essential for
the determination of boiling point, Expt. 21. If the tip is properly sealed,
return the pipet to its place upon the sheet of paper.

II
Fig. 20. Fractional Distillation According to EMICH

Repetition of Distilla,tion. Allow the fractionating tube to stand until

it has acquired room temperature. Then centrifuge it briefly to collect
all liquid in the asbestos, and repeat the distillation as before to collect an
additional fraction. Repeat this until no more liquid is obtained when
the tube is heated. Six to twelve fractions will be obtained in this manner.
These are used for the determination of their boiling points in Expt. 21
and any additional observations needed for identification.
Any non-volatile residue will remain with the asbestos which should
be removed with a probing wire and replaced by a fresh batch after fractiona-
tion of an unknown mixture.

Experiment 21
Emich's Method for the Determination of Boiling Points (1007)
Pyrex beaker, tall form, 250·ml capacity; stirrer, Fig. 21; thermometer for
melting point determinations; standard microscope slide; rubber band which
may be cut from wide tubing; bath liquid for melting point determinations.
Expt.21 Work on the Centigram Scale 103

The boiling point may be determined with 0.2.ul of liquid and less.
As described in the preceding experiment, the liquid is taken up with the
fine tip (0.05- to 0.2-mm bore and about 15 mm long) of a capillary pipet
of about 0.5-mm bore in the wide part. The amount of liquid should be
enough to fill a I-mm length of the wide capillary so that the drop will
not break while rising. The gas (air) bubble in the fine tip may be very
small; it may fill several millimeters of the fine capillary, but it must not
extend into the tapered portion of the pipet.
Set up the melting point bath of Fig. 21. Insert the top of the thermom-
eter into a cork and, using a rubber band, attach the boiling point capillary

t-

-
Fig. 21. Determination of Boiling Points According to EMleR. The enlarged tip of a
sealed boiling point capillary is shown at the right

to the bulb end. To determine in one heating the boiling points of several
liquids or fractions, attach a microscope slide to the bulb end of the ther-
mometer as shown in Fig. 21. Arrange the boiling point capillaries on the
slide in their proper order on both sides of the thermometer and at a height
that the drops in the capillaries are level with the center of the bulb. When
the apparatus has been set up as shown in Fig. 21, pour the bath liquid
into the beaker; 30 % sulfuric acid will do up to 108 0 C.
Adjust the illumination until the mercury in the thermometer and
the droplets in the capillaries can be seen without difficulty. Then heat
the bath rapidly with stirring until the boiling range of the droplets
(fractions) has been approached. At this time, reduce the size of the flame
to get a temperature rise of 2 C per minute and begin to stir continuously
0

so that the loop of the stirrer moves up and down through the whole height
of the bath.
104 Technique of Experimentation

Keep an eye upon the droplet representing the lowest boiling liquid
(fraction). When the boiling point is approached, the droplet begins to
quiver, and the rate of stirring should be increased. The droplet finally
rises through the capillary, and the temperature is read when it arrives
at the level of the surface of the bath. With an essentially pure liquid or
an azeotropic mixture, this happens at the boiling point. A mixture may
be recognized by the fact that its composition and boiling point changes
as it rises in the capillary, i. e., is lifted up by the vapor given off into the
gas space below the droplet. In such instances, the temperature at which
the droplet starts to rise and the temperature at which it arrives at the
bath surface should be recorded; the two may be many degrees apart.
Continue the experiment until all drops have arrived at the surface of
the bath. Then remove the flame. When the temperature of the cooling
bath approaches the highest boiling point observed, resume stirring and
record the temperatures at which the droplets begin their descent. The
experiment may be repeated until the droplets become so small, due to
distillation to the cooler parts of the capillaries, that they break. At the
conclusion of the experiment, the liquids may be saved by fusing shut
the ends of the wide capillaries and collecting the liquids at these ends by
whirling in a centrifuge.
GARCIA (436) has described a modification of the method, that permits
determination of the boiling point under reduced pressure with 2 to 5,u1
of liquid.

Work on the Milligram Scale

Reagents. A large number of solid reagents may be kept in small vials
systematically arranged in a Behrens reagent box. Plastic vials are
recommended for fluorides and strongly alkaline substances. It is advisable
to supply the solids as granules of 0.1- to I-mm diameter. Larger crystals
should be crushed in a mortar and then classified by spreading and tapping
the material on a sheet of paper and rejecting the very coarse and the fine
material.
Solutions and reagents giving off fumes must not be included in a
Behrens box but should be assembled on open shelves. Plastic containers
are suggested for fluorides, alkaline solutions, and reagents responding to
sodium, potassium, calcium, and silica. Closure with caps instead of
stoppers would facilitate maintaining the purity of the reagents. Bottles
or vials of 10- to 15-ml capacity will suffice for most reagents and may
be systematically assembled in plastic blocks or trays designed for easy
cleaning.
The Weighing of Samples may be done on an analytical balance or
on a Sauter balance, which may also serve for the calibration of capillary
pipets.
 Work on the Milligram Scale 105

For Measuring Liquid Samples, a centrifugal pipet (Fig. 39) is rec-

ommended. Measuring pipets of 0.2-ml total delivery and graduated to
read 0.01 ml together with capillary pipets (Expt. 22) suffice for the
measuring of reagent solutions. In addition, calibrated loops and stirring
hooks (Expt. 23) are used for dispensing small volumes of solutions.
One may prefer the use of commercially available microliter pipets
with a syringe control obtainable from various supply houses. Also the
simple shop-made pipet control of BACKUS (461), based upon a thumb
wheel rolling over a rubber tube, should be applicable.
Microcones of about 0.7-ml capacity may be given a shape and size
for convenient handling and efficient work with solution volumes from
10,u1 to 0.5 ml and precipitates from 10-,ug to 0.5-mg weight. The shape
shown in Figs. 22a and b is recommended for general work. Microcones
of the type shown in Fig. 22c are useful for the collection and investigation
of very small amounts of precipitates as well as for the estimation of the
volume of small to medium quantities of precipitates; in addition, they
are useful for the performance of liquid-liquid extractions with small
volumes of solution and solvent. The capillary forming the lower part of
this type of cone should have uniform bore and a length of 3 to 5 cm.
The diameter of the bore may be varied to fit the task, and usually will be
within the limits from 1 to 3 mm. They are made by fusing a thick-walled
capillary of uniform bore to tubing of 6-mm i. d. As an alternative, one
may start with a short length of tubing of 6-mm bore and heat it at the
center until the bore has shrunk to 3 mm before slowly drawing out to a
thick-walled capillary of about 6-cm length. Cutting the capillary at the
center and the wide tubing at points 2 cm from the tapering portions gives
two microcones.
Reagents are Added by depositing them on the inside surface of the
microcone just below its opening. Brief swirling in the centrifuge then
transfers them to the tip of the cone. Solids may be handled with a glass
thread (Expt. 61) or a microspatula obtained by hammering flat the very
end of a platinum, nickel, copper, or aluminum wire of 0.5- to I-mm diam-
eter. Liquids are added from measuring pipets or calibrated capillary
pipets. Very small volumes of liquid reagents may be added with a platinum
loop, a platinum hook, or a stirring rod. These may be inserted directly
into the mixture and twirled for mixing; when withdrawing, these tools
should be repeatedly touched to the inside wall of the microcone to strip
them of the adhering solution or slurry. To test the pH of the mixture,
the solution adhering to the tool may be transferred to test paper by
drawing the tool over a point of the edge of the paper and inspecting this
location with a magnifying glass. A platinum hook removes so little solution
that it is often permissible to transfer tlJ.e whole of the removed matter to
the test paper by simply catching its edge with the hook.
106 Technique of Experimentation

Mixing in the Microcone requires the use of stirrers which may be

obtained by heating one end of a capillary of about 9-cm length in the edge
of a roaring Bunsen flame; first rotate the capillary until a bead of glass
has formed, and then hold it quiet and horizontal with the end touching
the flame until the bead has dropped into the position shown in Fig. 22d.
If the volume of liquid is small or the cone is very narrow, more delicate
stirrers are obtained as follows. Fuse the center of a short piece of capillary
by heating in a microflame or in the edge of a Bunsen flame while slowly
rotating it around its axis, Fig. 22/. Then remove the capillary from the
flame and draw out the bead to a thread a few tenths of a millimeter in

6mm

T'
6rnm
...

~'"1~
~
~ lOmm
I

~
.. I
b c
d
e
f'

g
Fig. 22. Microcones, Blocks, and Stirrers. The cone fJ has been drawn on a larger
scale to show the details of wiring

diameter. Break the thread at a point about 3 cm from the capillary

handle and fuse the end of the thread into a tiny bead which is finally
allowed to droop to one side, Fig. 22e. If the stirrer is intended for a
straight-tip cone, Fig. 22c, hold in perpendicular position when fusing the
bead; only very small beads are permissible in this instance and their
centers must not deviate much from the axes of the rods.
For mixing, the bead is inserted into the mixture and the capillary handle
is twirled between the fingers. Simultaneous motion up and down is
needed when the height of the liquid exceeds a few millimeters. On with-
drawing, the stirrer is stripped by touching it to the inside wall of the
microcone or drawing it through a drop of solvent placed upon the inside
wall just below the opening. As usual, all material is finally combined by
brief swirling in the centrifuge.
Treating with Gaseous Reagents. A small volume of solution may be
rapidly saturated with a gas by bubbling the latter through the liquid.
To obtain suitably small bubbles of gas, proceed as follows. Draw out one
end of soft-glass tubing of 6-mm bore to a capillary of I-mm bore and
10- to I5-cm length. Insert a plug of cotton into the wide part of the tube
 Work on the Milligram Scale 107

to serve as filter for the gas. Then draw out the end of the wide capillary
to a long, fine capillary of about O.l-mm bore, which is broken off to give
it a length of 10 cm. By means of a cork, clamp the tube in a perpendicular
position, Fig. 23, and connect the wide tube to the source of the gas which
should be supplied with a pressure of at least 30 cm of
water column above the atmospheric. Start the flow
of gas before inserting the orifice of the fine capillary
into the liquid. A stream of very fine gas bubbles is
obtained, and there is no danger that solution could be
~I\
I ~
'.:
. 1
'." .1"i
thrown out of the microcone. Finally withdraw the
liquid from the capillary before shutting off the gas so
that no liquid can be lost by beeing drawn into the
capillary.
The technique is recommended for precipitation with
hydrogen sulfide. The gas bubbles rising from the very
tip of the microcone keep the solution thoroughly mixed.
For testing whether or not the precipitation is complete,
one may centrifuge the precipitate into the tip of the
cone and treat only the supernatant solution with the
gas by keeping the orifice of the capillary above the plug
of precipitate.
Before shutting off the stream of gas, remove the
microcone with the reaction mixture and wipe the end
of the capillary, which has been inside the microcone,
first with moist and then with dry filter paper. Finally
break off and discard the last centimeter of the fine
capillary before shutting off the gas stream.
Solutions may be neutralized by bubbling through
them air which has been passed through a gas washing
bottle containing ammonia or hydrochloric acid of
suitable concentration. Fig. 23. Passing Gas
Small volumes of liquid will soon be saturated with into a Solution. The
a gas by absorption through the relatively large surface. bore of the fine
capillary is ex-
The apparatus (866) shown in Fig. 24 has been used for aggerated
saturating the contents of microcones with highly corro-
sive gas, hydrogen chloride, which was generated in an apparatus similar to
that of SWEENEY (699). All connections were made with ground glass
joints. The inlet tube which extends almost to the bottom of the saturation
chamber has four glass rings fused on, that support the microcones. The
stopcocks permit to fill the chamber with gas and to remove the filled
chamber from the gas line so that the action may proceed undisturbed for
any desired length of time. Immersion into heating or cooling baths is
possible with the saturation chamber either connected to the gas supply
108 Technique of Experimentation

or detached from it. The apparatus was used for the precipitation of
AlCl 3 • 6 H 20 from ether solution.
Heating in Microcones. The advantages of electrically heated metal
blocks (155, 904) cannot be denied, but simple improvisations give satisfactory
service.
For temperatures below 100 ° C, a suitable bath liquid in a 150-ml
beaker will suffice, Fig. 25. The microcone may be attached to the rim
of the beaker with a strong wire of aluminum,
nickel, copper, or platinum. An inexpensive short
thermometer may pe used for stirring the bath
liquid.
The steam bath of Fig. 26 consists of a 250-ml
Pyrex Erlenmeyer flask with narrow mouth and
snugly fitting inset which may be easily made of
Pyrex glass tubing of 24-mm outer diameter. The
cross section of the inset at the level a is shown in
the side figure. The glass horns pointing toward
the center of the inset support the microcones and
other apparatus to be heated. If not too much
water is placed into the flask, steam may be
obtained in a short time with a Bunsen burner
or an electric hot plate supplying the heat. A
few granules of zinc added to the water will give
even boiling, but this device should be avoided
when searching for small amounts of zinc.
At times, solutions are heated for the expelling
of dissolved gases which are rapidly carried away
Fig. 24. atll!·a ion
if steam bubbles form in and escape from the
hombc,·: fib lit '/2 nat. heated solution. Since boiling will not ensue when
. iz heating in a steam bath, the expelling of gases
should be aided by passing a stream of gas bubbles
through the heated solution. This may be done by forcing a suitable gas
into the solution as suggested by Fig. 23, but frequently it will suffice
to use the simple device shown in Fig. 27. A stream of fine bubbles emerges
from the opening of the fine capillary as the air expands in the bulb. When
the bubbling stops, one removes the gadget; if desired, one may cool the
bulb in a stream of tap water and again insert the gadget into the cone.
Heating blocks of metal offer the simplest means for the maintenance
of controlled temperatures above 100° C. In addition to a well for the
thermometer, they are provided with holes for the insertion of apparatus.
They are best made of aluminum which is not only a good conductor of
heat but also has a remarkable resistance to the action of corrosive agents.
Copper blocks become a nuissance because of the oxide scales which, on
 Work on the Milligram Scale 109

heating the blocks, are frequently ejected by the oxidized surface with such
force that they may land in the reaction vessels and contaminate the

1
ig. 25. Wlttel' Bltth Fig. 26. team Bath
Fig . 27. Exp \ling Gas

material under investigation. Electrically heated blocks are more easily kept
clean, but heating with a gas flame is perfectly satisfactory for most purposes.

Fig. 28. Heltting Block. a Inset for microcones; b inset for centrifuge tubes; c inset
for distilling flask of Fig. 33; approx. 1/2 nat.. size

The heating block suggested by Fig. 28 permits accomodating various

apparatus by exchange of insets A, B, C, etc. The shape of the insets
shown in the figure brings about that the upper portions of the apparatus
110 Technique of Experimentation

will be more rapidly heated than the bottom parts, and this will aid in
the prevention of creeping. The fork E, made of aluminum wire, is used
in the removal of microcones from the hot block.
For heating under pressure, micro cones may be sealed with the use of
RACHELE'S pressure cap (413) shown in Fig. 29. The cone rests in a brass
ring with rubber lining, to which a metal stirrup is attached. The rubber
lined pressure plate is tightened on the rim of the cone by means of a screw
threaded through the center of the stirrup. SCHENCK (1272) cuts two 5-mm
thick slices with a very sharp knife from the thick end of a clean, new
rubber stopper of about 24-mm diameter. A hole, just large enough to

Rubbe<

Fig. 29. Pressure Cone (left) and RACHELE'S Pressure Cap Applied to Microcone (right);
approx. nat. size

snugly fit the microcone, is cut through the center of the disk with the
two cut faces, and this disk is shoved up to the rim of the cone. The smooth
face of the other disk is placed upon the opening. The top disk is backed
up by a stiff metal disk of like diameter (a 25-cent piece), and the three
disks are clamped together with two Hoffman screw clamps placed along
parallel cbords tangentially to the cone.
For further alternatives, the cone may be closed with a rubber stopper
which is then secured by means of a ligature with copper or aluminum
wire (13). If use of rubber or cork is not permissible, the reaction mixture
may be sealed into a pressure cone, Fig. 29, which is given approximately
the dimensions of a microcone. The neck should have a bore of not less
than 2 mm to facilitate the introducing of capillary pipets or siphons.
Previous to heating, the cone is sealed at a by the customary procedure of
heating in a small £lame until soft and then drawing out without removing
from the £lame so that the capillary which forms is fused through and shut.
After heating, the pressure cone is opened with the customary precautions
 Work on the Milligram Scale 111

(eye shield) taken when opening Carius tubes. If the absence of pressure
after cooling is assured, it may be simply cut open below the seal. Finally,
a standard microcone may be sealed, for heating, into a somewhat larger
tube, or one may use a device like the micro autoclave of GORBACH (155),
that uses the principle of the safety valve and permits adjusting the pressure
by means of a weight displaceable along a lever.
Fusions may be carried out on platinum foil or in a small spoon, cup,
or crucible of 0.05- to 1-ml capacity and made of porcelain, vitreous silica,
platinum, tantalum, silver, nickel, or iron. Pyrosulfate fusions may be
performed in a Pyrex micro cone, and it is recommended to do it at the
point where the taper changes to the cylindrical shape with the cone in
approximately horizontal position. In the following, detailed directions
are given for the performance of fusions on wires, a technique which is
specially suited for the scale of work.
Straight pieces of wire of 0.5-mm diameter and 4- to 5-cm length are
used. One end is either clamped into a locking forceps or sealed into the
end of a capillary of 5-cm length and somewhat more than 0.5-mm bore.
Assuming a density of 2.5 for the fluxes and that the volume of the wire
is compensated by the elongation of the beads, one may crudely estimate
that beads with 2-, 3-, and 4-mm diameter at the equator have weights
of 1.5, 6, and 13 mg, respectively. These estimates will aid in getting flux
in the correct proportion to sample to be treated.
The bead is formed at the end of the wire and held there by not heating
the bead directly, but by heating the wire at a suitable distance from the
bead which is at the end of the wire.
Sodium Oarbonate Fusion. Place some anhydrous sodium carbonate
upon a piece of platinum foil, and take up small portions of it by touching
them with the end of the heated platinum wire and later with the molten
salt on it until a bead of the desired size is obtained. Estimate the amount
of flux from the equatorial diameter of the bead, or determine it by weighing
the wire without and with the bead or by supplying a weighed amount
of salt on the foil. Use a microflame for heating, Expt. 19, and if the sulfur
content of the illuminating gas is objectionable, use a flame of purified
hydrogen. To transfer the material under investigation to the bead, either
make a slurry, transfer small portions of it from a capillary to the bead,
and fuse after each addition-or weigh the substance on a sheet of platinum
foil having a high polish and mop the substance up with the molten bead.
Finally, fuse until gas bubbles no longer appear in the bead; this takes
about one minute. For dissolving the melt, insert the end of the wire
with the bead into the required volume of the selected solvent which may
be contained in a microcone or a crucible.
Sodium Peroxide Fusion. Fuse a pellet of sodium hydroxide on a
sheet of nickel. Collect a bead by dipping the end of a nickel wire repeatedly
112 Technique of Experimentation

into the melt and each time allowing the hydroxide to solidify on the wire.
Finally add the material under investigation as a slurry in water or mop
it up from a platinum or nickel sheet with the cold bead which, to this
end, is exposed briefly to air so that its surface becomes moist. After fusing
briefly, take up sodium peroxide with the bead with the use of the latter
procedure and fuse again. Repeat the adding of peroxide and fusion once
or twice.
Pyro8ulfate Fusion. On a slide or watch glass, prepare a thick paste
of powdered potassium sulfate and concentrated sulfuric acid. Take the
paste up with the end of a platinum wire and fuse until a bead of proper
size is obtained. Heat the bead cautiously until fumes of sulfur trioxide
start to appear. Then add the material under investigation as a slurry
in concentrated sulfuric acid. Perform the fusion by lightly heating the
wire behind the bead in the edge of the lowest part of the Bunsen flame.
The bead should be 10 mm to 20 mm outside the flame. If the pyrosulfate
decomposes too rapidly, there may not be time enough for getting the
material completely dissolved. Thus, if undecomposed particles remain
visible in the melt, the fusion may: be repeated after adding to the bead
some more concentrated sulfuric acid.
Treatment with Hydrofiuoric Acid is sometimes used for the dissolution
of silicates. The following technique considers suggestions made by
VAN BRUNT (416,1271).
Out from platinum foil either a disk of 3-mm diameter or a rectangle
of 2 mm X 6 mm and weld it to a platinum wire of 0.3-mm diameter and
4-cm length, which serves for handle. Heat the foil and the wire in the
intended relative positions in a small hydrogen flame, burning at the
orifice of a capillary, so that they stick together. To perform the welding
proper, heat a small steel anvil to just above 100° 0 and mount the capillary
so that the hydrogen flame burns horizontal and close above the surface
of the anvil. With forceps hold the part to be welded into the flame so
that it touches the anvil and, when it sends out a red glow, tap it lightly
with a 4-oz. ball pein hammer. Finally clean by fusing potassium pyrosulfate
upon the foil and wire and then dissolving the melt in concentrated hydro-
chloric acid.
Make the substance to be treated into a slurry with dilute sulfuric acid
or water. Take the slurry into a capillary and transfer it to the foil at the
end of the platinum wire by adding small portions at a time and evaporating
after each addition. To this end, apply a microflame (burning from the
orifice of a capillary) to the wire at a suitable distance from the foil. When
the transfer has become complete, treat the residue upon the foil with
hydrofluoric acid or with hydrofluoric-sulfuric acid mixture, a drop of
which is held ready upon a platinum sheet. Add the acid by means of a
platinum loop to a part of the foil, that holds little or none of the substance.
 Work on the Milligram Scale 113

Then slowly evaporate the acid by heating the wire at a suitable distance
from the foil. Repeat the treatment as often as it appears necessary.
For the extraction of the residue immerse the foil in the solvent held ready
in a microcone.
The technique is also suited for the removal of fluoride by evaporating
with silica and sulfuric acid, and it may be used for the removal of organic
matter by ashing.
Separation of the Liquid from the Solid Phase. As a rule, it is necessary
to pack the solid by swirling in the centrifuge. Small amounts of liquid
are then lifted off by means of a capillary pipet, whereas large volumes
are removed with a siphon. Solid particles clinging to the walls of the

___ I

:J

Gf{ (
c
Fig. 30. Use of Capillary Pipet; a and b show the correct procedure. Approx. nat. size,
bnt the bore of the capillary is exaggerated

microcone are stirred into the liquid and combined with the bulk of the
solid by centrifuging.
To remove a small volume of supernatant liquid, select a capillary of
0.5- to I-mm bore and 20-cm length and draw out one end to a tip of
10- to 20-mm length and not less than 0.2-mm bore. Hold the microcone
almost horizontal and lay the pipet into the cone as shown in Fig. 30a.
Regulate the rate at which the liquid enters the pipet by inclining the cone.
Hold the cone in the left hand, and stepwise push the pipet into the cone
with the index finger of the right hand so that the opening of the pipet
always remains just below the meniscus of the liquid in the cone. Finally
bring the tip of the pipet to a point 1 mm from the precipitate, Fig. 30b.
If the cone is properly inclined, the liquid is completely taken up by the
pipet. Meniscus b disappears first, and very soon afterwards meniscus a.
Withdraw the pipet from the microflone and transfer its contents to another
cone (or to wherever it is needed) by blowing them out with the mouth.
Save the pipet for the transfer of washings, which operation will also rinse
the pipet. If there is any doubt concerning the completeness of the removal
of the liquid, centrifuge the microcone containing the solid and transfer
Benedetti-Pichler. Identification 8
114 Technique of Experimentation

any liquid collecting above the precipitate to the bulk of the centrifugate
as described.
If the capillary pipet is not perfectly straight, or if a straight pipet is laid
into the microcone in such a way that a narrow space results between pipet
and cone, Fig. 30c, liquid is drawn into this space by capillary attraction and
cannot be taken up by the pipet. When the latter is withdrawn, this liquid
spreads over the walls of the microcone.
To remove a large volume of liquid, convert the capillary pipet into a
siphon. Hold one end of the pipet so that the other end will give the desired
angle of about 60 degrees by slowly following the pull of gravity when
the middle is cautioUf~ly heated in the edge of the non-luminous Bunsen flame

Fig. 31. Use of Capillary Siphon. Approx. 1/2 n at. size, but the bore of the siphon
is exaggerated

1 cm above the barrel of the burner. Break the fine tip close to the taper
to obtain a reasonably fast flow of liquid.
The use of the capillary siphon is illustrated by Fig. 3l. Place the leg
with the wide opening into the microcone that is to receive the liquid.
Then insert the leg with the tip into the cone containing the liquid above
the solid so that the tip and the wide part of the capillary rest on opposite
sides of the microcone just as in Fig. 30a indicated for the straight capillary
pipet. Incline tl}e tubes properly, and the capillary siphon will fill by
itself. Proceed in princip~e just as described for the capillary pipet. Finally
bring the tip of the siphon close to the precipitate so that practically all
liquid may be removed in one operation. Withdraw the siphon so from
the cone with the solid that the leg with the wide opening stays in the cone
with the liquid and remains there when the cone is set aside. Proper
inclining and the narrow orifice at the tip make it possible to keep the
siphon continuously filled with liquid.
Whirl the cone with the solid in the centrifuge, and use the siphon
for the transfer of the liquid collecting above the solid. Set siphon and
centrifugate aside as before and hold them ready for the collection of
 Work on the Milligram Scale 115

washings. The first washing may serve to rinse the siphon into the
centrifugate.
Washing and Extraction of Solids. Deposit the measured volume of
solvent below the opening of the microcone. Without stirring up the solid
in the point of the cone, spread the liquid by means of the stirrer over
the inside wall of the microcone. Collect the liquid in the point of the cone
by brief swirling and then mix solid and liquid thoroughly by stirring so
that none of the solid is spread over the walls of the cone. At this time,
heat in a suitable bath if the washing or extraction shall be performed
with hot solvent. Finally, centrifuge and withdraw the liquid as described
above. The microcone with the liquid will, as a rule, remain sufficiently
warm to satisfy the requirement of extraction or washing with hot solvent
if the work is performed without delay.
Transfer of Solids. Whenever possible, dissolve the solid and transfer
the resulting solution by means of capillary pipet or siphon. If this is not
feasible, add some water or some other suitable liquid and stir to obtain
a slurry in the tip of the microcone while the main portion of the supernatant
liquid remains clear. Obtain a capillary of 0.5- to I-mm bore and 20-cm
length and cut it off evenly at both ends. Close one end with the finger
and insert the other into the microcone so that it touches the bottom of
the tip. Incline cone and capillary 45 degrees to the horizontal and then
remove the finger from the upper opening of the capillary. The liquid
rushes into the capillary with the slurry entering first. Again close the top
opening of the capillary with the finger, and transfer the contents to the
location where they are wanted.
Depending upon density and particle size of the solid, it will collect
more or less rapidly in the liquid at the lower end of the capillary if the
latter is held in a vertical position before being emptied. It may become
possible to transfer the whole slurry to the receiver while retaining most
of the transfer liquid in the capillary. Of course, the transfer is repeated
until it is sufficiently complete. Finally, the transfer liquid may have to
be removed, which is accomplished with the use of the technique for separating
liquid and solid phases and (or) evaporation.
Electrodeposition. The needle electrode, p. 186, may be used for side
tests and for the detection of very small amounts of metals. In general,
much simpler apparatus will suffice. A platinum wire of O.I-mm diameter
is fused into the tip of a microcone to serve as anode (1143). A like platinum
wire of 6-cm length and soldered to fine, insulated copper wire is inserted
through the opening of the cone, Fig. 22g. Only a few millimeters of the
end of the cathode are inserted into the electrolyte. A glass fiber of 8-cm
length is placed into the cone so that it touches the wire in the tip; this
makes that the individual gas bubbles rise as they form on the anode and
prevents the accumulation of gas in the tip. When the deposition is considered
S*
116 Technique of Experimentation

complete, the cathode is quickly withdrawn from the microcone and inserted
into the rinse solution, which may be held ready in a beaker, without
breaking the connection to the source of current. To remove the deposit
from the cathode for further treatment or the performance of a confirmatory
test, the end of the wire is dipped into the droplet of solvent held ready
in a microcone or upon a slide. Electrolysis with an e. m. f. of 10voits
and an electrolyte of 3-F KOH gave efficient separation of 10 flg zinc from
solutions containing up to 10 mg manganese, 0.5 mg cobalt, 0.5 mg nickel,
0.1 mg iron, 0.1 mg aluminum, 0.1 mg chromium, and 0.1 mg cadmium.

l
J c
b

Fig. 32. Siphons Operated by Suction; a using a thermometer capillary; b tube for
control of suction; c capillary siphon; d stand; about 1/2 nat. size

Liquid-Liquid Extraction is conveniently performed in microcones.

A straight-tip cone, Fig. 22c, is recommended when the volume of the
bottom layer or of both layers is small. Complete separation of the layers
may be obtained by use of centrifugal force; in addition, the surface of
the microcone may be treated with chlorosilane (13, 90, 423) so that the
liquids will not wet it, and the same treatment may be applied to the
outside (or outside and inside) of the pipet or siphon used for mechanical
separation of the layers.
Suction operated siphons or pipets, Fig. 32, permit removal of either
top or bottom layer. The suction may be supplied by mouth or with an
aspirator. In the latter instance, a tube b with side opening which is
closed with the finger should be inserted into the suction line to permit
instantaneous stopping of the action. Fig. 32c shows the use of a small
square of thin rubber sheet for securing a tight fit of the capillary siphon
 Work on the Milligram Scale 117

in the wide tubing. As a satisfactory alternative, one may secure the siphon
in the tube with sealing wax or some other suitable cement.
If a mechanica;l stand is available, the micro cone is clamped to it in
vertical position so that it may be raised and lowered with the rack-and-
pinion motion. As a substitute, the cone may be attached with rubber
bands to the tube of a microscope.
No difficulty should be experienced in lifting off the top layer so that
only a small fraction of it is left behind. Various devices may be used to
get the tip of the siphon or pipet into the bottom layer without having
some of the lighter liquid entering it. The tip may be finely drawn out
and sealed so that it may be broken open by pressing it against the bottom
of the cone. The tip may be inserted in a drop of the heavier liquid so
that it fills with it by capillary attraction before it is inserted into the cone,
and (or) air pressure is maintained in the siphon or pipet to prevent the
lighter liquid from entering; the last procedure offers little difficulty when
a pipet is used which is operated by a syringe control obtainable from
supply houses, a Pumpett (A. S. La Pine and Co., Chicago 29, Ill.), or a
pressure or levelling-bulb device used on the microgram scale (90, 141).
One will adjust the pressure so that an air bubble or a droplet of the heavier
liquid just starts to form at the orifice of the pipet tip when it enters the
heavier layer. If the outside of the pipet or siphon is not wetted by the
lighter liquid, there is no objection against immediately raising the micro-
cone until the tip of the pipet or siphon touches its bottom. Otherwise,
it seems preferable to raise the cone while the lower layer is withdrawn
so that only a short length· of the tip of the pipet or siphon is immersed
into the heavier liquid at any time.
If the volume of the layer to be removed is less than 10,ul, one will
prefer to use a capillary pipet. A tip of 0.2-mm bore and up to 5-cm length
will give good control and will be found advantageous when working in
straight-tip cones.
Evaporation may be carried out efficiently in the microcone by blowing
a stream of filtered air from a capillary on the liquid, Fig. 28d. The heating
device must be selected to fit the vapor tension of the liquid to be removed;
volatile solvents may be removed by simply blowing air into the cone,
but creeping will be prevented if the upper part of the cone is heated in
some manner (blowing steam upon the outside, heating coil, surrounding
by a hot metal shell, Fig. 28a, b, etc.). Insertion of the micro cone into a
steam bath suffices for the evaporation of aqueous solutions, but heating
in a metal block is required for the elimination of sulfuric acid.
Use of an infrared heat lamp is recommended for evaporation in crucibles,
small dishes, and watch glasses. Application of the heat from above
eliminates creeping, and a stream of air blowing over the surface of the
liquid gives rapid evaporation without boiling.
118 Technique of Experimentation

Description of elaborate heating devices (155, 162, 904) and apparatus

for evaporation by exposure to a stream of hot air (162, 887, 891) may
be found in the literatur~.
Distillation. Apparatus and technique have to be selected to fit the
individual case.
The apparatus of Fig. 33 is useful when a substance shall be more or
less completely removed from a relatively large volume of liquid (more
than 10,ttl) and transferred to some reagent solution as this happens in
the separation of arsenic from the other metals by distillation of the tri-
chloride. If the charge is to be concentrated to a certain volume, this
volume may be indicated by a graduation mark upon the tube.

Fig. 33. Distillation Apparatus Used by ANNE G. LOSCALZO (88); about 1/2 n at. size

To perform a separation, introduce the charge with a capillary, capillary

pipet, or capillary siphon through the short tube a of 2-mm bore. Then
place the tube into a heating block with windows for observing the progress
of the distillation and connect a with a supply of inert gas (air dried by
passing through a tube with Anhydrone is suitable for the distillation of
Asels)' Insert the capillary b into a microcone with suitable absorbent
and supply gas at a rate to obtain not more than two small bubbles per
second in the microcone. For cooling, the latter may be placed into a
dish with ice water, or ice water may be run over the outside of the cone
to be caught in a funnel. Then heat the block to the needed temperature
and hold it until the volume of the charge has been reduced to the required
amount. For stopping or interrupting the distillation, remove the tube
from the heating block and withdraw the microcone b before shutting off
the stream of gas. At the conclusion of the distillation, rinse the capillary b
 Work on the Milligram Scale 119

before stopping the gas stream by drawing it through a drop of suitable

reagent suspended in a platinum loop.
Analytical distillations with a few microliters of liquid may be carried
out quite efficiently by closely approaching the surface of the liquid which
has been heated nearly to the boiling point with
an efficiently cooled condenser. The procedure
which uses the principle of molecular distillation
has the advantages that bumping cannot occur
and that spray resulting from a boiling charge
will not contaminate the condensate.
The apparatus (869) may be assembled in the
laboratory. Fig. 34 shows the Pyrex distilling tube
of 4-mm bOre and 20- to 30-mm length in the
heating block. The aluminum block of 2 cm X
X 3 cm X 5 cm had the four sides covered with thin
sheet of mica to serve as electrical insulator. Over
this mica was wound a 60-cm length of No. 27
nichrome wire (0.35-mm diameter) to serve as
heating coil which was covered with asbestos board
held together by two bands of copper wire. The
insulating asbestos mantle extended above and
below the metal block, and also the top surface
of the block was covered by asbestos board. The
heating coil was connected through a 64-ohms
Biddle resistance to the 120-volt d. c. line. The
cold-finger condenserb, attached with De Khotinsky
cement to a rack-and-pinion movement, consisted
of a thin-walled capillary of 6-cm length and tapering
from 1.5-mm outer diameter at the lower end to
3-mm o. d. at the upper end. The inlet tubing for
the ice water used in cooling opened close to the
very tip of the condenser and was a capillary of Fig. 34. Analytical
less than 0.5-mm o. d. The outlet capillary of Distillation with a Few
Microliters of Liquid;
like diameter opened about 10 mm away from J. R. RACHELE (410);
the tip of the condenser so that only its lower about 2/3 nat. size
end was cooled. The outlet was connected to
a water pump which sucked the ice water through the condenser.
For efficient use provide the distilling tube with graduation marks
indicating the critical volumes to be considered and indicate the correspond-
ing positions of the condenser.
Heat the block to the proper temperature. Place the charge in the
distilling tube, and centrifuge it for collection at the bottom of the tube.
Start the circulation of ice water through the condenser. Place the distilling
120 Technique of Experimentation

tube into the block and immediately lower the condenser into it to bring
its tip close to the surface of the liquid. Hold the temperature constant,
and at suitable intervals (2 to 10 minutes) raise the condenser and collect
accumulated distillate by touching a capillary pipet to the tip of the
condenser.
Emich's method of fractional distillation was modified by MORTON and
MAHONEY (429) to give about two to three fractions for each microliter
of liquid available. This is accomplished by using quite small drops for
the determination of boiling points and by reducing evaporation losses
during fractionation by attention to various detail such as cooling the
condensate, use of a shielded block for heating, and proper timing.
The fractionating is performed with the use of a copper block, 38 mm X
X 38 mm X 15-cm high, which is heated with either a gas flame or a coil
of 12 meter of No. 30 (0.25-mm diameter) oxidized resistance wire wrapped
around the lower 4 cm of the block. A sheet of asbestos, 1 mm thick,
covers the top surface of the block, which contains two holes bored to a
depth of 95 mm. The one with 6-mm diameter receives a thermometer;
the other of 8-mm diameter is for the fractionating tube.
The fractionating tube is made of thin-walled tubing of 2-mm bore
and 15-cm length. The lower end is sealed and blown up to form a bulb,
6- to 7-mm outer diameter. Glass wool is ground in a mortar until it is
fine enough to pass into the tube, and the fractionating tube is packed
with it so that the packing will not quite reach up to the top surface of
the metal block. Immediately above the packing, the tube is made to
form a short (3 to 4 mm) constriction reducing the bore to one half or
one third. The constriction should appear just above the level of the
asbestos sheet covering the top of the block, and the tube is cut off 20 mm
above the constriction.
The fractionating column is shielded to reduce the rate of the transfer
of heat from the metal block by slipping over it a loosely fitting tubing
of glass or asbestos paper of 7-mm outer diameter. The tubing should
rest upon the bulb and should reach up to nearly the surface of the copper
block. Finally, the fractionating tubing is pushed through the center of
a circular disk of asbestos paper, 10-mm in diameter and the underside
lined with aluminum foil, which provides a heat shield that covers the
opening of the well during distillation.
To perform a distillation, first prepare a suitable number of boiling
point capillaries (Expt. 21) with a bore from 0.5 to 0.2 mm in the wide
part of the tubing and very fine tips of 10- to 20-mm length. The longer
and finer the tips, the less liquid is lost when fusing shut their orifices.
Before applying the technique to an unknown, practice it first with 10,u1
of benzene and then with a like volume of a mixture of equal volumes
of benzene and xylene.
 Work on the Milligram Scale 121

To start with, weigh the fractionating tube on an analytical balance.

Introduce the liquid to be distilled, centrifuge it to the bottom of the
tubing, and again weigh the latter to obtain the weight of the sample.
Slip the insulating jacket (glass or asbestos tubing) over the packed column
so that it rests on the bulb and top it off with the circular heat shield.
Then wind a strip of wet filter paper, 15 mm wide and about 4 cm long,
around the end of the capillary to serve as condenser; the ring of wet paper
must not cover the first three millimeters above the constriction and the
heat shield, where the condensate will collect. Insert the fractionating
tube into the block, push the heat shield down to cover the well, turn the
heat on, and direct a fine stream of air from a capillary upon the condenser,
i. e., roll of wet filter paper.
If the approximate boiling point is known, heat first rapidly and then
slowly until a tiny droplet forms above the constriction. Quickly turn off
the heat and read the thermometer. The task is now to find the lowest
temperature that will give, within 60 to 90 seconds of heating the fractionating
tube in the well of the block, a fraction of not more than the size required
for the determination of boiling point. To this end, withdraw the fractionating
tube from the well, remove heat shield and insulating jacket, and whirl
the tube in the centrifuge to return all liquid to its bulb. Wait until the
temperature of the block has dropped 4 to 5 degrees, and then return the
fully armed fractionating tube into the well. First hold the temperature
for 90 seconds; if no condensate forms, slowly raise the temperature.
Repeat the cycle of heating until a droplet forms, centrifuging, and again
heating until the proper temperature has been found. Then collect the
first fraction by touching the droplet of condensate with the tip of the
capillary pipet intended for the boiling point determination. Wait until
the wide part of the capillary is filled to a length of 1 mm; then withdraw
the pipet, seal its tip as described in Expt. 20, and remove the fractionating
tube for cooling and centrifuging.
The time may be used for determining the boiling point of the fraction.
The size of the fraction may be determined by weighing the pipet before
and after collecting the fraction, or it may be estimated from the bore
of the pipet and the height of the liquid column.
Regulate the heating of the block so that its temperature will not
drop more than two degrees. For the collection of the second fraction
and following fractions just return the fractionating tube armed with
the cold insulating jacket and heat shield to the well and repeat the procedure
used in collecting the first fraction.
Up to 70 fractions may be obtained with relatively large volumes
of liquid, 25,u1, and boiling point graphs may be plotted with little effort
if fractions of like volume are collected, which is simple when using pipets
of the same bore. If the liquids have high vapor tension, evaporation
122 Technique of Experimentation

losses may be reduced by substituting for the paper condenser a coil of

No. 13 copper wire (1.8-mm diameter) forming one half of the wire and
the other half shaped to give a wire basket into which are placed a few
lumps of dry ice or solid carbon dioxide.
Overheating the insulating jacket must be avoided by holding the time
for collecting a fraction down to the 60 to 90 seconds specified. If the
heating is prolonged much beyond these limits, the jacket has time to
acquire the temperature of the block, about 10 degrees above the boiling
point of the fraction to be collected, and the separation of the components
may completely fail.
The work cannot be hurried, and the collection of a large number of
fractions does require three to four hours. Obviously, when the boiling
point rise to a higher boiling fraction occurs, it is advisable to again determine

b c d
,-
o Scm

1em
o
'===---h
c<
e
F ig. 35. Subl ima ion on tho Milligram call"

the proper minimum temperature of the block before collecting the next
fraction. MORTON and MAHONEY (430) describe a copper block for the
determination of the boiling points.
Sublimation. The techniques described in connection with work upon
the centigram scale for obtaining crystalline sublimates are suited also
for the milligram and submilligram scales. A modification of procedure
is suggested for sublimation in a stream of gas.
Sublime in thin-walled tubing of 3- to 4-mm outside diameter and
15-cm length, which is ,:lrawn out to a capillary of 0.1- to 0.2-mm bore at
one end, Fig. 35. For introducing the charge, use a technique suitable
to its nature. If it is a small object, place it into the tube and push it into
proper position by means of a rod. If it is a dry powder, place it into a
trough made of platinum foil, 5 mm X 10 mm, which is then pushed into
the tube, or introduce it scooped up into a thin-walled capillary of 1- to
2-mm bore and 2-cm length. Place a solution or slurry into the trough of
platinum foil and evaporate to dryness before introducing the latter into
the tube. As an alternative, take up a slurry into a wide capillary and
collect it at the point of the capillary with the procedure described in
 Work on the Milligram Scale 123

Expt.47. Finally dry the capillary g, Fig. 35, before pushing it into the
tube for sublimation.
Slip over the tube a helix of thick copper wire, which is wound loosely
enough to permit observation of the charge during heating, and then
connect with a supply of suitable gas by means of the wide tubing a,
containing a loose plug of cotton, which is drawn out first to the diameter
of the sublimation tube and then to a short capillary of O.I-mm bore.
Turn on the stream of carrier gas, insert the fine capillary e into a drop
of water or suitable absorbent, and adjust the gas stream to get about
two tiny bubbles per second. Push the wire helix over the charge and heat
with a small non-luminous flame. If only a very small amount of sublimate
is obtained, draw out the tube at d to a suitable small bore (0.5 to 1 mm)
and drive the sublimate into the resulting capillary by slowly advancing
the heated zone. For collecting sublimate and residue, cut the tube at c
and (or) other suitable locations to obtain material for investigation near
the end of a section of tubing of 6- to 7-cm length. For the extraction,
place the piece of tubing into a microcone as shown in Fig. 35/ or mount
it in the opening of the cone by means of a cork with fitting hole. Use a
small volume of solvent as indicated by the illustration and get it to the
material by means of a stirrer or platinum loop inserted through the tubing
or by simply heating the solvent to boiling. Finally centrifuge the microcone
with the tubing in place. The liquid collects in the tip of the cone, and
the tubing may be withdrawn without causing loss of any material.
Liberation of Gases. Apparatus and techniques described for the
centigram scale, p. 96, are suited for the milligram scale also. If the amount
of gas liberated is very small, a gas reaction cell as described for the sub-
milligram scale, p. 159, is recommended. As an alternative, the gas may
be liberated in the tip of a microcone; a wad of glass wool may be placed
into the conical portion of the microcone, and the small fragments of test
papers, test fibers, and reagent droplets may be deposited upon the inside
wall of the cylindrical part. A paraffin-coated ring zone in the conical
part would safely prevent diffusion along the glass; a test droplet could
be deposited upon a fragment of a cover slip, that fits into the cylindrical
part of the microcone. Obviously, a stopper should be applied, that does
not absorb or react with the liberated gas. A microcone of thin-walled
tubing and somewhat longer than usual in the cylindrical part could be
sealed by fusing it shut.
Confirmatory Tests. Customarily used are the slide tests and the spot
test technique described for the submilligram range. Obviously, the
sensitivity of the confirmatory tests should be somewhat better than
that of the general working technique to permit frequent testing for
minor and trace constituents in small aliquots of centrifugates or pre-
cbpitates.
124 Technique of Experimentation

The cell shown in Fig. 36 is useful for the microscopical inspection

of small precipitates collected with the centrifuge in the point of the cone.
To this end, the cell is filled with water. Illumination with daylight and
observation of the reflected light reveal the true color of the collected solid.
The microcone is not suited for revealing light colorations occurring
in small volumes of liquid. To this end, the liquid is best taken up with
a capillary, Expt. 55.
Storing Material under Investigation. Without delay dissolve precipitates
which age or oxidize upon standing. In general, it will suffice to stopper
microcones containing liquids, slur-
ries, or liquids containing precipitates
or saturated with a gas. Liquids may
also be taken up into a capillary
which is then fused shut at both
c

c
Fig. 36. Cell for the Observa tion of the Fig. 37. Suction Flask for the R in sing of
Contents of Microcones Microcones and Pipets; approx. 1 / 3 n at . size

ends. In this manner, the volume of gas in contact with a liquid may be
reduced to a minimum.
Cleaning Apparatus. Feathers or pipe cleaners are useful for the
scrubbing of microcones. Rinsing and drying are best performed with the
use of suction. The suction-operated siphon, Fig. 37 a, permits efficient
treatment with cleaning solution and rinsing. For quick drying, the micro-
cone is whirled in the centrifuge, whereafter the liquid collected in the
point is removed with the siphon; the remaining film of moisture is removed
within one or two minutes by drawing air through the cone in position a.
The suction flask s should be connected to the line with a 30- to 50-cm
length of thin-walled rubber tubing of 4-mm outer diameter; this usually
requires the use of tapering connectors c at both ends. Stiff, heavy-walled
suction tubing is generally undesirable, and its use should be restricted
to those occasions which demand a good vacuum. The siphon a may be
exchanged for a stopper b with a short length of 6-mm glass tubing g
and rubber tubing r for the rinsing of thermometer capillaries, measuring
pipets, microburets, etc., or it may be replaced by the gadget c for the
cleaning of calibrated capillary pipets which are pushed through a pinhole
in the center of the small square of thin rubber sheet, sq.
Expt.22 Work on the Milligram Scale 125

Experiment 22
Oalibration of Oapillary Pipets and Oentrifugal Pipets
Analytical balance, millimeter rule, and microscope with calibrated micro-
meter scale.
To obtain a useful capillary pipet, make it from a capillary of 0.3- to
0.5-mm uniform bore and 20-cm length. TEN EYCK SCHENCK (127~)
points out that for drawing such capillaries, the glass tubing must be
rotated in the flame (a Meker flame is preferable to a Bunsen flame) until
the whole heated section is uniformly soft, which may take twice as long
as the time required for the first softening occurring in the hottest part
of the flame. Proper drawing gives sharp tapers on both sides of a uniform
capillary; gently sloping tapers indicate a capillary with the bore increasing
from the center to both ends.
For determining the uniformity of the bore, cut from each end of
the capillary a length of 1 cm. Scratch with the sharp edge of broken

Ir 1 I::;
.0 9
i
8
I ii' 1'1'1'1'1 '}
7 6 5 ~ 3 Z 1 ~
G
Fig. 38. Use of Calibrated Pipet

porcelain or with an ampule cutter to get a clean break vertical to the

axis of the tube. Mount the short lengths on the stage of the microscope
and parallel to its optic axis as outlined in Expt. 55, and focus upon the
squarely cut upper ends with a 10 X objective. Get the image of the opening
into the center of the field, and read the diameter off the eyepiece micro-
meter scale; the capillary may be considered satisfactory if the bore
is the same at both ends within 5 %, i. e., 25 f-lm.
The capacity of a given length of the capillary may now be calculated
from the diameter D. The distance in millimeter which the meniscus is
to travel if 1 f-ll shall be delivered may be estimated from 4/3 D2 if D is
substituted in millimeters; 1 mm = 1000 f-lm. It will be between 3 and 8 mm.
Some allowance for the wetting of the glass is made by substituting 3 for n.
Draw out one end of the capillary to a tip of 0.2-mm bore and cut it
off at a distance of about 5 mm from the taper. The diameter of the orifice
may be measured by attaching the capillary near the wide end to a clamp
which is finally to rest upon the table -top, and then inserting the tip of
the pipet (after removal of mirror and condenser) from below through
the central opening in the stage of the microscope; the narrow top orifi'ce
is brought into focus after moving the pipet into the optic axis of the
microscope and casting light upon it.
To check the calibration, insert the tip of the pipet into a drop of water.
Allow the water to enter until about 10 cm of the capillary are filled.
126 Technique of Experimentation Expt.22

Then measure the length of the water column with a millimeter ruler, Fig. 38.
Make allowance for the water in tip and taper by adding to the length
of the cylindrical column one third of the length of the taper. Without
delay, weigh the filled capillary to the nearest tenth of the milligram.
Then empty the pipet slowly by holding it horizontally and touching the
tip to filter paper, and weigh the empty capillary. The delivery is given
by the difference of weights. Again compute the distance which the
meniscus has to travel in the wide part of the pipet to deliver l,ul. The
holdup upon the walls of the capillary and consequently the amount of
delivery depend much upon the rate of outflow. Thus, one should attempt
to keep the latter constant within limits suggested by the
precision requirements (0.1, relative, when measuring
reagents).
Use the pipets as in the determination of the delivery,
i. e., measure the liquid column with the millimeter ruler.
For cleaning, use the suction device of Fig. 37 c. The
rubber sheet provides a tight seal when suction is
applied. Transfer cleaning solution to the upper opening
by touching it with a drop of the reagent hanging at the
end of a glass rod. For rinsing with water, touch the nozzle
of the wash bottle to the opening. The water is sucked out
Fig. 39. Trans· of it and through the pipet. Finally, lift up the rubber square
fer from Centrif-
ugal Pipet to
and apply some water to the tapered tube t. Again allow the
Microcone rubber square to form the seal, rinse the pipet once more,
and then dry it by sucking air through it for one minute.
Store the calibrated pipet between two plugs of cotton in a test tube.
Indicate upon the label the length that corresponds to a delivery of l,ul.
The centrifugal pipet, Fig. 39, may be calibrated for capacity since use
of centrifugal force permits transferring its whole contents to the point
of the micro cone when the two are whirled together. It may be readily
shaped from glass tubing of 6-mm o. d. by drawing a capillary of 2-mm o. d.
and adjacent a capillary of 0.5- to I-mm bore. For calibration, weigh it
first empty and dry. Then add the desired weight (10 mg for 10,u1) and
allow water to enter through the tip of the pipet until the weight is balanced.
Mark the location of the meniscus by a line drawn with ink suited for
writing on glass. Of course, a ring mark may be etched and filled with
graphite (lead pencil) or ferric oxide. Keep the pipet between plugs of
cotton in a test tube with label indicating the delivery and the distance
of the mark from the tip of the pipet. For rinsing and drying, insert the
tip of the pipet into the rubber tubing r of the device Fig. 37 b and apply
suction.
For adjusting the meniscus to a given mark, hold pipets horizontally
and touch the tip either to the liquid to be taken in or to filter paper.
Expt.23 Work on the Milligram Scale 127

Experiment 23
Preparation and Calibration 01 Platinum Loops and Hooks
Platinum wire, 3- to 4-cm lengths of wire No. 31 (O.3-mm diameter); vise,
steel needle or paper clip, flat-tipped forceps, and calibrated capillary pipet.
Select a stiff wire or brad of the thickness wanted for the diameter
of the loop; 1 mm is suited for general work. Clamp the wire so that a
l-cm length of it is easily accessible. Use a 2- to 4-cm length of platinum
wire of 0.3-mm diameter. While holding it in one hand, grasp one end
with forceps and bend it around the steel wire to
obtain a completely closed loop, Fig. 40a. Press
the loop between two flat surfaces to get it into
one plane. Bend the wire close to the loop so that its
plane is inclined at an angle of 30 degrees to the rest
of the wire (169), Fig.40b. Finally, seal the other
end of the wire into a capillary drawn out of a glass
tube which is mounted in a test tube by means of a
stopper. It has been suggested that the loop is soldered
closed with gold (154), but this is not necessary if
the loop is handled with care.
Cleaning the Loop. Keep at hand a 30-ml glass ~.
stoppered bottle with concentrated hydrochloric acid. b
Immediately after each use and before use, rinse
the loop in running tap water, hold for a minute
into the acid, rinse it in running tap water and in
distilled water, and finally ignite it in a non-luminous
Bunsen flame until it glows without imparting a
Fig. 40. Platinum
color to the flame. A convenient supply of running Loop; a and bare
tap water is obtained by connecting one end of a approx. twice nat.
U-tube to a faucet and letting the water overflow at size
the other opening.
Calibration. Dip the clean loop into a beaker with water and withdraw
it slowly with the plane of the loop perpendicular to the surface of the
water. Without delay, touch the tip of the calibrated capillary pipet
to the liquid caught by the loop so that it is drawn into the pipet. Measure
the length of the liquid column with the millimeter ruler, and compute
the volume. Repeat the calibration five times and compute the mean
volume. The procedure may be improved by having some water in the
pipet from the start so that the displacement of the meniscus occurs only
in the cylindrical part of the tube.
Rate of withdrawal of the loop from the liquid and its position relative
to the surface of the liquid during withdrawal greatly affect the volume
of liquid taken up by the loop so that it· may change by a factor of two.
128 Techniques of the Sub milligram Scale

Use of Loop. The loop is filled as in calibration and then repeatedly

touched to or set down upon the surface which is to receive the reagent
(slide or wall of microcone) until the loop is empty. The loop is emptied
at the first contact with a clean glass surface. If the surface is slightly
oily, a succession of droplets will result, which may be wiped together
with the loop held so that its plane is vertical to the surface.
Platinum Hook. Mount platinum wire of 0.3-mm diameter and 3- to
4-cm length in a glass tubing just as this is done with a loop, Fig. 40.
Ignite the wire, and then give it a sharp bend 2 mm from the free end
so that an angle of 30 degrees results. Clean and use the hook like the
platinum loop. When slowly withdrawing it from water, it will hold a
droplet of about 0.1 Ill. The hook, like the loop, may be used for introducing
reagents directly into solutions, for stirring by twirling, and for withdrawing
small fractions of reaction mixture for side tests.

Techniques of the Sub milligram Scale

Under this heading, a collection of techniques shall be presented, which
may be applied to amounts of material ranging from several milligrams to
fractions of micrograms and attain sensitivities from several micrograms
to fractions of nanograms when applied to identification tests. The spot
tests and slide tests performed with approximately 0.5 to 1 III of solution
are customarily considered adjuncts of the milligram technique of working
in microcones, but a quick computation will show that, as a rule, only
a few micrograms of solute derived from the sample under investigation
are involved.
Spot Tests
To improve the sensitivity and sometimes also the specificity of spot
tests, they are performed on paper so that the test solution is slowly added
to a very small area of the paper. To this end, the solution to be tested
is taken up into a capillary with fine tip which is then touched to the paper
so that the solution slowly enters the paper at one point and spreads
radially from this point because of the capillary action of the interstices
in the paper. It seems that this procedure was first recommended by
F. L. HAHN (403, 859) and then systematically investigated by CLARKE
and HERMANCE (410) who also pointed out the advantages of papers
impregnated with insoluble reagents.
Adsorption or precipitation of the solute in the test solution will occur
in the small zone a where the latter enters the paper, Fig. 41. If the paper
contains a soluble reagent, the spreading solution may carry it out of zone b
and concentrate it in zone c before it gets a chance to react. The distribution
of an insoluble reagent in the paper, however, will not be changed by the
passage of fluid, and the zone a of precipitation or reaction will grow in
 Spot Tests 129

proportion with the amount of reactant introduced with the test solution.
Under all conditions, solutes which do not react and are not adsorbed
spread in the paper with the solvent. They may be more or less completely
washed into zone c and beyond by applying rinse liquid at a with a capillary
pipet in the same manner in which the test solution has been added.
Spot t.ests on paper have the disadvantage that the outcome of a test
must be judged from the observation of a mere change of color, whereby
it remains undecided whether the colored substance is dissolved in the
liquid, adsorbed upon the fibers of the paper, or is separated as a third
phase, solid or liquid. Obviously, attention to the functions of the various
zones in the paper may indicate that a color change in zones b or c does

-,::.'
~
1
L
~
I'
It
c
a,

Fig.41. P erforma nce of Spot Tests on the Milligram Scale ; approx. twice nat. size

not snpport the same conclusions as a color change at the spot where the
liquid enters the paper. Consequently, proper performance and attention
to the general appearance of the test (location of the phenomena) may
prevent incorrect interpretation and improve the reliability of the tests.
The sensitivity of spot tests on paper is naturally greatly influenced
by the visibility of the color phenomena. Consequently, one will prefer
reagents that lead to the formation of strongly colored substances, dyes.
If the reaction product is a colored solid, the perceptibility of the color
will greatly depend upon the particle size. Very small particles approaching
colloidal dimensions or deposition as film upon the surface of the fibers
will be most desirable. It should not surprise, therefore, that the procedure
of adding the reagents as well as quality of the paper may profoundly
affect the sensitivity. Statements of the latter without reference to a
definite procedure must be considered approximations. Strongly absorbent,
thick drop test paper is suited for some tests, and thin, "ash-free" paper
for others (900, 920). Provided that the test solution is first allowed to
spread in the paper and that the latter is then sprayed with the reagent
solution, ACKERMANN (954) has shown that the sensitivity of tests for
Benedetti-Pichler, Identification 9
130 Techniques of the Submilligram Scale

anions and cations having little affinity to paper (alkalies) is the poorer,
the larger the area is, which is wetted by the test drop; this is an obvious
consequence of Beer's law. Since the wetted area will decrease with
increasing thickness of the paper, thick papers should favor tests for anions
and alkalies if the mentioned procedure is followed. The ion exchange
capacity of the paper determines its suitability for the detection of cations
with good affinity to the carboxyl groups of cellulose; thus, for uranyl,
lead, stannous, cupric, hydrogen ion, etc., the suitability of the paper is
determined by its action as ion exchanger, and the area which is wetted
by the test drop is of no consequence.
Blank tests with the paper will show whether or not contaminants
are present in objectionable concentration; the ash constituents (iron,
barium, calcium, magnesium, potassium, silica, and phosphate) are rarely
objectionable, but even good filter paper may contain up to 5 per cent
degradation products of cellulose, that give colorations with iodine and
may act reducing (1112).
The filter paper is best cut into squares of 2-cm edge or strips 2 cm X 6 cm
and stored in stoppered wide-necked bottles or covered Petri dishes. It
should not be touched with fingers, but handled with forceps having ivory
or plastic tips.
For impregnation with reagents, strips of paper are either sprayed
with the reagent solution or immersed in it for 20 to 30 minutes so that
they float freely in the bath. Then they are slowly withdrawn from
the bath and pressed between filter paper to remove the excess of
solution.
To precipitate reagents in the paper, CLARKE and HERMANCE (418)
soak the paper briefly in one solution, pass it through a wringer to remove
excess liquid, dry it, insert into the precipitant, etc. To remove the excess
of the last reagent, they place the paper upon an inclined glass plate and
run water over it. The papers are dried in an electrically heated oven
and stored in black envelopes like photographic papers.
If the substance to be identified is either strongly adsorbed by the paper
fibers or reacts with a reagent or dye in the paper, that is not extracted
by the test solution, the akro technique of SKALOS (881) may be used to
improve the limit of identification. The test paper is cut into strips, 15 mm
long and 3 mm wide, which are then cut to form a triangle of 15-mm height
upon a base of 3 mm. The test solution is applied to the very point of the
triangle by means of capillary pipet, loop, or hook-or by touching the
droplet of the test solution with the very point of the paper. The effect
may be observed with the aid of a magnifying glass. Additional improvement
of the sensitivity may be obtained by concentrating the test drop before
taking it up with the point of the reagent paper, by evaporating the test
drop to dryness and mopping up the residue with the moist point of the
Expt.24 Spot Tests 131

paper, or by adding small fractions of the test drop to the point of the
paper and drying after each addition.
Obviously, the cut surfaces constitute a major part of the area of the
test, especially at the very point of the paper. Consequently, special
care must be taken to avoid contaminating the paper when cutting it
into shape. A sharp edge on glass or porcelain must be used for cutting
if a test for iron shall be carried out.
Usually, the appearance of spot tests changes during drying. Frequently,
the color change is better perceptible when the paper has become dry,
but the appearance of the test may change within a few hours if it is exposed
to air and light. If the test shall be preserved for evidence, one may try
to improve its stability by thorough washing after development, drying,
exclusion of air by spraying with. suitable lacquer, impregnation with
paraffin, enclosing between glass plates or sheets of cellophane, and storing
protected from light in black envelopes.
A more or less crude estimation of quantity is frequently possible by
comparison with tests performed under like conditions with a series of
solutions containing known amounts of the identified substance.

Experiment 24
Test for Mercuric Mercury and Lead
Iodide test paper: Soak drop test paper in 1% KI solution, allow to dry,
cut into strips, and store in an amber bottle.
By means of a capillary pipet add l,ul mercuric test solution (10 mg Hg
per ml) to iodide test paper. An orange-red spot forms where the solution
enters the paper. Repeat the test with dilutions of the test solution
containing 5 mg, 2.5 mg, l.2 mg, 0.6 mg, and 0.3 mg Hg per milliliter,
respectively, obtained by repeated dilution with an equal volume of
I-F HNO s' Start with a drop of test solution and a drop of I-F HNOs
on a slide. Take into a wide capillary with narrow tip a l-cm length of
each solution and blowout the mixture upon a slide; this is the first dilution.
Use l,ul for the test, and take a l-cm length back into the wide capillary
to make the second dilution; etc. - Observe that the appearance of the
test changes with the more dilute solutions: the center of the spot remains
white, and the HgI2 precipitates in a ring zone. The color is seen best
when the paper has become completely dry.
Repeat the experiment with l,ug lead test solution (10 mg Pb per ml).
A yellow spot of PbI 2 appears where the solution enters the paper.
Prepare stannite reagent, just before use, by adding 2 drops of stannous
chloride reagent to 3 ml 2-F NaOH in a test tube and mixing to get a
clear solution.
9·
132 Techniques of the Submilligram Scale Expt.25,26

To confirm the presence of mercury and lead, treat the colored spots
by adding to each a drop of stannite reagent. The mercuric iodide is reduced
to gray or black mercury, and the lead iodide dissolves to give colorless
plum bite solution.
Also other metals of the hydrogen sulfide group give spots on iodide
paper when present in sufficient quantity. Silver gives a pale y~llow spot
of AgI which turns black with stannite; bismuth gives a brown to black
spot which turns black with stannite; cupric and stannic ions give blue
spots (reaction of liberated iodine with the cellulose) which fade with
stannite reagent. Cadmium, antimony, and arsenic leave the iodide paper
unchanged.

Experiment 25
Chromate Test for Silver
Solution of ammonium chromate, 2 g in 100 ml water; sodium chloride,
1% in water.
By means of a glass rod, place drops of ammonium chromate solution
upon a strip of drop reaction paper. Allow the paper to dry for five minutes
at room temperature. Then, by means of a capillary pipet, add 1 pI silver
test solution (10 mg Ag per ml) to the center of one of the yellow chromate
spots. A brownish red spot forms where the solution enters the paper;
if only little silver is present, the brown spot may appear 2 to 5 minutes
after adding the test solution.
The test may be made more distinct by adding with the capillary pipet
10 to 30,t! distilled water to the center of the brown spot. The soluble
yellow (NH4)2Cr04 is carried away to leave the brown Ag 2Cr0 4 in the
white paper.
Confirm the presence of silver by finally adding a drop of sodium
chloride solution to the brown spot. The silver chromate is converted to
chloride, and the brown spot disappears immediately.
Mercuric mercury, cadmium, tin, and antimony do not give a test.
Lead and bismuth produce yellow spots. The brown spot given by cupric
ion after rinsing with water remains unchanged when treated with sodium
chloride.

Experiment 26
Tests for Bismuth and Antimony
Quinine iodide reagent: dissolve 1 g quinine hydrochloride in 50 ml water
and add 0.2 ml 6-F HCI; dissolve 2 g KI in 50 ml water; before use, mix equal
small volumes of the two solutions. - Cinchonine iodide reagent: dissolve 1 g
cinchonine hydrochloride in 19 ml water and add 1 ml 6-F HCI; dissolve 2 g KI
in 10 ml water; before use, mix 2 small volumes of the former solution with
Expt.26 Spot Tests 133

1 volume of the iodide solution. - Iodide-acetate reagent: dissolve 25 g ammonium

acetate and 10 g KI in 50 ml water. - Stannous chloride reagent: 11 g SnOl 2 • 2 H 20
dissolved in 17 ml 12-F HOI and diluted with water to 100 ml; keep in bottle
containing 1 g metallic tin.
Mark six test points on a strip of drop test paper. Place the paper
upon a glass plate and press down upon it with the end of a glass rod
that has been drawn out to a point and cut and polished to obtain a flat
circular surface of 2-mm diameter (948). Do not press hard enough to
tear the paper. Using a lead pencil and writing at some distance from the
indentations, label them Bi, Cu, Sb, Sb, Bi, and Pb.
With capillary pipets add to each of the marked spots about 1 #1 of
the indicated test solution (10 mg ion per ml) and then lay the paper aside
for drying. It is not necessary to'use a calibrated capillary pipet; in this
instance, it will suffice to crudely estimate the diameter of the bore and
from this the length of the capillary, representing 1#1. As an alternative,
the test solutions may be measured with a platinum loop and taken from
the loop into the capillary pipet for transfer to the paper. The contaminated
end of the pipet may be cut off before proceeding to the next solution,
and a fresh tip drawn.
When the droplets have evaporated on the paper, use a glass rod or
a dropper to treat the first three spots (Cu, Bi, Sb) with small drops of
freshly mixed quinine iodide reagent (433, 1135) and the last three, with
freshly mixed cinchonine iodide reagent (1158). Record the color of the
spots. When they have become dry, treat the first three spots with a drop
each of freshly prepared stannite reagent (Expt. 24) and the last three
spots by adding to each 1 drop of iodide-acetate reagent.
Quinine iodobismuthite gives an orange-red precipitate (spot) which
is reduced to black metallic bismuth upon adding stannite reagent; the
corresponding iodoantimonite is orange-yellow and dissolves in stannite
reagent to give a colorless solution. The cinchonine iodobismuthite gives
a bright orange-red spot which changes little on adding iodide-acetate
reagent; the corresponding antimony compound'is orange-yellow, and the
spot changes immediately to white when iodide-acetate reagent is added.
A brown ring may form around the white central area, but it disappears
when the spot dries.
Mercuric mercury, cadmium, and arsenic do not visibly react with
the reagents. Lead gives yellow PbI 2 which remains unchanged with
iodide-acetate reagent and dissolves in stannite reagent. The pale yellow
AgI is reduced to black metallic silver by stannite reagent. Cupric ion
gives a brown to black spot which remains nearly unchanged when stannite
reagent is added. Stannic tin may give a blue spot of liberated iodine
acting upon the cellulose.
134 Techniques of the Submilligram Scale Expt. 27, 28

Experiment 27
Test tor Copper, Nickel, and Cobalt (1150)
Rubeanic acid, diamido-dithio-oxalic acid, 1% solution in ethanol.
With a capillary pipet transfer to different locations on a strip of drop
test paper about l.ul each of copper, nickel, and cobalt test solutions
containing 10 mg ion per ml. Expose the moist spots to ammonia fumes
by holding the paper over the opening of a bottle containing strong ammonia
solution. Finally treat each spot with a drop of rubeanic acid solution,
and record the colors of the precipitates. The tests are quite specific.
The copper salt is insoluble in acetic acid, whereas the salts of nickel
and cobalt will not precipitate or precipitate only partly from acetic acid
solution, depending upon the concentration of the acid. Flow through
paper will produce concentration gradients aiding in the separation of
the metal ions (121).
Mix small equal volumes of the copper, nickel, and cobalt test solutions,
and treat the mixture with an equal volume of 4-F acetic acid. Soak a
strip of drop test paper in rubeanic acid solution and let it dry. With a
capillary pipet transfer about l.ul of the acidified salt solution to the
impregnated paper. The olive green or black coloration obtained where
the liquid enters the paper indicates the presence of copper which may
be thus detected up to the limits Ou: 00 : Ni = 1 : 2000 : 20000. Try to
improve the simultaneous identification of nickel and cobalt by first
rinsing from the center of the spot with 10 to 20.u1 of I-F acetic acid added
from a capillary pipet. Finally expose to fumes of ammonia_

Experiment 28
Test tor Cadmium (590, 591)
Cadion test paper: soak ash-free filter paper or drop test paper in 0.02%
solution of Cadion 3 B, benzenediazoaminobenzene-4-azo-4'-nitrobenzene, in
ethanol. Press between blotting paper and allow to dry; cut the dry paper in
strips. - Rochelle buffer: dissolve 10 g Rochelle salt in 100 ml water and 0.1 ml
glacial acetic acid.
Mix a small volume of cadmium test solution with an equal volume
of Rochelle buffer solution, and transfer about l.ul of the mixture with a
capillary pipet to the Oadion test paper. Allow the test drop to evaporate
at room temperature (heating is not permissible), and then add with a
glass rod a drop of a mixture of 4 volumes 2-F KOH with 1 volume ethanol.
The test paper becomes purple, but the circular area where the cadmium
solution entered the paper turns pink or salmon red.
The diameter of the pink spot is related to the amount of cadmium.
For the detection of small amounts of cadmium, place one drop of reagent
Expt.29, 30 Spot Tests 135

solution on drop test paper, add the test drop to the center of the spot,
and then add the KOH.
Interferences by copper, nickel, cobalt, iron, chromium, and magnesium
are avoided by the addition of the tartrate. Mercuric mercury gives a
yellow spot which might be mistaken for cadmium, but changes to gray
on adding stannite reagent, Expt. 24. Tin solutions produce a white spot,
and the dye is also destroyed by heat or strongly acid solutions. Other
common elements of the hydrogen sulfide group have no visible effect.

Experiment 29
Precipitation of Silver Arsenate
Buffered silver solution: dissolve 1 g AgN0 3 and 7.7 g ammonium acetate
in a mixture of 6 ml glacial acetic acid and 200 ml water.
With a capillary pipet transfer about l,ul slightly ~cid or alkaline
arsenate test solution (10 mg As per ml) to ash-free filter paper. Lay the
paper aside to dry, and then add 5 to 10,u1 buffered silver solution by means
of a capillary pipet.
Brown Ag3As04 separates. An amount of chloride ion equal to that
of the arsenate does not hinder the detection of the latter, but the appearance
of the brown spot is somewhat delayed. Large amounts of halide must
be absent. Antimony and tin do not give colored reaction products.

Experiment 30
Molybdenum Blue Test for Tin (121)
Phosphomolybdate paper: soak strips of ash-free filter paper in 1% aqueous
solution of phosphomolybdic acid; pour a few milliliters of strong ammonia
solution into a 400-ml beaker, and expose each strip of paper for lO to 20 seconds
to the ammonia fumes by holding it into the gas space of the beaker immediately
after removing it from the phosphomolybdic acid bath; allow the strips to dry,
and store them in bottles of amber glass. - Magnesium ribbon.
Place a clean glass plate upon a sheet of black paper. Cut magnesium
ribbon first into strips of about I-mm width, and then cut the strips into
0.5-mm lengths and allow the small squares to drop on the glass plate.
Take about l,ul of stannic test solution into a capillary pipet. Deposit
a large drop of 12-F HOI upon a glass slide, and allow about l,ul of this
acid to enter the pipet containing the tin solution. Blowout the contents
of the pipet upon a glass slide, and there treat the droplet with a small
square of the magnesium ribbon. First dip the end of a glass thread into
the tin solution; then touch the metal square with it and transfer it into
the test drop.
When the evolution of hydrogen has stopped, insert the tip of the
capillary pipet into the droplet and transfer the clear solution without
136 Techniques of the Submilligram Scale

delay to phosphomolybdate test paper. A blue spot forms, which becomes

more distinct when the paper is laid aside and allowed to dry.
Arsenic, antimony, and mercury solutions do not give reduction to
molybdenum blue under the conditions specified for the tin test.
Repeat the test (inclusive reduction with magnesium) with 1,tt1 cupric
test solution, and note that a blue spot is obtained, which disappears
when the paper is laid aside to dry. The presence of copper also promotes
the fading of the molybdenum blue produced by stannous tin if there is
not present at least four times as much tin as copper.

Slide Tests
Typical slide tests are based upon the separation from solution of more
or less coarsely crystalline precipitates. They are performed upon micro-
scope slides, and the shape of the crystals is observed under the microscope.
When comparing the different types of chemical confirmatory tests
with regard to specificity and reliability, the slide tests must be given
the highest rating. With spot tests and fiber tests, the outcome rests
essentially upon the observation of color phenomena. At times it will
be impossible to tell whether or not a new phase has formed. The test tube
technique is somewhat superior, for, aside from the color effects, the separa-
tion of phases and their general appearance can be observed without
difficulty. In addition to all these criteria, however, slide tests reveal
also the shape of the particles of a new phase formed. The convincing
finality of microscopic identification recommends slide tests for general
use on any scale of work.
Since the particles of precipitates may remain so small that even the
microscope is unable to reveal their shape, care is taken to favor the
formation of few nuclei so that they may grow to reasonable size with
the small amount of material available. The reactions are intentionally
selected and conducted to favor the formation of relatively large crystals,
the dimensions of which range from a few micrometers to a few millimeters.
Rather soluble precipitates are preferred, and in all instances conditions
are established, which give a low rate of nucleation as a consequence
of the continuous maintenance of a low degree of supersaturation. In some
instances, crystals of the desired size are obtained by recrystallization
from suitable solvents.
Examples of the various techniques employed may be found in Expts. 31
to 41. The reagent is frequently added in the solid state so that it may
spread slowly through the test drop by gradual dissolution followed by
diffusion. If the reagent is a liquid, the test solution may be evaporated
and the solid residue treated with the liquid reagent. The test drop and
the reagent drop may be placed side by side upon the microscope slide
 Slide Tests 137

and then connected by a narrow channel which prevents instantaneous

mixing. If a gas is involved, the rate of precipitation may be controlled
by the rate at which the gas is added or removed.
It is desirable to begin the microscopical observation of the test at least
immediately after adding the reagent. Observation of the dissolution of
the reagent, accompanied or immediately followed by the separation
of the test forms, demonstrates thc connection between cause and effect
in a very cGnvincing manner. If the test is set aside after adding the
reagent and the crystalline precipitate is observed after lapse of some time,
one cannot be certain whether the crystals have been formed by the action
of the reagent or because of the evaporation of solvent. To arrive at a
decision in such instances, it will be necessary to compare with a second
test drop that has been standing equally long without addition of reagent,
but even such comparison may fail to furnish equally convincing proof.
The time of appearance and the location of the precipitate in the test
drop provide valuable clues that help avoiding misinterpretations. Crystals
which are slowly growing along the edge of the drop form as a consequence
of evaporation and the resulting rise of concentration, which is most
pronounced along the edge of the drop. Crystallization caused by the
immediate action of the reagent takes place close to the reagent, and,
as a general rule, the closer the crystals are to the reagent, the smaller is
their size.
The general appearance of tests makes it possible to draw conclusions
concerning the conditions in the test drop, and the information may be
used to recognize the changes necessary for a more successful repetition
of the test. If the solid reagent does not dissolve or dissolves very slowly
in the test drop and very few small crystals or none separate close to the
reagent, the concentration of the sought-for substance may be so high
that the reagent becomes coated with the insoluble test form so that it
cannot dissolve; diluting the test drop may be followed by copious separation
of characteristic crystals. Also instantaneous separation of a dense, powdery
or gelatinous precipitate close to the reagent indicates that the concentration
of the test drop is too high; larger crystals may form after some time at
the periphery of the area of granular precipitation, but even these may
remain quite small if too little reagent was added so that the granular
precipitate consumed all or nearly all of it. The most favorable conditions
prevail when small crystals form close to the reagent and large ones at
a distance. With somewhat lower concentrations of the sought substance,
the number of small crystals separating close to the reagent decreases,
and there is a distinct trend toward the formation of medium-sized crystals
only. With further dilution of the test solution, size and number of the
crystals of the test form decrease; finally, no precipitation or crystallization
follows upon adding the reagent, and crystals of the test form separate
138 Techniques of the Submilligram Scale

only after the concentration has been raised by evaporation. Under these
conditions, the crystals are small and usually appear first at the edge of
the drop. Experienced workers usually succeed in keeping near optimum
conditions by suitable adjustment of the concentration of the test solution
by proper choice of solvent volume, diluting the solution, or concentrating
it by evaporation. Naturally, the solution volume may be adjusted at will,
and the quantity of sought substance present may be estimated from the
amount and nature of the sample or the volume of precipitates and other
observations made in the course of the analysis.
Obviously, all dissolved substances will appear in the residue when a
test drop evaporates to dryness. The soluble substances contained in the
test solution usually form crystals that are much larger than those of the
test form, and complete evaporation of the drop gives a crust in which
the crystals of the test form become hidden. If the latter are quite insoluble,
adding a drop of solvent will render them again observable by dissolving
the incidental solids; if they are reasonably soluble, however, the appearance
of the test must be restored by the very cautious adding of small amounts
of solvent. This may be done by exposure to solvent vapors (breathing
upon the residue of the test drop). When working in a very dry atmosphere,
the rapid evaporation of test drops may become a nuissance; it may be
prevented by frequently breathing upon the slide, covering the test drop
with a small watch glass, or placing the slide into a humid atmosphere
maintained inside a desiccator or under a bell jar; the slide with the test
drop may be made the floor or ceiling of a gas chamber containing a drop
of solvent.
The volume of the test drops is optional; naturally, the limits of
identification will improve in proportion with the reduction of volume.
In the following experiments, drops of 0.3- to I-pI volume are used to get
well below the milligram range. Working with much smaller drops requires
use of a moist chamber to prevent their rapid evaporation, p. 198.
For work with aqueous solutions, the microscope slides should be
slightly oily so that small drops do not spread without, however, assuming
hemispherical or even spherical shape. This condition may be obtained
by wiping the slides with tissue paper containing a trace of sebum
(oil) from the face or scalp. Cover slips are rarely used; if they
are placed upon the drop after adding the reagent, the characteristic
appearance of the test may be completely destroyed, and· crystals of the
test form may be lost by being carried to the edge of the cover slip where
they remain hidden to view. Omission of the cover slip also makes it
possible, after washing and (or) evaporation of the drop, to treat
individual crystals with reagents for further convincing confirmation of
their identity.
Expt. 31,32 Slide Tests 139

Experiment 31
Silver Dichromate (125)
Potassium dichromate, granular, I-mm diameter.
Upon a slide, mix 5,tt1 silver test solution (10 mg Ag per ml) with 1,tt1
16-F HNOs. Take up the mixture into a capillary pipet and transfer it
to another slide to obtain the droplet in a suitably small area. Using
transmitted light and a magnification of 20 to 80 diameters, focus the edge
of this drop under the microscope. At some distance from the drop, deposit
upon the slide some K 2Cr20 7 •
Draw a glass thread of 0.1- to 0.2-mm diameter from the end of a capillary.
Moisten the end of the thread by dipping it into the test drop, and then
pick up with it a kernel K 2Cr20 7 of about I-mm diameter (0.5-,tt1 volume)
and place it into the edge of the test drop (this may be done while observing
through the microscope, compare Expt. 61). Without delay observe the
test under the microscope.
Spears of Ag 2Cr20 7 grow out of the reagent kernel and form also at
some distance from the reagent which spreads in the test drop as may
be seen from the yellow coloration. The crystals are yellow to deep red
depending upon their thickness; they are triclinic and show strong bi-
refringency and a slight degree of pleochroism, light to dark. Prepare a
sketch and collect into it all typical forms observed at this time and on
later occasions. Relatively large crystals of yellow K 2Cr2 0 7 and colorless
KNO s will be observed when the test drop dries out.
Distilled water or glycerol may be used for moistening the end of
the needle previous to picking up the solid reagent. The use of saliva is
not advisable for obvious reasons, and the removal of reagent directly
from the vial with a moistened needle is likewise objectionable. Naturally,
the unused reagent should not be returned to the reagent bottle or vial.
The portion of the glass thread, which 'has come into contact with the
reagent or the test solution, is broken off and discarded.

Experiment 32
Recrystallization 01 Silver Ohloride (125)
Watch glass, 2.5-cm diameter. - Silver chloride, powder.
With a spatula, transfer some silver chloride to a slide. Select a cluster
of 0.5-mm diameter (0.05-,tt1 volume), push it to an empty area of the slide,
and treat it with 5,tt1 6-F NHs from a capillary pipet. Without delay, stir
the mixture with a glass thread, take up the clear solution into a capillary
pipet and transfer it to another slide. Immediately cover the droplet with
a small watch glass so that it projects slightly over the edge of the slide,
Fig. 42. The small opening between slide and watch glass permits slow
140 Techniques of the Submilligram Scale Expt.33

escape of the ammonia, and the rate of its escape from the solution is
sufficiently retarded to give relatively large crystals of AgCl.
Allow to stand for 10 minutes with the watch glass in place. The
separation of AgCI is indicated by the appearance of a turbidity visible
to the unaided eye; a clean watch glass does not interfere with the inspection
of the drop under the microscope when a low magnification is used.
Finally, remove the watch glass and examine the crystals with
a magnification of 80 diameters or more. When the drop has evaporated
completely, place upon the residue a fragment of a cover slip with a droplet
of 4-F HNO a hanging on its underside. The drop spreads between slide
and cover slip and dissolves all solids but the AgCl. In addition, the cover
slip makes possible the efficient use of medium magnifications of 100 to
300 diameters.
Colorless octahedra, tetrahedra, cubes, and combinations may be
observed. When opportunities offer themselves, add to the sketch drawings

Fig. 42. Retarding Evaporation; 2/3 nat. size

of skeletal forms and spherulites (168) which may be obtained if the silver
chloride has been precipitated from solutions containing ions of other
heavy metals.
Considering that the insolubility in acids and the solubility in ammonia
is demonstrated in addition to the cubic nature of the crystals, the test
appears specific for silver. In addition, the identification of AgCI is aided
by its high refractive index which gives heavy outlines to the larger crystals
and renders small crystals entirely black in transmitted light (reflected
light shows that they are colorless).

Experiment 33
Lead Iodide
Potassium iodide, granular.
This test has found special favor with HEMMES (154).
Mix equal small volumes of lead test solution (10 mg Pb per ml) and
6-F acetic acid. With a capillary pipet, transfer 10,u1 of the mixture to
the center of a microscope slide. Using a magnification of about 20 diameters
and transmitted light, focus the edge of the drop under the microscope.
With a glass thread, place a KI grain of 0.5-mm diameter into the edge
of the drop and observe that it dissolves rapidly while yellow hexagonal
Expt.34 Slide Tests 141

plates of PbI 2 appear in a circular zone around the reagent. As the KI

diffuses through the drop, the zone of precipitation moves away from the
center where the PbI 2 is dissolved in the excess of KI. Prepare a sketch
showing the general appearance of the test and collect in it drawings of
all typical shapes observed.
The plates of PbI 2 are usually thin enough to show interference colors.
Use a dark background and strong light from the side, and observe during
crystallization when the particles are kept in motion by the convection
currents in the drop. When the test drop finally becomes more concentrated
by evaporation, fine needles of colorless KPbI 3 • 2 H 2 0 appear in the
center of precipitation where the reagent was introduced. Observe with
crossed nicols, and rotate the stage.

Experiment 34
Potassium-Lead-Copper Nitrite (1l5, 116)
Potassium nitrite, 5.8-F: di9solve 5 g KN0 2 in water to give 10 ml solution;
acet<1te buffer solution: treat 4.5 g sodium acetate trihydrate with 1 ml glacial
acetic acid and add water to get 10 ml solution.
To prepare the reagent, transfer to a micro cone 0.05 ml of 5.s-F KN0 2
and a like volume of sodium acetate buffer solution. Mix thoroughly.
This reagent will keep for two days if the cone is stoppered and kept away
from strong sources of heat or light (S75). After preparation, it should be
tested with a known lead residue before using it on an unknown.
By means of a platinum loop, deposit several small droplets o.f lead
test solution near one end of a microscope slide and 5 to 10 mm apart.
Hold the slide about 5 cm above a small Bunsen flame, and evaporate
the droplets to dryness without overheating the residues; just heat the
slide until the evaporation starts, and then remove it from the flame
and finish the evaporation by blowing upon the droplets with the mouth.
Quickly cool the slide by placing it upon metal, a metal block, or the base
of the microscope. By means of a capillary pipet, moisten each residue
with I % (I-F) CuS0 4 solution (10 mg Cu per ml), and again evaporate to
dryness and cool to room temperature. Finally, with a platinum loop,
hook, or capillary pipet, treat one residue with a small amount of nitrite
reagent. To avoid seeding, do not touch the slide with the tool, and
add so little reagent that it does not completely cover the residue. Without
delay, inspect with a magnification of SO to 100 diameters and strong
transmitted light. Use the front lens of the condenser.
The small squares and rectangles of K 2CuPb(N0 2 )6' of 10- to 25-ttm
edge, form either immediately or, with small quantities of lead in a relatively
large drop, after a few minutes. The color varies from yellow to black,
depending upon the thickness of the crystals. Bubbles of nitrogen oxide
142 Techniques of the Submilligram Scale Expt. 35

have circular outline. When evaporation nears completion, various green

and colorless salts begin to crystallize and the observation of the triple
nitrite becomes difficult or impossible.
The appearance of the precipitation clearly indicates the changes that
should be made to improve the test.
1. If crystals of the triple nitrite are not obtained, or only a few small
ones grow near the edge of the drop within 5 minutes from the addition
of the reagent, either the latter is spent, or too much of it has been taken.
The triple nitrite is rather soluble, and it is necessary to keep the volume
of the reagent at a minimum. Repeat the test by adding a much smaller
volume of reagent to the next residue. If no improvement is obtained,
it is obvious that the reagent does not have the proper composition. Prepare
a new batch of reagent; at times, it becomes necessary to prepare a fresh
solution of potassium nitrite.
It should be understood that the slow appearance of few small crystals
of triple nitrite must be expected whenever the quantity of lead approaches
the limit of identification.
2. If a brown or black mass of precipitate forms immediately and the
individual crystals are so small that their shape cannot be recognized,
the supersaturation in the drop is too high. Repeat the test with another
residue as follows: breathe upon the slide so that the residue liquifies,
and then add the reagent.
3. If relatively large black crystals form within 3 minutes, the conditions
in the drop are properly established.
The triple nitrite test is specific for lead since no other element is known
which could substitute for it without changing the color of the crystals.
Barium, strontium, and calcium may substitute for lead, but the crystals
will be colorless or green. Nickel replaces copper without giving much
change in the appearance of the precipitate; rubidium, cesium, and thallous
ion may be substituted for the potassium, which decreases the solubility
of the triple salt. Concerning the amount of copper to be added in the
test for lead, the ratio Pb : Cu = 1 : 10 gives the best results; it need not
be closely adhered to since the limiting proportions allow large deviations.
The test may be directly applied to insoluble lead salts such as the sulfate,
see Expt. 47.
Experiment 35
Oesium Iodobismuthite (115, 128) and Oesium Iodoantimonite(115, 168)
Cesium chloride, granular; potassium iodide, granular; stannite reagent,
Expt.24.
With a capillary pipet, transfer 5,al bismuth test solution to a slide.
For adjusting the acidity of the test drop, evaporate just to dryness and
dissolve the residue by adding 5,ul 2-F HNOs when the slide has attained
Expt.36 Slide Tests 143

room temperature. Place a grain of CsCI of I-mm diameter into the edge
of the drop; it dissolves quickly, and colorless cesium chlorobismuthite
separates around the dissolving reagent. Hexagonal plates are usually
seen in the midst of a variety of other forms.
Without removing the slide from the stage of the microscope, introduce
a grain of KI of I-mm diameter into the edge of the drop and just outside
the zone of precipitation of chlorobismuthite. Yellow, red, and nearly black
hexagons and stars form where the zones of diffusion of the CsCI and the
KI meet. The intensity of color is, as usual, determined by the thickness
of the crystals. When the diffusing iodide reaches the colorless crystals
of the chlorosbismuthite, they become gradually converted to orange
iodobismuthite.
Set the slide aside until the test drop has evaporated to dryness. Then,
using a magnification of 20 to 50 diameters, illumination with reflected
light, and a green or blue background, focus upon a portion of the prepara-
tion, which contains well developed crystals of the iodobisml1thite. While
observing through the eyepiece, add a large drop of stannite reagent
from a medicine dropper, but be certain that none of the reagent gets on
the lenses of the microscope (it strongly acts upon optical glass, and reduction
of lead glass may cause blackening).
Observe that the crystals of iodobismuthite turn black and opaque
while mostly retaining their shape. They become pseudomorphs of tiny
particles of metallic bismuth, which remain clustered together to keep
the shape of the iodobismuthite. The shape of pseudomorphs reveals
their history but bears no relation to their lattice structure.
Repeat the test with 5,Ill of antimony test solution (10 mg Sb per ml).
Adjustment of the acidity by evaporation is not possible because of the
volatility of the chlorides of antimony, and it is not necessary since the
HCI of the test solution will not oxidize the iodide. Thus add the CsCI
and then the KI directly to the drop of test solution. Observe that antimony
behaves much like bismuth. Finally, allow the test drop to evaporate
to dryness, and then treat it with stannite reagent which dissolves all
antimony compounds and thus permits distinguishing between antimony
and bismuth.
It may be mentioned that solutions of tartar emetic, K(SbO)C 4 H 4 0 s ·
.1/ 2 H 20, must be acidified with HCI in order to give the test with CsCI

and KI.
Experiment 36
Bismuth Gobalticyanide Pentahydrate (1143)
Potassium cobalticyanide, granular (the preparation of the salt is described
in the paper cited above); stannite reagent, Expt. 24; bismuth-lead test solution:
dissolve 1.2 g Bi(N0 3 h· 5 H 20 and 8 g Pb(N0 3 )2 in 3-F HN0 3 to make 100 ml
solution.
144 Techniques of the Submilligram Scale Expt.37

Transfer 5.ul of bismuth test solution to a slide and adjust the acidity
as in Expt. 35 by evaporation and dissolution of the residue in 5.ul 2-F HNO a.
Place a grain of KaCo(CN)s of I-mm diameter into the edge of the drop
and observe with transmitted light and a magnification of 70 to 100 diameters.
Close to the reagent, the BiCo(CN)s· 5 H 20 precipitates as a fine powder,
but larger crystals grow slowly at the outermost boundary of the area of
precipitation and exhibit very characteristic shapes, drawings of which
should be collected. The crystals grow rather slowly, and the observation
should be continued for 5 minutes or more. Remarkable twin crystals are
occasionally observed.
Finally set the test aside and allow it to go to dryness. Using a magnifica-
tion of 20 to 50 diameters, illumination for observation in reflected light,
and a colored background, focus upon characteristic crystals and add a
large drop of stannite reagent as in Expt. 35. Black pseudomorphs of
metallic bismuth will be obtained.
The shape of the crystals of bismuth cobalticyanide is greatly modified
by the presence of lead, stannous tin, and mercuric mercury. For a
demonstration, repeat the above experiment with a solution containing
50 mg of lead and 5 mg of bismuth per milliliter. The resulting precipitate
which contains about 10% Pb consists of lens-shaped forms that, seen
from the front, exhibit the appearance of oil drops; side views have sharp
outlines. Stannous tin behaves very much like lead, and the lens shapes
are converted to black pseudomorphs by adding NaOH solution; the
required stannous ion is already contained in the lenses (and in the residue
of the test drop).
Experiment 37
The Mercurithiocyanates of Copper, Zinc, Cadmium,
and Cobalt (115, 118, 896)
Potassium mercurithiocyanate, granular: dissolve 27 g HgC1 2 in lOO ml
boiling water and precipitate by slowly adding a solution of 19 g KCNS in
20 ml water; cool to room temperature, collect the Hg(CNS)2 in a BuchnEr funnel,
and wash it with water and dry; dissolve 16.8 g KCNS in 15 ml water, and add
40 g Hg(CNS)2; stir and add water until solution is complete; transfer to a flat
dish and place for evaporation into a desiccator with 18-F H 2S0 4 ; grind to grains
of 0.5- to I-mm diameter. - Ammonium mercurithiocyanate solution: dissolve
5 g HgC1 2 and 5 g NH~CNS in 6 ml water.
Deposit l-.ul portions of copper, zinc, cadmium, and cobalt test solutions
(10 mg metal ion per ml) upon a slide so that the drops are kept 5 to 10 mm
apart. Complete the collection by adding l-.ul portions of the following
six mixtures of test solutions: 1 volume of Zn (test solution, 10 mg ion
per ml) + 10 vols. Cu (test solution); 10 vols. Zn +
1 vol. Cu; 1 vol. Zn +
+ 10 vols. Co; 10 vols. Zn + 1 vol. Co; 10 vols. Cd +
1 vol. Co; equal
volumes of Zn, Cu, and Co test solutions.
Expt.37 Slide Tests 145

If necessary, warm the slide and blow air upon it until all test drops
are evaporated to dryness. Treat one residue after the other as follows:
dissolve it by adding to it I,ul 2-F HN0 3 , and place into the edge of the
resulting drop a grain of 0.8-mm diameter of K 2Hg(CNS)4; observe the
resulting precipitates with suitable magnification and illumination (strong
transmitted light as well as reflected light with white background). Prepare
drawings.
The tests for zinc, copper, and cobalt are quite specific. In practice,
one will separate these metals from other heavy metals before applying
the test. To increase the certainty of identification, it is recommended to
repeat the zinc test after adding to a fresh test drop a small amount of
copper, to repeat the copper test with some zinc added, and to repeat the
cobalt test with some cadmium added. In all instances, it will be wise
to adjust the amount of added metal to that of the metal to be identified
to a ratio giving a striking effect (1 copper to 5 or lO zinc, and 1 cobalt
to 10 cadmium).
Ferrous and auric ion give yellow needles and dendrites which may
be mistaken for those of the copper mercurithiocyanate. Ferric salt
imparts a red coloration to the solution, and dark red, almost black hexagons
may separate. Manganese, lead, and silver give colorless crystals which
may be mistaken for the cadmium or zinc mercurithiocyanate. Only strong
solutions of nickel give small pale yellow disks and masses of radiating
needles which may appear brown in transmitted light (Tyndall effect) (88).
Mercuric mercury precipitates Hg(CNS)2; iodide and cyanide should be
absent since they bind the mercury of the reagent.
SCHOORL (168) has demonstrated in the instance of the cobalt that
the identification limit may be considerably improved by decreasing the
volume of the test drop. In addition, one may increase the bulk of the
test form by adding a second metal ion and precipitating the mixed
mercurithiocyanates copper-zinc (1: 10), zinc-copper (1: 3), cobalt~zinc
(1: 10), and cadmium-cobalt (1: 10) instead of the simple mercurithio-
cyanates of copper, zinc, cobalt, and cadmium, respectively. Determine the
resulting limits of identification of the cobalt and of the zinc tests as follows.
Prepare a small volume of a dilution of 1 volume of cobalt test solution
with 9 volumes of water. Mix, and dilute 1 volume of the mixture with
4 volumes of water to obtain a solution containing 0.2 mg Co per ml.
Continue diluting in the ratio 1 to 5 to obtain solutions with 0.04, 0.008,
0.0016, and 0.0003 mg Co per ml. Start with the strongest solution (1 mg Co
per ml), and transfer two I-,ul portions each of every dilution to a microscope
slide which is sufficiently oily to prevent the spreading of the droplets.
Warm the slide and blow upon it to evaporate the droplets to dryness.
After cooling to room temperature, moisten the first, largest residue with
ammonium mercurithiocyanate reagent, and inspect under the microscope.
Benedetti-Pichler. Identification 10
146 Techniques of the Submilligram Scale Expt.38, 39

Add the reagent with a platinum loop of about 0.3-,al capacity, and avoid
touching the glass with the platinum, which might give seeding and
numerous small crystals. Repeat the test with residues containing less
cobalt until the dilution has been found, that does no longer give a positive
test in every trial. Then return to the next stronger solution and try to
obtain the test with smaller volumes of the solution: 0.5, 0.3, 0.1 ,aI, which
may be transferred to the slide with a platinum loop or hook. It is essential
that the evaporation residues cover correspondingly smaller areas on the
slide and that correspondingly smaller amounts of reagent are added by
means of smaller loops or platinum hooks. Compute the smallest mass
of cobalt, which reliably gives a positive test; it represents the limit of
identification.
In a similar way determine the limit of identification for copper by
starting with a mixture of 1 volume copper and 9 volumes zinc test solution.
Analogous mercuriselenocyanates are mostly of theoretical interest
because of the instability of the reagent, but they may be used to test
cobalt salts for presence of small amounts of iron, copper, and nickel (863).

Experiment 38
Test lor Oadmium with Brucine and Bromide (136, 560)
Brucine bromide reagent: dissolve 0.65 g brucine and 1 g NaBr in 5 ml
4-F acetic acid and 15 ml water; sodium sulfide-hydroxide reagent: dissolve
24 g Na 2 S· 9 H 20 and 2 g NaOH in water to make 50 ml solution.
Transfer l,al cadmium test solution (10 mg Cd per ml) to the microscope
slide and deposit next to it a drop of like volume of brucine bromide reagent.
With a glass needle or the platinum loop, draw a channel to connect the
two drops. Prepare a drawing of the general appearance of the test showing
the characteristic clusters of supposedly monoclinic plates. - It is necessary
that the cadmium solution does not contain free acid; if it does, evaporate
the test drop on the slide to dryness and dissolve the residue in water.
When the test has evaporated to dryness, treat the residue with a
relatively large drop of sodium sulfide-hydroxide reagent, which has been
diluted with 10 volumes of water, and inspect the "pseudomorphs" of CdS
with reflected light and a white or purple background.

Experiment 39
Test for Mercury with Zinc, Oopper, and Thiocyanate (119)
Potassium thiocyanate, 5% solution; mixture of equal volumes of zinc and
copper test solutions (5 mg Zn and 5 mg eu per ml).
This reversal of the zinc test of Expt.37 is specific for mercury.
Evaporation is carried out in presence of nitric acid to reduce volatilization
of the mercuric chloride.
Expt.40 Slide Tests 147

Deposit a series of droplets of about 0.3-,ul volume of the copper-zinc

solution upon a microscope slide and evaporate them just to dryness.
Reserve the slide with the residues.
Transfer about 0.3,ul mercury test solution (10 mg Hg per ml) to a
slide, add a like volume of 6-F nitric acid, and evaporate just to dryness
without heating much above 100° C. After cooling, add from a capillary
pipet 0.5.al thiocyanate solution, and make certain that the whole residue
is dissolved. Take up the solution with the capillary pipet and cover
with it one of the copper-zinc residues without touching either the residue
or the slide with the point of the pipet. Observe the test with suitable
magnification and strong transmitted light, top lens of condenser inserted.
Compare with the observations made in Expt. 37. The bundles of charac-
teristic needles of purplish hue are replaced by tiny grains if the amount
of mercury is small, and daylight illumination becomes essential for the
observation of the color of the precipitate.

Experiment 40
Magnesium-Ammonium Arsenate and Silver Arsenate (1l5, 152, 168)
Magnesium acetate tetrahydrate, granular; or calcium chloride hexahydrate.
Transfer 0.5,ul arsenate test solution (10 mg As per ml) to the center
of a microscope slide, and add 2,ul 2-F HN03. If the drop should have
spread out too much, take it up into a capillary pipet and transfer it to
the center of another slide. Invert the slide and place it upon the opening
of a bottle containing strong ammonia so that the hanging test drop is
exposed to the ammonia gas. It requires only a few minutes to render
the test drop ammoniacal. Then remove the slide from the bottle, and
place a grain of 0.6-mm diameter (O.I-,ul volume) of magnesium acetate
or calcium chloride into the edge of the test drop. Observe with a total
magnification of 70 to 100 diameters and transmitted light of low intensity.
Prepare a sketch of the general appearance of the test, and include drawings
of characteristic forms such as X shapes, prismatic crystals, and feathery
dendrites of the NH4(Mg or Ca)As04 · 6 H 2 0.
The crystals of the corresponding phosphate have the same shape.
To confirm the presence of arsenate, set the test aside and allow it to
evaporate to dryness. Then place a large drop of I-F NH3 upon the residue
so as to cover it completely.
If the surface of the slide does not repel the solution strongly, drag
off the wash liquid as described by BEHRENS (ll5, ll6, ll8). With the
slide held horizontally, touch the point of a glass needle (stirring rod of
0.5- to I-mm diameter) to the edge of the drop and slowly draw it over
the slide away from the drop so that the liquid follows the point and forms
a narrow channel. The crystals of the test form cling to the slide and
10·
148 Techniques of the Submilligram Scale Expt.41

remain in place if the motion of the liquid is slow. While regulating the
rate of flow by suitably inclining the slide, start a circular motion with
the point of the needle in the channel to widen it to a pool about 1 cm
from the test drop. When most of the wash liquid has been collected
in this pool, remove it by taking the drop up with filter paper. Inspect
the area of the test drop under the microscope to make certain that the
crystals of the test form have not been disturbed. Then add another large
drop of I-F NH a, and remove the second wash liquid like the first.
If the liquid is so strongly repelled by the surface of the slide that itis impossible
to draw a channel, proceed as follows. Cut strongly absorbing filter paper (drop
test paper) into squ~res of l-cm edge. While inclining the slide so that the liquid
must flow upward to reach the paper, insert the corner of a square into the edge
of the drop, Fig. 43a. Observe the precipitate and regulate the flow by proper
inclining of the slide so that the crystals remain in place. When most of the wash
liquid has been taken up by the paper, treat the moist residue with the second
drop of dilute ammonia, which is then removed in like manner.
Finally, set the slide aside until wash liquid left with the precipitate
has completely evaporated. Then focus crystals to be tested with low
magnification, 20 to 80 diameters depending upon the size of the crystals.
Using reflected light and a green or blue background, treat individual
crystals with silver test solution; add the latter by means of a platinum
hook or platinum loop while observing through the eyepiece. The crystals
become opaque and change their color, but they usually retain their shape.
The resulting pseudomorphs of AgaAs04 are reddish brown; Ag aP0 4 is
yellow.
If desired, repeat the experiment with phosphate test solution and
magnesium acetate.

Experiment 41
Rubidium Chlorostannate (125, 168)
Rubidium chloride, granular.
Dilute 1 volume of stannic test solution (10 mg Sn per ml) with 4 volumes
3-F HOI, and transfer 1 to 2,ul of the mixture to a microscope slide. The
diluting may be done by taking into a capillary pipet 1 loop (0.3 to 0.5,ul)
of test solution and 4 loops (l.2 to 2,ul) HOI, taken from a drop on a slide,
and finally emptying the pipet upon a slide.
Introduce a grain of RbOl of 0.6-mm diameter (O.I-,ul volume) into
the edge of the drop, and observe with transmitted light and a magnification
of first 20 and later about 80 diameters.
A fine powder separates close to the reagent. Isometric crystals,
predominantly octahedra and tetrahedra, form outside the area of rapid
precipitation or, if only little tin is present, after some time near the edge
of the drop. Because of lesser solubility, the corresponding cesium com-
Expt.42 Slide Tests 149

pound affords a more sensitive test. The crystals of Cs 2 SnCI 6 , however,

are smaller and interferences are more to be feared since cesium has a
greater tendency than rubidium to form insoluble compounds.

Experiment 42
Estimation of the Relative Quantities of the Metals in a Slurry Containing
Arsenic, Antimony, Copper, and Silver
Slurry in a microcone. The slurry is prepared by placing into the cone first
a measured volume (I to 10 ,ul) of silver test solution (10 mg Ag per ml) and
adding 20,u1 of a solution obtained by taking measured volumes (I to 10 ml)
of arsenic, antimony, and copper stock solution (50 mg metal per ml), adding
25 ml 12-F HCI, and diluting with water to 100 ml.
Centrifuge; 6 microcones; 2 straight-tip cones with 2-mm bore; equipment
for working with micro cones including a heating block of the type shown in
Fig. 28, distilling apparatus, Fig. 33, and supply of dry air. - Reagents used
in Expts.32, 35, 37, and 40; supply of hydrogen sulfide; ammonium sulfide,
7-F: saturate 14-F NH3 with H 2 S, and then add a like volume of the NH 3;
ammonium chloride, 3-F; ammonium acetate, 6-F; hydrobromic acid, 9-F; KI,
KCNS, and KBr0 3, granular.
Treat the slurry in the microcone with H 2 S and separate the copper
and arsenic groups as directed in P. 64 to 66; consider that the volumes
given in these directions must be divided by one thousand when working
on the milligram scale.
The hydrogen sulfide precipitate will probably contain a total of 200 to
600,ug of metals, and it might be compared with a sulfide precipitate
obtained from 300,ug Hg. Reject supernate 2, P. 64.
Dissolve the copper group, residue 3, in 20 ftl 3-F HN0 3, p. 401, separate
the clear solution from any sulfur, and then treat it with l-,ul portions
of 12-F HCI until all silver is precipitated. Use the volume of the AgCI
for estimation of quantity. Lift off the supernate, treat it with an excess
of NH 3, and estimate the amount of copper from the intensity of the blue
color of the ammonia complex.
Treat the washed sulfides of the arsenic group according to P. 68, p. 410,
"Dissolution of Sample" and "Distillation of Extract AI". Treat the
distillate with l-,ul portions of 3-F NH 2 0H· HCI until it is colorless and
then with H 2 S. Use the yellow sulfide for estimation of the quantity of
arsenic. Treat the residue of the distillation with l,ul 3-F NH 2 0H . HCI,
20,u1 H 2 0, and finally with H 2 S. Use the orange sulfide to estimate the
quantity of antimony. Confirm the presence of Ag, Cu, As, and Sh.
The estimation of quantity is based upon the experience that like
amounts of a substance give precipitates of nearly the same volume,
provided that the precipitation is carried out with the same reagent under
comparable conditions and the precipitates are collected and compacted
by the same centrifugal force acting for the same length of time. In practice,
150 Techniques of the Submilligram Scale

solutions of known content may have to be precipitated after the precipitate

has been obtained in the analysis of the unknown substance. It is possible,
however, to postpone the whirling of the precipitate until the precipitates
of known amounts have been prepared so that all precipitates to be compared
may be centrifuged simultaneously. Experience (143) has also shown that
like amounts of different metals of the hydrogen sulfide group give sulfide
precipitates of nearly the same volume; it is thus possible to estimate
the sum total of metals in a mixture of sulfides by comparing the volume
with that of some sulfide precipitate having a known metal content and
to crudely estimate the quantity of one metal by comparing the volume
of its sulfide precipitate with that of the sulfide precipitate of another
metal. It may be expected that a like reasoning may be applied to hydroxides,
sulfates, silver halides, etc. Simplifying assumptions of this kind will
be frequently permissible since crude sedimetric comparison will not give
a relative precision better than ± 0.1 of the estimated quantity, which
is more than sufficient, however, for distinguishing between majors,
minors, and traces.
Estimation of quantity by colorimetric comparison with standards
may be practiced to advantage whenever a stable coloration is obtained.
Titrimetry, advocated by SWIFT (53), could be performed with the use
of simple capillary pipets and a millimeter rule; the meniscus at the fine
tip will serve for stopcock (90, 888).

Working Upon the Surface of a Slide

The performance of chemical separations upon a microscope slide was
introduced by BEHRENS (115, 116) to extend the usefulness of the slide
tests by the removal of interfering substances. He succeeded in performing
complete qualitative analyses with the material under investigation being
continuously upon a slide. His use of decantation by "dragging off" or
"drawing off", described in Expt. 39, in place of filtration requires nicety
of judgement based upon wide experience and may have been responsible
for the fact that the technique never found general favor and that a search
for more efficient techniques started about 1900. At this time it may
be stated, however, that the technique of BEHRENS is still useful in its
original form and in a great variety of modifications. The principle of
working with the material under investigation clinging to a flat or slightly
curved surface is used in some of the most sensitive procedures (944).
The limits of identification are related to the areas actually occupied by
the matter under investigation; the smaller the areas used, the better
is the sensitivity.
The decantation technique of BEHRENS has been modified in many
ways (118, 1070) and it may be greatly simplified whenever either only
Expt.43 Working Upon the Surface of a Slide 151

the liquid or only the solid is to be collected. Whether or not a clear filtrate
is obtained depends upon the extent to which the solid adheres to the slide.
Any form of decantation is impossible, of course, if the solid does not
settle. On the other hand, solid matter clinging tenaciously to the slide
may be washed by simply running solvent (water from the wash bottle
or faucet) over the slide. Aside from these extremes, most solids divide
themselves, part going into the decanted liquid and the washings and
some remaining on the slide. BEHRENS found, however, that many solids
can be made to behave by evaporating the mixture of solid and liquid
to dryness; when afterwards the residue is extracted, the insoluble solid
adheres to the slide in a degree which makes it possible to obtain a clean
separation by decantation.
A variation of the technique has been used by KOFLER (159, 607)
for the purification of organic substances by heating and the removal
of the eutectic melt. The melting point of the eutectic mixture is determined
by a preliminary experiment, in which a small amount of the substance
is heated on the hot stage under the microscope. Then, a square of l-cm
edge of hardened filter paper (S. & S. No. 576 which is replacing No. 575
used by KOFLER) is put upon a slide resting on the hot stage. A thin layer
of the substance is spread upon the paper, and a second slide is layed on
top of it. The stage is heated until the eutectic temperature is exceeded
by a few degrees, whereupon a downward pressure is put on the top slide
(with the raser end of a pencil). The melt is taken up by the paper which
becomes translucent in places. Usually, the solid residue sticks to the
top slide, and the paper square may be replaced by a fresh one after lifting
the top slide. The paper square is changed several times, and after each
change, the temperature is slightly raised. In this manner, a satisfactorily
pure substance may be obtained in 10 to 15 minutes.
Filterpaper may be used up to 300 C if each square is heated for a
0

short time only, which will suffice for efficient work. As an alternative,
one may heat the substance upon small squares, 18 mm X 18 mm X 1.5 mm
thick, of porous clay (605, 890). See also Expt. 54.

Experiment 43
Conversion of Silver Chloride to Silver Dichromate (152, 154)
Handling Precipitates and Solutions, Fusion, and Electrolytic Reduction
Magnesium ribbon; potassium dichromate, granular; ammonium chromate,
2 g in 100 ml water.
Obtain narrow slides, about 8 mm by 75 mm, by cutting standard slides
parallel to the long edge. If a glass cutter is not available, try the points of a
triangular file. With most cutting tools, it is necessary to find by trial the position
that gives a barely visible trace with a humming rather than scratching sound
when going with a smooth motion and moderate pressure across the surface.
152 Techniques of the Submilligram Scale Expt.43

If the trace is properly drawn, the slide will readily break along the desired
line when it is lightly tapped along it with the metal tool. Use the file to blunt
the resulting sharp edges and corners.
The silver chloride is collected, washed, dried, and fused; the solidified
melt is reduced with metallic magnesium. The silver sponge is washed
free of chloride and then dissolved in nitric acid for testing with chromate.
From a measuring pipet, deposit 0.05 ml of silver test solution near
the end of a narrow slide. Place a large drop of 12-F HCI upon another
(standard) slide, and transfer small portions of this acid with a platinum
loop to the drop of silver test solution until all silver is precipitated as
AgCI, i. e., until adding another portion of acid no longer causes further

e===6
/ 7'
c

~~,.,----"'?
~=<!!.::>=="'l/
6, =======i!V
d
Fig. 43. Working upon the Slide ; 1/2 n a t. size

precipitation. It is understood that the mixture is stirred with the loop

after each addition of acid.
Removal of Solution and Washing of Precipitate. Use filter paper or
drop test paper cut into squares of about 2-cm edge. Remove the major
part of the liquid as described in Expt. 40 and shown by Fig. 43a. Then,
with the dry edge of a fresh piece of paper, scrape the precipitate together
into a compact little tablet, Fig. 43b. Whenever the edge of the paper
becomes moist and soft, take another part of the edge, which is still dry
and stiff.
For washing, place a large drop of water upon the slide near the pre-
cipitate, Fig. 43c. Then tilt the slide so that the drop flows over the
precipitate. Do not stir, but allow to stand for a minute; then remove
the wash liquid with filter paper and dry out the tablet of precipitate as
before. Repeat the washing once, and do it so that finally the precipitate
is collected into a small tablet near the end of the slide. The technique
of collecting the precipitate is simple, but rather wasteful. Obviously,
it is not suited for the collection of finely powdered solids.
Drying and Fusing. Hold the slide with the precipitate a few centimeters
above a microflame, and perform a slight lateral oscillatory motion which
ensures the heating of the entire width of the slide and prevents its cracking.
Expt.44 Working Upon the Surface of a Slide 153

When the precipitate appears dry, lower the slide until it nearly touches
the microflame. Heat until a clear drop of fused AgCl is obtained (m. pt.
450 0 C). Then place the slide upon an asbestos board or upon wire gauze,
and allow it to cool to room temperature.
Reduction to Metallic Silver. The fused AgCl adheres tenaciously to
the slide, which simplifies the following operation. Place a large drop of
I-F acetic acid upon the AgCl. Cut to a point one end of a magnesium
ribbon of 3-cm length; bend the ribbon into the form of a Z, and place
it upon the slide so that the pointed end touches the AgCl, Fig. 43d. The
liberation of hydrogen starts immediately. Remove the magnesium ribbon
when the reduction has become complete, which is usually indicated by
the phenomenon that the spongy metallic silver does no longer adhere
to the slide and floates to the surface of the drop.
Removal of Solution and Washing the Silver. The metallic silver adheres
to paper, but there is no reason why it should adhere to the point of a
pipet rather than to the surface of the slide. Consequently, remove the
solution and the washings by means of a capillary pipet, but otherwise
use the procedure of decanting as when working with the squares of paper.
The tip of the pipet is too narrow to permit passage of the rather coarse
silver particles. Regulate the intake of liquid by properly inclining the
capillary pipet and the slide. For the practically complete removal of
chloride, wash twice with water; use a large drop each time. Use the same
capillary pipet for the removal of all liquids, and reject filtrate and washings.
Dissolving the Metal and Adjusting the Acidity. Warm the slide over
a small Bunsen flame until it is completely dry. Then add to the silver
from a capillary pipet 2,ttl 16-F HNO a, and warm the mixture to start
the reaction of the acid with the metal. The dissolution requires seconds
only. Warm the slide and blow upon the droplet until the solution is
evaporated just to dryness. Allow the slide to cool to room temperature.
Confirming the Presence of Silver. Dissolve the residue of AgNO a in
5 to 10,ttl of distilled water. Take up the solution into the capillary pipet
in which the water has been measured; transfer half of the solution to
a (standard) microscope slide, and use the rest for the spot test of Expt. 25.
Warm the slide so that the drop of silver solution evaporates. After cooling
to room temperature, dissolve the residue in 2.5 to 5,tt1 2-F HNO a and
test by adding K 2Cr 20 7 , Expt. 31.

Experiment 44
Separation of Bismuth and Lead (1143)
Evaporation, Extraction of "Invisible" Residues
Bi: Pb = 1: 100000; Pb: Bi = 1: 100 or better.
Phillips beaker, 250.ml; watch glass, 9 cm. - Solution containing 0.1 mg Bi
and 1 mg Pb per milliliter: dilute 1 ml of Bi·Pb test solution (5 mg Bi and
154 Techniques of the Submilligram Scale Expt. 44

50 mg Pb per ml) with 3-F HN03 to 50 ml , and mix well. - Potassium iodide
paper; quinine iodide reagent ; stannite reagent; triple nitrite reagent; potassium
cobalticyanide (i. e. the reagents used in Expts. 24, 26, 34, and 36).
Bismuth and lead are completely separated by repeated evaporation
of their' nitrates with water. Bismuth is hydrolized to the water-insoluble
basic nitrate BiO· NO a• Lead nitrate is extracted from the final residue
with water, dilute ammonium acetate, or ammonium nitrate solution (750,
1250).
Take l,ul of solution containing 0.1 mg Bi and 1 mg Pb per milliliter
if slide tests are to be used for final confirmation; take 5 ,ul, if spot tests
shall be used. Measure the volume in a calibrated capillary pipet, and

Fig. 44. Evaporation up on the Steam Bath; 1/2 n at . size

transfer the solution to the center of a clean watch glass so that the residue
will occupy a very small area; proceed as described below.
Evaporation Upon the Steam Bath. Place into the Phillips beaker,
Fig. 44, a few granules of metallic zinc and about 100 ml water. Heat on
the wire gauze to even boiling. Place the watch glass upon the beaker
so that the steam escapes through the lip, and blow filtered air from a
capillary toward the center of the watch glass to hasten evaporation.
Deposit upon the center of the hot watch glass about 0.5,ttl solution
at one time, and allow to evaporate to dryness before adding the next
portion to the residue. Continue until all of the solution has been evaporated
in a small spot. The residue clings to the glass, and caution may be relaxed
during the following evaporations with water.
Without removing the watch glass from the steam bath, add 10 to
20,u1 water to the evaporation residue without touching the latter with
the tip of the pipet. As a rule, the whole residue will be covered by the
large drop of water; if this should not happen by itself, spread the drop
with a glass thread. When the first drop has completely evaporated,
repeat the treatment four more times. This requires only a few minutes.
Expt.44 Working Upon the Surface of a Slide 155

Finally, remove the watch glass from the steam bath, wipe its underside
dry, and allow it to acquire room temperature.
Extraction of Residue. Inspection will show a relatively copious residue
upon the watch glass, whereas the total quantity of lead and bismuth
salts is less than 2 fig. The residue consists mainly of substances extracted
from the glass apparatus and of impurities in the reagents. Exclusive
use of vitreous silica and platinum in the experiment and in the preparation
and storing of the reagents would be required to reduce the amount of
residue to the theoretical size.
Whether or not a residue is visible, treat the area which was originally
occupied by the test drop with 5 to 10 fil distilled water. If necessary,
spread the water with a glass thread, but do not loosen the residue. After
standing for 3 minutes, take up the clear extract into a capillary pipet.

Fig. 45. Evaporation upon the Slide; 1/2 nat. size

Fractional Evaporation. To obtain the residue in as small an area as

possible, transfer 0.5-fil portions of the extract to a clean microscope slide
and, each time, evaporate to dryness before adding the next portion to
the residue. For evaporating, heat the slide over a microflame, Fig. 45,
while moving the slide in a horizontal plane so that the drop performes a
circular motion around the point of the flame. This causes the slide to
be heated in a circular area around the drop, and creeping of the drop is
prevented. Remove the slide from the flame when the drop begins to
vanish. Complete the evaporation by blowing upon the slide, and then
add the next fraction while the slide is still hot. Continue in this way
until the extract has been used up.
Repeat the extraction of the residue upon the watch glass with two
more 5- to 10-fil portions of distilled water. Using always the same capillary
pipet, concentrate all extracts in the small area upon the microscope slide.
Confirmation of Lead. To perform the triple nitrite test, proceed with
the evaporation residue upon the slide as directed in Expt. 34. If a spot
test shall be used, dissolve this residue in 1 fil water, take up the solution
into a capillary pipet, and transfer it to potassium iodide paper, Expt. 24.
Dissolution of Basic Salt and Test for Bismuth. Treat the residue upon
the watch glass with 10 fil 6-F HN0 3 • Use a glass thread to spread the acid
156 Techniques of the Submilligram Scale Expt.45

over the whole area treated during the extractions with water. Finally,
collect the solution in a capillary pipet, and transfer it in 0.5,a1 fractions
to a slide for evaporation in a small area. Evaporate just to dryness and,
after cooling to room temperature, cover the small residue with 0.5,a1 or
less 2-F HN0 3 • To perform a slide test, without delay add a grain of
potassium cobalticyanide of 0.5-mm diameter, Expt. 36. Otherwise take
up the solution into a capillary pipet and proceed according to Expt. 26.

Experiment 45
Separation of Silver, Lead, and Mercurous Mercury (168), Sublimation,
Extraction with Boiling Solutions
Narrow slides. - Solution containing 10 mg each of silver, lead, and mercurous
ions per milliliter solution: mix 2 ml each of the corresponding stock solutions
(50 mg ion per ml) and dilute with I·F HN0 3 to 10 ml. - Potassium iodide paper;
nitrite reagent; potassium chromate, 2% solution; sodium chloride, 1% solution;
potassium cyanide, granular.
Silver, lead, and mercurous mercury are precipitated as the chlorides
and washed to remove the nitrate ion. Mercurous chloride is isolated by
sublimation, and lead chloride is extracted from the residue with boiling
dilute acid. Boiling water is not suited since it does not dissolve basic
chloride which sometimes forms during the sublimation.
Precipitation of the Chlorides. With a measuring pipet transfer about
0.1 ml of silver-lead-mercurous solution (10 mg of each ion per ml) to the
end of a narrow slide. Deposit 20,a1 12-F HCI close to the test drop and,
with a glass thread, combine the two drops and stir until the white precipitate
flocculates.
Remove the solution with filter paper as in Expt. 43. Wash the pre-
cipitate with two 0.02-ml portions of 2-F HCI, and collect it as a small
tablet near the end of the slide. Reject filtrate and washings.
Sublimation from Slide to Slide. Hold the narrow slide with the washed
precipitate in the left hand and a microscope slide of standard dimensions
in the right. Rest the latter on the edge of the former so that the two
form an angle of 5 to 10 degrees and enclose a narrow wedge-shaped air
space, Fig. 46a. Briefly heat the precipitate by holding it about 2 cm
above the point of a microflame. Every few seconds, remove the narrow
slide from the flame and, along the edge, move the microscope slide forward
over the precipitate and hold it there until the amount of condensate does
no longer increase, Fig. 46b. It should be understood that the angle between
the slides and the wedge-shaped air space are maintained when moving
the upper slide back-and-forth. When condensation ceases, move the
upper slide back into its former position, Fig. 46a, and again heat the
precipitate for a short time.
Expt.45 Working Upon the Surface of a Slide 157

A quickly evaporating condensate of water will be obtained in the first

two or three tests. When no more water condenses, assume that the residue
has become dry and begin the sublimation. For the brief periods of heating,
decrease the distance between precipitate and tip of flame to 5 mm. The
vapor usually cannot be seen. Thus, test at intervals of a few seconds if
vapor is given off by moving the large slide over the heated substance.
The white sublimate fills a circular area above the heated precipitate. When
condensation ceases, draw the top slide back and heat the precipitate some
more, etc.
Collect the whole sublimate in the same spot upon the microscope slide
and in as small an area as possible. Finally prove the completeness of the
separation by further heating of the precipitate and then holding over it

Fig. 46. Sublimation from Slide to Slide; about 1/2 nat. size

a fresh portion of the microscope slide. If no sublimate is obtained, assume

that all Hg 2Cl 2 has been removed from the precipitate.
A temperature of 300 C suffices for the volatilization of Hg 2C1 2 •
0

Excessive heating will cause the residue to melt and will render difficult
the subsequent extraction of PbCI 2 •
Identification of Mercurous Chloride. Place the slide upon a sheet of
white paper, and use a platinum loop to place a droplet of conc. NH3
upon part of the sublimate. The moistened portion turns black. This
test will not succeed if the HN0 3 is not sufficiently removed and is able
to oxidize Hg 2CI 2 to HgCl 2 when the precipitate is heated for sublimation.
Extraction with Hot Solvent. The heated precipitate clings to the
glass, and this makes it possible to use the following simple technique.
Treat the residue upon the narrow slide with 0.05 ml 2-F HCI. Hold ready
in the right hand a capillary pipet with relatively wide tip for the quick
removal of the hot extract. Grasp the narrow slide with the left hand
and hold the residue with the drop of acid about 3 cm above the tip of
the microflame. When steam bubbles just start to appear in the drop,
remove from the flame and quickly insert the tip of the capillary pipet
so that the whole hot solution is quickly taken into the latter.
158 Techniques of the Submilligram Scale Expt.46

Identification 01 Lead. Evaporate the HOI extract in small portions

on a slide to obtain the residue in a small area, Expt. 44. Perform the
triple nitrite test directly with the residue, Expt. 34, or dissolve the residue
by heating close to boiling with about 20,tt1 water; take the hot solution
into a capillary pipet, and use it for the potassium iodide test, Expt.33.
Identification of Silver. From a capillary pipet, add 5,tt1 12-F NHs
to the residue from the extraction on the narrow slide. Without delay.
stir with a glass thread, and then take the clear solution back into the
capillary pipet and transfer it to a clean slide. Oover the test drop and
continue as outlined in Expt. 32 or (and) 43.
To obtain a spot test, dry the AgOI residue upon the narrow slide by
warming over a flame, and then heat until the salt melts. After cooling.
add a grain of KON about twice the volume of the AgOI, and heat over a
small Bunsen flame until the KON melts and the reduction of the AgOI
is complete. The metallic silver adheres firmly to the slide. Wash it free
from cyanide and chloride by running distilled water from the wash bottle
over the slightly inclined slide. After drying by holding the slide over a
flame and cooling to room temperature, dissolve the silver by warming
it lightly with 5,tt1 16-F HNO a• Evaporate just to dryness. and dissolve
the residue in 5 to 1O,tt1 water. Use the solution for the test of Expt. 25.

Experiment 46
Test for Ammonium Ion, Use 0/ the Gas Reaction Cell (149)
L. I .• O.l,ttg NH 4 •
Glass ring, 25-mm outer diameter, 5 mm high, polished at top and bottom.
For substitutes may serve: crucible of 0.5- to I-ml capacity; bottom part of
wide vial, cut 5 mm above the bottom and polished flat; screw cap of plastic. -
Sealed capillary, 5-mm long, containing iron wire, p. 97; strong permanent
magnet (Alnico or Cunife). - Chloroplatinic acid, 5%: keep in vial of plastic
or of vitreous silica and test from time to time by allowing 5,tt1 of it to evaporate
at room temperature in a desiccator for the microscopical inspection of the residue.
The test substance is treated with sodium hydroxide. The liberated
ammonia is absorbed by a solution of chloroplatinic acid and precipitates
ammonium chloroplatinate. The test is performed in a completely closed
gas reaction cell which is assembled from two standard microscope slides
and the glass ring.
Assemble the cell of Fig. 47 and use a strip of writing paper (width
nearly equal the height of the ring, and length nearly equal to the diameter
of the ring), folded to give an angle of 90 degrees, to divide the chamber
into two compartments. Cellophane may be used in place of paper, or
the paper may be impregnated with paraffin wax if one wishes to exclude
adsorption on the paper.
 Working in Capillaries 159

Deposit in the smaller compartment on the floor of the cell l,ul of

ammonium test solution (10 mg NH4 per ml) which has been diluted with
nine volumes of water and next to it l,ul 6-F NaOH. Place the capillary
with the iron core into the same compartment. Deposit l,ul H 2PtCl 6
solution upon the top slide so that it hangs on the ceiling of the larger
compartment of the cell. Place the assembled cell upon the stage of the
microscope and focus through the top slide upon the drop of chloroplatinate.
Observe for a period of 5 minutes and ascertain that the drop remains clear.
Without opening it, remove the cell from the stage, and by applying
the magnet from below the bottom slide, wipe together the droplet of NaOH
and the test drop. Return the cell to the stage of the microscope and again
focus upon the drop of chloroplatinate. Use a magnification of 30 to

Fig. 47. Cell for Gas Reactions; about nat. size

100 diameters for observing the yellow tetrahedra and octahedra of

(NH4)2PtCl6. If a large number of very small crystals is formed, exchange
the top slide for another one with a fresh droplet of reagent; repeat this
until the NH3 concentration in the chamber has become low enough to
make possible the slow separation of large crystals.
Instead of chloroplatinate, one may use tiny fragments of test papers,
moistened and pasted to the ceiling of the cell. Suitable are red litmus
paper, pH test paper (Hydrion paper for the range 7 to 10), and filter paper
dipped into Nessler reagent. The limit of identification will improve as
the size of the test papers decreases; squares or triangles of 1-mm edge
are convenient, but a few fibers suffice if the color may be observed with
magnifiers or microscope.

Working in Capillaries
The technique of working in glass capillaries of 0.3- to I-mm bore
was developed by EMICH and his coworkers (154, 1007, 1010, 1016, 1139,
1141). It may be used for the preparation of derivatives of organic substances
on a milligram scale; in inorganic analysis, it is advantageous for the
performance of certain tests with about I to 100,ug of solid material and
a few microliters of solvent. In addition, it is often useful as an adjunct
160 Techniques of the Submilligram Scale Expt.47

to the technique of working upon the microscope slide and in the microcone.
The relatively large wall surface of capillaries several centimeters long,
however, presents disadvantages in the performance of a sequence of
separations and, for this reason, lengthy analytical procedures have never
been performed in capillaries of these dimensions.
In the performance of confirmatory tests on the sub milligram scale,
the capillary takes the place of the test tube of the gram scale .

. Experiment 47
Lead Sulfate, Triple Nitrite, Lead Ohromate
The Oapillary as Adjunct to Working Upon the Micro8cope Slide
Nitrite reagent and 2% ammonium chromate solution as used in Expts. 25
and 34; ammonium acetate, solid.
Place 5,ttl of lead test solution (10 mg Pb per ml) upon a slide, and
deposit a large drop of 4-F H 2 S04 upon another. With a platinum lopp,
transfer portions of the acid to the drop of lead solution until the precipita-
tion is complete. Note that the PbS0 4 does not adhere to the slide and
that decantation upon the slide would be difficult to perform.
Transfer of Solid as a Slurry. Cut off squarely both ends of a capillary
of about 0.5-mm bore and lO-cm length. Close one of its ends with the
index finger, and with the other end stir the mixture upon the slide into
a slurry. When a rather uniform mixture has been obtained, incline slide
and capillary, and lift the index finger off the top of the latter so that
the slurry enters quickly into the tube.
Separating the Phases in the Capillary. By means of a microflame,
fuse the capillary into a bead at a point 3 cm from its dry end, Fig. 48a.
When the bead cools, contraction of the air inside causes the slurry to go
further into the capillary; Fig. 48b.
Cut off all except 1 cm of the empty capillary beyond the glass bead,
and place the capillary, bead downward, into a microcone. After swirling
in the centrifuge, cut the capillary at the boundary line, Fig. 48c, so that
one piece contains the clear solution and the other just the precipitate.
In this instance, reject the solution which is not needed.
Transferring the Precipitate to a Slide. If the triple nitrite test is to
be used, place upon a microscope slide 5,ttl of copper test solution (10 mg Cu
per ml). Grasp the empty part of the capillary with fingers or,forceps,
and dip the end containing the PbS0 4 into the copper solution. Without
lifting it out of the drop, repeatedly tap the end of the capillary against
the slide. The PbS0 4 usually leaves the capillary as one lump, Fig. 48d,
if the bore of the tube tapers bluntly at the bead and care is taken that
no air bubble forms in the capillary. Sometimes it is necessary to loosen
the precipitate in the capillary by stirring with a fine glass thread. To.
Expt.48 Working in Capillaries 161

G(Wt1plete the experiment, evaporate the mixture of lead sulfate and copper
solution to dryness, and treat the residue with nitrite reagent as directed
in Expt. 34. Note that the slow dissolution of the PbS0 4 obviously favors
the separation of relatively large crystals of the triple nitrite.
If the experiment shall be concluded with a spot test, place 5,ttl water
upon a slide, and transfer the precipitate from the capillary to this drop
as outlined above. Treat the mixture of PbS0 4 and water with a crystal
of ammonium acetate about 1.5 mm in diameter (l,ttl volume), and aid
the dissolution of the PbS0 4 by stirring with a glass thread. Add to the
clear solution l,ttl 6-F acetic acid and take it into a capillary pipet for
performance of a chromate test as described for silver in Expt. 25.

c==z==,,~,,'~
' I================~.-==== ~

."
•
==============C=====\
~
' .8C== c
Cvt

Fig. 48. Separation of Solution and Precipitate; about nat. size

Experiment 48
Recrystallization of Lead Iodide (154)
Place upon a slide so that they do not mix one large drop each of lead
test solution, iodide test solution (10 mg ion per ml), and 2-F acetic acid.
With a capillary pipet of 0.5-mm bore in the wide part, take up portions
of the three solutions in the following order: first a 5-mm length of lead
solution, then a 6-mm length of iodide solution, and finally a 20-mm length
of acetic acid.
Sealing Capillaries. Use a microflame, pilot flame, or the edge of a
Bunsen flame. About 6 cm from the drop, draw out a fine capillary,
Fig. 49a, but do this without removing the tube from the flame so that
this fine capillary is fused shut, Fig. 49b, while it is drawn. When the sealed
end cools, the solution is drawn into the tube, Fig. 49c. Break the now
empty tip close to the taper, and seal the opening by touching it to the
edge of the flame, Fig. 49d. Inspect this second seal under the microscope.
Mixing. Cut open the end of the capillary, that has been sealed first,
and thoroughly mix the contents with a fine glass thread with a bead
Benedetti-Pichler, Identifica tion 11
162 Techniques of the Submilligram Scale Expt.49

at the end. Again draw out and seal the capillary near its opening. Repeated
opening and sealing a capillary will not shorten it much if, before sealing,
a thin glass rod is fused to the opening of the capillary to serve for a handle .
Heating Liquids in Capillaries. Place the capillary into a test tube,
and add water to submerse the whole capillary. Heat the water in the test

==~==~============~====~==~

==C===~==========~~ F.~=== b

==~====~==============~~ c

ec::::Ii ==:::::r========::::;:, d
Fig. 49. Sealing a Capillary. The bore is exaggerated

tube and keep it boiling until all of the PbI 2 in the capillary has dissolved.
Then set the test tube with its contents aside, and allow to cool slowly
to room temperature.
Examination of the Contents of a Capillary. Pour the water out of the
test tube. Then remove the capillary, place it upon a slide, and examine

(1

Fig. 50. Cell for Use in Examining the Content.s of a Capillary. The bore of the
capillary is exaggerated

with magnifying glass and under a microscope with both transmitted and
reflected light. Use low magnification and improve the optical conditions
by immersing the capillary in water. A suitable cell may be improvised
with two glass rods of 2-mm diameter and a cover slip, Fig. 50.
Finally, cut the capillary open at both ends, and blowout its contents
upon a slide for additional examination: measurement of the angles of
the hexagonal plates, testing in polarized light.

Experiment 49
Isolation of Metallic Mercury, Conversion to Iodide (1140)
rt _____ ......:M /) 1 ITHn it; " ,mp.t,p.r : iodine. powder; ammonium oxalate, saturated
Expt.49 Working in Capillaries 163

Metallic mercury is precipitated by inserting copper metal into the

solution which has been treated with ammonium oxalate to assure a
satisfactory deposit (651). The amalgamated copper is heated, and a
condensate of droplets of mercury is obtained. Exposure to iodine vapors
gives conversion of the metal to mercuric iodide.
Into a capillary pipet having 0.5-mm bore in the wide part, take up
l,ul mercuric test solution (10 mg per ml) and 3,u1 ammonium oxalate
solution. If a microscope is not available for the inspection of the test,
take ten times larger amounts. Seal the capillary as outlined in the preceding
experiment and illustrated by Fig. 49.
Cut open the end of the capillary, which was sealed first, and introduce
with forceps a 2-mm length of clean, bright copper wire. Place the capillary,
sealed end down, into a cone and centrifuge briefly; this collects wire and

•• ,p.

·e
~

Fig. 51. Distillation of Mercury. In c the bore of the capillary is greatly exaggerated

solution at the sealed end. Draw out and fuse shut the open end, and
heat the sealed capillary for at least 1 minute in a steam bath or a test tube
with boiling water. Finally, lay the capillary upon filter paper, and cut
it so that the copper wire becomes accessible.
Wash the wire upon the filter paper by adding a drop of water from
the wash bottle. When the water has been absorbed by the paper, pick
up the wire with clean forceps, and transfer it to dry filter paper. Remove
all moisture by gently pressing the wire between sheets of the paper.
Distillation of Mercury. The capillary must not contain any moisture.
Thus, prepare it from a freshly drawn capillary of 0.5-mm bore and 15-cm
length. To prevent flame gases from entering the tube, heat and fuse shut
the center portion. This gives two dry capillaries which are sealed at one end.
With forceps, introduce the amalgamated copper wire into one of them,
and bring it down to the sealed end by tapping with the capillary on the
bench top or by centrifuging, Fig. 51 a.
For the distillation, heat the sealed end of the capillary in the edge
of a non-luminous Bunsen flame so that the rest of the tube remains cool.
Continue heating until a bead of glass has completely enclosed the copper
metal, Fig. 51 b.
u*
164 Techniques of the Submilligram Scale Expt.50

Examination of Distillate. Place the capillary upon a microscope slide,

and use reflected light and a magnification of 20 to 50 diameters. The
mercury condenses in small drops near the heated end. Thus, first focus
upon the glass bead, and then search the whole length of the capillary.
The silvery color of the metal and the mirror action of the surface of the
drops are clearly visible with a black background. When the droplets
have been found, observe them also with transmitted light. They will
appear as black circular disks when the hands are cupped around the stage
to exclude light coming from above. The directions apply, in a general
way, when a magnifying glass is used for examination. Holding the capillary
in front of a lamp gives the effect of observation with transmitted light;
for observation with reflected light, lay the capillary on dull, black paper
or cloth.
Conversion to Iodide. Push one end of a capillary of about 0.3-mm
outer diameter into the iodine contained in the vial so that some of it gets
wedged into the bore of the fine tube. A very small amount does suffice (1272).
Wipe clean the outside of the capillary, and then insert it into the capillary
containing the distillate of mercury so that the iodine approaches the
droplets within a few millimeters, Fig. 51 c. Without delay, place the
combination upon a slide and observe with reflected light at intervals of
several minutes. Depending upon the temperature, the conversion to
yellow and orange HgI2 proceeds more or less slowly. The droplets become
coated with a film of the iodide. At times, the coating bursts and liquid
mercury is exuded to acquire a film of iodide that assumes the shape of
strings and sausages.

Experiment 50
Bettendortf's Test tor Arsenic (1009)
Stannous chloride reagent: dissolve 11.5 g SnOI 2· 2 H 20 in 17 ml 12-F HOI
and dilute with water to 100 ml; place some metallic tin into each bottle containing
this reagent_ - Isoamyl alcohol, b. pt_ 130 O. 0

Arsenate and arsenite are reduced in strongly acid solution to elemental

arsenic; selenium, tellurium, mercury, and the noble metals interfere since
they too are reduced to the elemental state.
Take up 1,u1 arsenic test solution (10 mg As per ml) into a capillary
pipet of 0.5-mm bore in the wide part. Deposit drops of stannous chloride
reagent and 12-F HCI upon a slide, and from there take into the capillary
pipet about 1,u1 SnCl2 reagent and 6,u1 HCI. Estimate these volumes
from the length of capillary filled by the 1-,ul drop of arsenic solution.
Take the HCI into the pipet last because it does not contain solid matter
and is, therefore, suited for rinsing the tip previous to sealing.
Seal both ends of the capillary pipet by the standard procedure described
in Expt.48 and illustrated by Fig. 49.
Expt.51 Working in Capillaries 165

Mixing. Mter inspecting the last seal under the microscope, mix the
contents of the capillary by centrifuging them two times from one end
of the capillary to the other.
Heating. Place the capillary into a clean, dry test tube, and add about
2 ml of amyl alcohol. Boil the amyl alcohol over a small Bunsen flame
for 3 minutes so that the ring of condensate remains about 5 cm below
the opening of the test tube while the capillary is entirely immersed in
the boiling liquid and its vapor. Mter heating, set the test tube and contents
aside for 2 minutes. Then pour the amyl alcohol back into the reagent
bottle, and remove the capillary from the test tube.
Examination of Precipitate. With the centrifuge, collect the precipitate
at one end of the capillary, preferably the one which tapers to a finer point.
Then place the capillary upon a slide, and inspect its contents with reflected
light before a white background and a magnification of 20 to 30 diameters.
Immerse the capillary in water to better the optical conditions, Fig. 50.
The arsenic separates either in the black or in the brown form. Save the
capillary with the precipitate for the next experiment.

Experiment 51
Oxidation of Arsenic to Arsenic Acid
Carius' Treatment in Capillaries (1141)
Goggles or safety shield. - Buffered silver solution as for Expt. 29 or
magnesium acetate tetrahydrate, granular, as for Expt. 40. - Capillary with
arsenic precipitate obtained in Expt.50.
The elemental arsenic is oxidized to HaAs04 by heating with concentrated
nitric acid in a sealed tube.
Separation of Precipitate and Solution. Cut open the empty end of
the capillary containing the arsenic precipitate. If necessary, whirl again
in the centrifuge to collect the arsenic as a compact plug in the point
of the tube. For the removal of the liquid, either use a suction-
operated siphon, Fig. 32a, with the intake arm drawn out to a capillary
of 0.2-mm outer diameter, or use a so-called "contraction pipet", Fig 52 (401).
The latter is quickly made from one of the pipets with elongated bulbs
obtained as a by-product of the drawing of capillaries from glass tubing.
For lifting off the liquid, insert the fine tip a into the capillary containing
the arsenic, but do not yet immerse the orifice of a into the liquid. Now,
heat the bulb c of the pipet with a small Bunsen flame to 200 or 300 C.
0 0

Remove the flame, and push the tip of the pipet into the liquid while the
bulb is still hot. The liquid is drawn into the pipet as the air in bulb c
cools. Advance the pipet into the capillary as the meniscus of the liquid
recedes until the orifice of the tip gets close to the precipitate. When air
begins to enter the tip of the pipet, withdraw the latter from the capillary
166 Techniques of the Submilligram Scale Expt.51

and heat bulb c again to expel the contents of band a. In this instance,
the centrifugate is rejected, but it may be received in another capillary,
in a microcone, or on a slide.
Since it cannot be avoided that a considerable amount of liquid remains
behind on the surface of the capillary, centrifuge the latter and remove
the collected liquid as before.
Washing Precipitates in Capillaries. Place a large drop of water upon
a slide. Heat the empty part of the capillary containing the arsenic by
drawing it through a Bunsen flame, and quickly insert the opening of the
capillary into the drop of water. The latter enters the capillary as the
glass cools. Transfer the water to the sealed end of the tube by brief
swirling in the centrifuge, and then mix thoroughly by means of a glass
thread with a small bead at the end. Centrifuge again, and remove the
wash liquid in the same manner as the centrifugate. Repeat the washing
once, and take special care to remove the second wash liquid as completely
as possible since it would dilute the HNO a added in the next step.

Fig. 52. Contraction Pipet. The bore of parts a and b is exaggerated

Oxidation According to Carins. Place a large drop of 16-F HNO a upon

a slide. Briefly heat the empty part of the capillary containing the arsenic,
and insert the opening of the hot capillary into the drop of HNO a• Allow
not more than 2,ul of the HNO a to enter the tube. Bring the acid to the
arsenic at the sealed end of the tube by whirling in the centrifuge, and
then seal the tube by drawing out just below the opening and fusing shut,
Fig. 49. Inspect the seal under the microscope.
For heating, place the sealed capillary into a clean, dry test tube of
Pyrex glass. To avoid serious consequences in the event of an explosion,
use goggles or a safety shield and keep the test tube pointed toward a
wall as long as the capillary remains sealed. Hold the test tube in a nearly
horizontal position. Heat the part containing the capillary by moving
the test tube backwards and forwards over a non-luminous Bunsen flame,
2 cm high, so that the entire length of the capillary is uniformly warmed.
Do not heat too strongly; the temperature need not exceed 250 C and 0

the HNOs should not completely evaporate at any time. Uniform heating
keeps the HN0 3 at the end of the capillary and in contact with the
arsenic.
When the solid is no longer visible, allow the test tube to cool to room
temperature. Keep the eyes protected until the tube is open. First let
the capillary slide into a centrifuge cone, and whirl it to collect the contents
at one end. Then remove it from the cone, and cut it open in the middle.
Expt.52 Working in Capillaries 167

Testing for Arsenic Acid. Take the liquid contents of the capillary
into a contraction pipet, and transfer them to the center of a 9-cm watch
glass for evaporation on the steam bath.
To perform a slide test, dissolve the residue in 5,al 2-F HNOs. Transfer
the solution to a microscope slide and treat with magnesium acetate as
directed in Expt.40.
To get a spot test, dissolve the residue in 2,al 0.5-F HNOs. Take the
solution into a capillary pipet and proceed as in Expt. 29.

Experiment 52
Oonversion of Aniline to Acetanilide (1010)
Melting point apparatus consisting of beaker with bath liquid, stirrer, and
thermometer, 250 0 C; supply of air under low pressure. - Aniline, purified;
anisol; benzene, reagent grade; carbon; asbestos for Gooch crucibles.
Aniline is heated with acetic acid in a sealed tube. The resulting
acetanilide is washed with water and recrystallized from benzene until
the melting point remains constant. To conserve material, the procedure
is arranged so that a change of container is not required. Capillaries are
able to withstand high pressure; consequently, procedures involving the
handling of substances with high vapor tension are advantageously performed
in sealed capillaries, which simplifies heating of reaction mixtures and
working with boiling solvents so that the customarily used reflux condensers
are not needed. Explosions occur only rarely if unreasonably high temper-
atures are avoided. They are quite harmless since the exploding capillary,
as a rule, is unable to destroy the surrounding test tube.
Heating the Reaction Mixture. Draw out a fine tip at one end of a
capillary of 1- to 1.5-mm bore and 12-cm length. Take into the capillary
2 ,al aniline and 3 ,al glacial acetic acid; measure the volumes from a calibrated
capillary pipet into a microcone, and then take the mixture into the wide
capillary. Seal both ends of the capillary by the standard procedure,
Fig. 49, inspect the seals under the microscope, and then heat the capillary
15 minutes at 150 0 C. To this end, either place the capillary into a drying
oven, or use a bath of anisol (b. pt. 154 0 C) and the technique described
in Expt. 50; the test tube may be provided with a cork stopper carrying
a glass tube of 20-cm length, which will serve for reflux condenser. Finally,
remove the capillary from the bath and, if necessary, collect its contents
at one end by means of the centrifuge. Cut it open at the empty end.
Washing and Drying in the Capillary. From the bore of the capillary
compute the length of it, that holds 10,a1. Introduce 40,al water either
by means of a capillary pipet with long fine tip, or by drawing the empty
part of the tube through a Bunsen flame and quickly inserting the opening
into a drop of water. Whirl in the centrifuge to bring the water to the
168 Techniques of the Submilligram Scale Expt.52

sealed end. Stir and mix with a glass thread having a small bead at the
end. This causes crystallization of the acetanilide. Mix, centrifuge, and
remove the aqueous solution with a suction-operated siphon or a contraction
pipet, Expt. 51. Centrifuge a second time, and again remove the collected
liquid. Repeat the washing twice, and use each time 30.u1 water. Test
the removed wash liquids with pH test paper.
For drying the acetanilide in the capillary, clamp the finely drawn
out tube shown in Fig. 23 to the stand holding the melting point apparatus,
and connect it to a supply of clean air, p. 97. Place a wad of cotton into
the wide part, and bend the fine capillary about 4 cm from the orifice
through an angle of about 10 degrees, just sufficient to make it act as a
spring when it is inserted into the capillary containing the acetanilide.
Insert it far enough so that the orifice of the fine capillary is only a few
millimeters from the wet substance. Turn on the flow of air and insert
the wide capillary into the melting point bath by suitable adjustment
of the position of the fine capillary which holds the wider tube by its
spring action, Fig. 53a.
Slowly raise the temperature of the bath to 100 C and hold this
0

temperature for 5 minutes. Then shut off the flow of air, and raise the
temperature for the determination of the melting point which probably
will be found below 114 0 C.
Treatment with Carbon and Recrystallization. First, determine the
suitable amount of solvent. Introduce a volume of benzene about equal
to that of the acetanilide. Whirl in the centrifuge to get the solvent to
the substance, Fig. 53b. Then fuse a glass thread to the open end of the
capillary, draw out close to the opening, and seal at the same time, Fig. 53c.
Place the capillary into an empty test tube which serves as air bath
and as protective tube, compare Expt. 51. Hold the test tube in a nearly
horizontal position, and heat the part containing the capillary by moving
the tube backwards and forwards over a non-luminous Bunsen flame,
2 cm high, so that the entire length of the capillary is uniformly warmed.
If solvent distils and condenses away from the solution, it may be brought
back to the solution by a flip with the hand holding the test tube or by
more strongly heating that part of the capillary in which condensation
occurs. If the amount of solvent is correct, the acetanilide will completely
dissolve in the hot benzene and crystallize when the solution cools, which
may be hastened by sliding the capillary out of the test tube. Crystallization
may be delayed by supersaturation, but may be started by further cooling;
to this end, wrap the capillary in some cotton, and let ether drip on it.
If necessary, adjust the amount of benzene until testing shows that
recrystallization may be performed with success. If too much benzene
has been added, the excess may be evaporated by the technique used for
drying the substance.
Expt,02 Working in Capillaries 169

Finally, cut open the empty end of the capillary and introduce a small
amount of carbon, Fig. 53d. Seal again, and place the capillary with the
freshly sealed end first into the test tube. Heat to dissolve the acetanilide
and transfer the solution to the end containing the carbon. Do this by
either flipping or heating, and complete the transfer by means of the
centrifuge. In this manner, it is possible to treat the solution with adsorbent
without getting the latter all over the tube (1270).
Again cut open the empty end of the capillary, and heat with a micro-
flame 10 mm from the solution to obtain a constriction. With forceps,

~''''~!====================b

@
- c

~!"~C/!================== d

'C ' " -- e

Fig. 53. Preparation of Derivatives in the Capillary

introduce some freshly ignited asbestos fibers into the capillary and,
with a thin rod, push them against the constriction to form a filter mat
of not more than 1- to 2-mm length. To hold the mat in place, make a
constriction also on the other side of it, Fig. 53 e.
Seal the open end of the capillary, and place it with the frashly sealed
end last into the test tube. Heat the whole length of the capillary to get
complete dissolution and mixing of the solution with the coal. Use flipping
and suitable heating to keep all solvent with the solution in tha smaller
compartment of the tube. Finally, allow to cool, reval'se the position of
the capillary in the test tube, warm to assure complete dissolution, and
then heat mostly the sealed end of the smaller compartment to force
the hot solution through the filter. Collect the filtrate at the end of the
larger compartment by the jerky motion, the direction of which is indicated
by the arrow of Fig. 53/. (As an alternativa, the filtration may be brought
about by whirling in the centrifuge with the capillary immersed in a hot
bath liquid.)
170 Techniques of the Submilligram Scale Expt.53

Allow the capillary to cool, and then cut it open so that the filter is
removed, Fig. 53g. For separating the mother liquor from the crystals,
press the latter somewhat together by means of a glass rod of I-mm diameter,
centrifuge, and lift off the liquid. Repeat the pressing, centrifuging, and
decanting several times. Then dry the crystal mass as before by heating
while blowing air through the capillary, and determine the melting point.
Omit the adding of charcoal, but repeat the recr~ stallization until the
melting point remains constant, 115 C. 0

The removed mother liquors may be evaporated, and the residues

used for slide tests, etc.

Experiment 53
Conversion of Urea to Symmetrical Diphenyl Urea (152)
Heating block, 3 cm X3 cm X 5 cm long, with horizontal wells for a short
thermometer and for the bulb (25 mm deep, 6-mm bore); short thermometer;
melting point bath as in precediug experiment. - Urea; aniline, purified; nitro-
benzene (b. pt. 211 0 C).
The urea is heated with aniline in a sealed tube. Distillation and
sublimation under reduced pressure are used for the removal of unreacted
aniline and the isolation of the diphenyl urea. The latter is then purified
by recrystallization from ethanol.
Heating with Aniline. From glass tubing of 6-mm outer diameter,
draw out the tube h shown in Fig. 53. The bulb should be somewhat less
than 6 mm in diameter; the capillary, 1 to 1.5 mm in bore and 8 to 10 cm
long. Introduce 3 mg urea into the bulb 1, and, with a capillary pipet,
add 9,ttl aniline. Fuse the tube shut at 6, and inspect the seal under the
microscope. Place it into a test tube, add nitrobenzene, and boil the latter
for 15 minutes so that the capillary is completely imma.csed in the hot
vapor; compare Expt.50. Finally, allow to cool, return the warm nitro-
benzene to its storage bottle, remove the capillary from the test tube,
and wipe dry its outside. Cut open the seal at 6.
Distillation of Aniline. Insert the bulb 1 into the well of the heating
block so that not more than 20 mm of the capillary are inside the well.
Connect 5 to the vacuum line (water pump giving about 60 mm Hg),
and begin heating the block. When its temperatura reaches about 120 0 C,
droplets of aniline condens at 2; fan the capillary with a small Bunsen flame,
and drive the droplets into the wide part 5. When all aniline has been
driven off, cool at 3 by hanging a strip of filter paper over the tube and
keeping it wet with water.
Sublimation of Diphenyl Urea. Raise the temperature of the block.
The sublimate will start collecting when a temperature of approximately
180 0 C is reached. Hold the temperature until the amount of sublimate
does no longer increase. Then remove the tube from the heating block
Expt.54 Working in Capillaries 171

and, without breaking the vacuum, draw it out and seal it shut at 2;
finally cut it at 4, Fig. 53h.
Recrystallization from Ethanol. The sublimate, which usually is
slightly discolored, is now contained in a capillary of about 7- to 8-cm length.
Recrystallize twice from ethanol and determine the melting point by using
the technique outlined in the preceding experiment. The pure substance
is supposed to melt at 240 C. It has been suggested (1260) to get complete
0

removal of the last mother liquor by use of centrifugal force as advocated

by RICHARDS (147, 149, 696, 698). To this end, follow the directions of the
next se3tion.
Spindrying in the Capillary. The crystals of diphenyl urea get large
enough that they cannot pass through a fine capillary; thus it is not
necessary to use a filter mat.
When sealing the capillary for heating with the solvent, draw out the
end to be sealed to a capillary of 0.1- to 0.2-mm bore, and fuse shut the
fine capillary at a distance of 10 mm from the taper to the wide part.
Place the capillary with the fine end first into the test tube, and heat to
obtain dissolution. Then, by one-sided heating and flipping, get the solution
to the end which has been drawn out to the fine capillary, but do not
force it into the latter. Thus, crystallization will take place in the wide
part when the tube is allowed to cool. For separating the mother liquor
from the crystals, first cut open at the wide end, and then break the seal
of the narrow capillary. By means of a cork or rubber stopper with a
lengthwise slit, mount the capillary in the opening of a microcone as
indicated by Fig. 53k. When the cone is whirled in the centrifuge, the
mother liquor collects in the tip of the microcone, and the melting point
of the crystals may be determined in the usual manner after sealing the
end of the fine capillary.
If recrystallization and spindrying shall be repeated after determination
of the melting point, first seal the end of the wide part of the capillary.
Then place it with the wide end first into a centrifuge cone, and heat by
fanning with a flame or by placing into a drying oven to melt the substance.
Without delay, whirl in the centrifuge to get the melt to the wide end
of the capillary. Open the end of the fine capillary for taking in the
solvent, etc.

Experiment 54
Purification of Benzene; Separation by Partial Melting in the
Capillary (755, 1016)
Apparatus for the observation of 8chlieren as described in Expt. 5; V-shaped
cell, not more than 0.5 mm thick, to permit observation of 8chlieren with 1O,ttl
in the cell. - Mixture of 98% benzene with 2% p-xylene; freezlllg mixture of
sodium chloride and ice.
172 Techniques of the Submilligram Scale Expt.54-

The liquid to be purified is sealed into a capillary with two compartments

separated by a filter. The liquid is frozen in the larger compartment.
The capillary is centrifuged while its temperature rises. Only part of the
solid is allowed to melt during centrifuging so that the liquid containing
most of the impurity passes through the filter and collects in the smaller
compartment.
A rather wide capillary is used in the following experiment to get
sufficient liquid for the demonstration of the change of refractive index
by means of schlieren observation. Much smaller amounts of material

Z J
::31 a,

=+ b

-=- c

~
' ____ d

2 J
:::sQ
>-e

,= 2
==*
3
f
""'"'"'
I ,
o 5 10 em

Fig. 54. Separations by Fractional Melting

will suffice if the bore of the capillary is reduced to 0.4 mm, which will
not affect the ease of manipulation.
Procedure Giving Two Fractions. Get a capillary of 12-cm length and
1.2- to 1.5-mm bore with a constriction about 4 cm from one end as shown
in Fig. 54a. Insert through 3 a few fibers of freshly ignited asbestos, and
press them lightly together by means of a glass rod of I-mm diameter
to obtain a filter mat at 2. By means of a microflame, attach a glass thread
at 3 and draw out to obtain a fine tip of I-cm length, Fig. 54b. Through
this, take into the pipet 50,u1 of the benzene-xylene mixture: c. Draw
out and fuse shut the wide end, and while this end is cooling and the liquid
is drawn into the capillary, fuse shut also the tip of the fine capillary,
Fig. 54d.
Place the capillary with the end 1 first into a heavy-walled centrifuge
tube of 10-cm length with a small plug of cotton in the tip for the protection
of the sealed end of the capillary. For an alternative, a test tube may
be made from glass tubing of 6-mm outer diameter by sealing one end;
Expt.54 Working in Capillaries 173

it should be at least 10 cm long and by all means 1 cm longer than the

metal shell of the centrifuge so that it may be easily lifted out for inspection.
Balance the tube for centrifuging.
Immerse the tube, with the capillary inside, into the freezing mixture
and leave it there until the contents of the capillary have solidified. Then
remove the tube, wipe dry its outside, and without delay place it into the
shell of the centrifuge and set the latter in motion. Stop the centrifuge
at suitable time intervals (20 seconds for a first tri~l) to inspect the contents
of the capillary. When nearly half of the crystal mass has melted, stop
centrifuging, remove the capillary from the tube, and place it upon a metal
block so that it may acquire room temperature: Fig. 54e. First break
open the seal at 3, and then cut apart at 2.
To test the effect of the interrupted melting, transfer the mother
liquor in 1 to the schlieren cell by means of a capillary pipet. Draw a
long fine tip from end 3, and allow the liquid obtained from the melting
of the crystals to flow through this tip into the mother liquor contained
in the cell. Note the character and estimate the strength of the
schlieren.
Procedure Giving Three Fractions. Repeat the experiment, but use a
capillary of l.5- to 2-mm bore with three compartments, Fig. 54/, and a
filter mat between 2 and 3. Take 80,ttl into compartment 3, freeze, and
centrifuge to collects 15,ttl of mother liquor in compartment 1. Fuse
shut at the constriction between 1 and 2, and set aside the first mother
liquor which has been sealed into compartment 1. Again freeze the contents
of compartment 3, and centrifuge to get 40,ttl of mother liquor into compart-
ment 2. Allow the capillary to acquire room temperature. Break the seal
at the tip of 3, and cut off compartment 2 with the second mother liquor.
Transfer the first mother liquor into the schlieren cell, and allow the melt
of the crystals to flow into it; again note the character and estimate the
strength of the schlieren.
Instead of observing the schlieren phenomenon, one may determine
the melting point (and possibly boiling point) of the crystallized substance
remaining in compartment 3. To this end, it is only necessary to seal
both sides of compartment 3 with the liquid inside and to attach it with
the liquid close to the bulb of a thermometer which is mounted in a stoppered
test tube. The test tube is filled with a suitable liquid (ethanol) so that
an air bubble of about 3 ml is left. The test tube is immersed into the
freezing mixture. When the contents of the capillary have crystallized,
the test tube is removed from the bath and wiped clean (this may be followed
by wiping with a glycerol-treated cloth). For the observation of the melting
point, rock the test tube around a horizontal position so that the air bubble
mixes the liquid in the test tube. Read the thermometer when the contents
of the capillary become liquid.
174 Techniques of the Submilligram Scale Expt.54

By means of a centrifuge permitting controlled heating during operation,

the technique could be extended for the purification of solids by removal
of the eutectic melt. The use of suction in place of the centrifugal force
provides a simpler solution which may, however, be less efficient and,
in addition, may cause loss of substance if its vapor tension is high.
FISCHER (421) first determines the melting point of the eutectic mixture
by heating a small sample on the hot stage under a microscope. The
purification is carried out in a capillary of 1.3- to 1.4-mm bore and 20-cm
length, which has a filter mat 8 cm from one end. FISCHER prepares the
filter mat by sintering glass powder, but asbestos or even paper filter
may be substituted depending upon the required temperature. The
substance is introduced through the opening closer to the mat and pressed
against the latter by means of a glass rod which is left in the capillary.
For heating, the capillary is placed upon the hot stage between two metal
strips, 10 mm X 60 mm X 2.5 mm thick, which are held to the sides of the
capillary by means of clamps. Two small holes must be drilled into the
ring of the KOFLER hot stage so that the capillary may be placed along a
diameter across the center of the stage. A microscope slide may be placed
upon the two metal strips before putting the circular glass plate upon
the ring. The capillary is moved so that the substance is in the field of
vision and may be observed during heating; the end of the capillary which
is farther away from the substance is connected to vacuum line and mano-
meter. The stage is heated to a temperature about 3 degrees above the
melting point of the eutectic mixture. When the material in the capillary
has sintered, the suction is turned on and the material is pressed together
with the glass rod until most of the eutectic melt has been removed.
For the final purification, FISCHER suggests three means: (a) gradual
raising of the temperature close to the expected melting point of the
purified substance with continuous or intermittent (if the vapor tension
is high) application of suction while the substance is pressed together;
(b) pressing a small wad of asbestos with the rod against the substance so
that it will absorb melt left on the walls of the capillary and near the free
surface of the substance; and (c) pulling the intake end of the capillary
so far out of the hot stage that the substance may cool to room temperature,
and applying 4 to 5,ttl of wash liquid while the suction is on; the wash
liquid should be a poor solvent for the substance to be purified; when the
suction has removed the wash liquid, the capillary is pushed back into
the hot stage for the drying of the substance.
The capillary is finally removed from the hot stage and cut at the
location of the dry purified substance which may be easily removed.
The procedure may be practiced with a mixture of 80 % salicylic acid
and 20% benzoic acid, the eutectic mixture of which melts at 110 0 C;
the purified salicylic acid may be washed with ethanol.
Expt.oo Working in Capillaries 175

Experiment 55
Cupric Ammonia Complex, Observation 01 Color m the
Capillary (1004, 1005, 1139)
Capillary clamp or substitute.
Faint colorations in small volumes of liquid are made perceptible by
taking the liquid into a long, narrow capillary and looking lengthwise
through the column of liquid. Lengths of 1 cm to 10 cm of heavy-walled
capillary of 0.1- to 0.5-mm bore (80 nl to 19 III capacity) may be used and
polished flat at both ends. For filling, the clear liquid is added at one end
until convex menisci are obtained at both orifices, Fig. 55a; long capillaries

, I

.1cm (I

c d e
~~b
Fig. 55. Coloriscopic Capillaries. The bore of the tubes is somewhat exaggerated

are better held horizontally until the cell is completely assembled. One
end is closed by placing the tube upon (against) a microscope slide. If
necessary, more liquid is added to the other end to get a convex meniscus,
whereupon this end is closed with a fragment of a cover slip. Air bubbles
must be absent, but may be dislodged by taking off the cover slip and
stirring with a fine platinum wire. The slide is placed upon the stage of
the microscope, Fig. 55b; the condenser is used to concentrate the light
upon the lower orifice of the tube, and the microscope is focused upon
the top opening. Obviously, it is desirable that the glass of the capillary
is colorless or nearly so.
Capillaries of black glass are recommended for the observation of the
absorption spectrum. To this end, a spectroscopic eyepiece may be used
in conjunction with the microscope (49), or the capillary is horizontally
mounted in the path of the light before it enters the slit of a spectroscope;
in the latter instance, the capillary is mounted in a stand for filling to
avoid warming it with the hands, and both ends are closed with fragments
of cover slips (1005). Cells have been designed for use with spectrophotom-
176 Techniques of the Submilligram Scale Expt.66

eters (441). The technique described in the following uses simple capillary
tubes.
Development of Color and Filling of Tube. With a capillary pipet of
0.5-mm bore, take up 1 JlI of a mixture of 1 volume ferric test solution
and 10 volumes copper test solution (1 mg Fe and 10 mg Cu per ml). Place
a drop of 6-F NHa on a slide, and insert the tip of the pipet containing
the test solution. Allow ammonia to enter until it fills 6 cm of the pipet.
Then draw out and seal the wide part of the pipet about 2 cm from the
meniscus of the solution, and fuse shut also the tip of the pipet, compare
Fig. 49. Mix the contents of the capillary by briefly whirling the liquid
first to one end of the capillary and then to the other. Finally collect
the precipitate at one end by somewhat prolonged centrifuging which
must give a perfectly clear liquid.
First open the empty end 1 of the capillary, Fig. 55c, and then cut
and break at 2 to remove the collected precipitate. Fuse shut the end 1 so
that a rounded-off seal is obtained, Fig. 55d. Whirl in the centrifuge to
transfer the liquid to the sealed end 1, Fig. 55e; lay the capillary upon a
slide and inspect under the microscope to make certain that no solid has
collected at the seal 1.
Observation of Color. Cut the capillary just below the meniscus 2 so
that it breaks evenly. Without delay, mount the capillary in a vertical
position upon a slide so that there is room for applying a drop of immersion
liquid (cedar wood oil, lubricating oil, or water) where the sealed end rest
upon the slide. The commercially available clamps are most convenient (417,
862), but many substitutes are available: a stopper with a slit for inserting
the capillary, Fig. 55/; a metal block with vertical wells of different bore
to fit the capillaries; etc.
Concentrate the light with the condenser upon the sealed end of the
capillary and, first using a 2 X to 5 X objective, focus upon the top rim
of the capillary. If the liquid does not absorb strongly, the bore of the
capillary showing the color of the liquid will be brighter than the ring
representing the cross section of the tube, Fig. 55g. From time to time
and by means of a platinum loop, add water to the top opening of the
capillary to maintain a plane or convex meniscus. Otherwise, the image
of the bore will turn gray or black depending upon the curvature, and
the observation of the coloration is rendered impossi~le when the meniscus
becomes concave and recedes into the capillary because of evaporation.
The evaporation may be delayed by placing a tiny fragment of a cover
slip upon the convex meniscus projecting above the top opening. The
immersion liquid applied to the base of the capillary improves the optical
path.
The recognition of light coloration, especially if it has a hue similar
to the color of the glass of the capillary, requires comparison with blanks
 Working in Filter Paper 177

pedormed in capillaries of the same kind. To this end, suitable capillary

clamps should permit getting the bore of two or more capillaries into
one microscopic image (focusing the ends of a bundle or row of capillaries).
The absorption spectrum may be observed with the use of a spectroscopic
eyepiece (88); for an alternative, the image of the colored circle may be
projected onto the slit of a standard spectroscope (1283).
The microscope provides a convincing demonstration, but the color
of the liquid may be seen with unaided eye or a magnifying lens when
viewing the cut end while holding the capillary over a brightly illuminated
sheet of white paper. Even when the light is directed to the side of the
capillary, enough of it is reflected in an axial direction to show the color
of the liquid when looking vertically toward the cut end.

Working in Filter Paper

Only such procedures shall be considered in this connection, in which
the paper is an essentially inert medium supporting and retaining solids
and permitting the flow of liquids. The separations may be considered
filtration, washing, and extraction in the plane of the paper, whereby
only one liquid phase is involved and the paper acts neither as adsorbent
nor as ion exchanger as it does in chromatography.
Whereas it is unavoidable that separations of this type occasionally
occur in the pedormance of spot tests on paper, it probably was the sulfide
fiber of EMlCH (550) rather than spot analysis, that lead CLARKE and
HERMANCE (410, 418) to the pedormance of chemical separations in filter
paper. The separation is brought about by a process of extraction. Conse-
quently, a solid sample may be deposited upon the paper and washed
by flowing a solvent upon the sample, which dissolves only certain substances
that are then carried away by the spreading of the liquid in the paper.
Solutions are deposited in a small area of the paper and sometimes allowed
to evaporate; the moist spot or the residue is treated with some precipitating
reagent, and this is followed by extraction of the solubles. Finally, the
paper may be impregnated with a reagent that immediately precipitates
certain substances while the liquid sample spreads through the paper.
In all instances, the insoluble part of the material is held in the paper at
the point where the sample is added, and the soluble substances are carried
away and diffused through the paper. If the soluble substances shall be
collected, it is necessary to regulate the process in such manner that the solvent
flows to a certain, well-defined, and small area, where it evaporates to leave
the solutes behind in the paper. So far, actual practice has followed two ways.
CLARKE and HERMANCE (874) added the sample and the solvent to
the lower end of a narrow strip of paper which was suspended inside a
vertical glass tube. The solvent rose to a level at which a current of air
was passed across the strip of paper, which was entering and leaving through
Benedetti-Pichler, Identification 12
178 Techniques of the Submilligram Scale

side arms on opposite sides of the vertical tube. The zone of evaporation
was thus determined by the level of the side arms and could be moved
at will by raising or lowering the strip of paper in the tube. The arrangement
gave the additional benefit that it must have eliminatEd the effect of
increased evaporation along the edge of the strip (947). The sensitivity
could have been increased by trimming the strip to a narrow waist at the
height where the solute is collected by evaporation.
The substance on corroded electrical contacts was collected about
I cm from the lower end of the paper strip which was then immersed in
organic solvent. Lubricating oil and tar were caught in an evaporation
zone 12 cm above the sample; after changing the solvent, soaps were
collected in a band 10 cm above the sample. After this followed extraction
with water, which was repeated after digesting the water-insoluble residue
with nitric acid. The water-soluble substances were collected 8 cm above
the sample, and the acid-soluble, 6 cm above it. The acid-insoluble resi(].ue
and the above listed fractions were cut out of the strip and individually
subjected to analysis.
For the separation of traces of nickel from the solution of a steel, the
lower portion of the paper strip was impregnated with barium carbonate
which precipitated the iron, but allowed washing the nickel ions into the
zone of evaporation, where they accumulated and could be detected with
dimethylglyoxime. In a somewhat similar manner, but using a sewing
thread in place of the strip of paper, MAHON (779, 963) was able to separate
and identify down to 0.015 pg Co or Fe, 0.1 pg Cu, and I pg 7n in solutions
containing these metals (limiting proportions of I: 5 and better).
Paper may be impregnated with "insoluble" reagent like dimethyl-
glyoxime by simply soaking it in the alcoholic solution and drying. The
precipitation of reagents in paper has been described on p. 130, and papers
loaded with precipitates of silver chromate, zinc sulfide, cadmium sulfide,
antimony sulfide, zinc ferro cyanide, cadmium ferrocyanide, and cadmium
xanthate have been used.
WEISZ (172, 916, 917, 922) escapes the effect of the edge of the paper
by allowing the solvent to flow from the center of a disk of paper toward
the circumference. This would give a very undesirable dissipation of the
dissolved solutes, but resting the paper upon a heated ring and supplying
the solvent at a suitable rate permit concentrating the solutes in a circ:ular
zone of only 0.3 to 0.1 mm and less in width so that the amount of substance
per unit area may become larger than it was in the original central spot.
Thus, this ring-oven technique collects the solutes with high areal concentra-
tion into a long (about 7 cm) but very narrow zone which may be cut
into ten or more sections for the performance of various tests. If the tests
are performed so that the solutes remain in the ring zone, a reasonably
precise estimation of quantity is possible (918, 931), and comparison
 Working in Filter Paper 179

with a standard silver sulfide scale suffices for a whole series of different
sulfides (951). Autoradiography may be used for locating and estimating
substances in the evaporation zone (928).
The concentrating effect· of the narrow evaporation zone is good enough
to invite use of the ring oven for evaporating solutions previous to the
pedormance of spot tests. Thus WEST and MUKHERJI (473) separate the
metal ions by liquid-liquid extraction into five groups. The solution of
each group is then fed to the center of a paper disk resting upon the ring oven,
and the solutes are washed into the evaporation zone. The circle containing
the solutes is cut into sections which are used for the confirmatory tests.
In this manner, the presence or absence of 35 metallic ions may be determined
within one hour.
Quite lately, WEISZ (973) has demonstrated that a narrow paper strip
may be used with the ring oven in place of the circular disk. In this manner,
the length of the narrow zone of evaporation may be decreased from 7 cm
to 6 mm, which should improve the sensitivity of the technique by about
ten fold. To the same end, the disk could be replaced by a 5- or 6-pointed
star, possibly with the tips ending in I-mm wide ribbons that pass through
the evaporation zone.
For work with disks, WEISZ uses "ash-free", fast filter paper, Schleicher
& Schuell No. 589, black ribbon!, and transfers 1.5,tt1 of the solution
to be treated with a capillary pipet to the center of the disk2. A reagent
is added, which precipitates at least some of the substances present, and
gaseous reagents are preferred for this purpose since they do not cause
spreading of the initial spot. The disk is laid upon the ring oven, and
solvent is added to the center of the spot to rinse the soluble substances
through the paper to the inner edge of the heated ring. Portions of 10,tt1
of wash liquid are added with a thick-walled capillary with a conical tip
that is polished flat at the orifice to provide good contact with the paper.
The amount of liquid added at one time must be adjusted to the capacity
of the paper within the ring zone; too much would cause flooding and
overrunning the heated sudace of the ring.
The flow of wash liquid regulates itself since it is drawn out of the
capillary at the rate dictated by the spreading in the paper due to capillarity.
When the capillary runs empty, it is again filled and applied to the spot;
1 Glass fiber paper does not seem to have been tried, but it might be useful
for working with strongly acid or alkaline solutions.
2 Minute amounts of insoluble substances are fused with a suitable reagent,
P. 42, on a square of platinum foil, 5 mm X 5 mm X 0.03 mm thick. The foil
with the melt is placed upon the center of the (55-mm diameter) disk of paper
and covered with a disk of lO-mm diameter, which is provided along its circum-
ference with 5 or 6 tiny droplets of a suitable cement. The large disk is placed
on the ring oven, and the melt is extracted by adding solvent to the center of
the sandwich, etc. (974).
12*
180 Techniques of the Submilligram Scale Expt.56

this has to be repeated five to ten times for washing the solutes completely
into the zone of evaporation.
The paper disk is finally dried in an oven. The initial spot is isolated
by punching it out of the dry disk with a circular die of 10-mm diameter.
The circle of evaporation of 22-mm diameter is cut into as many sectors
as the problem requires. The small circle with the initial spot may be
placed upon the center of a fresh disk of paper, treated with reagent,
and again washed to collect another group of soluble substances in the
evaporation circle of the new disk.
The ring oven consists of a circular cylinder of metal, 35 mm high and
55 mm in diameter, which has been drilled along its axis to provide a
central hole of 22-mm diameter. It may be made of aluminum or of copper
and plated with gold. It is supported by a tripod and heated electrically
to a temperature 5 to 10 degrees above the boiling point of the solvent.
The disk of paper is placed upon the top surface and secured there by
laying on it a ring of glass or porcelain with an inner diameter of 25 mm.
The market also offers a ring oven made of glass (Paul Haack, Garnison-
gasse 3, Vienna IX, Austria), which is heated by the vapor of boiling
liquids (953). Tetrachlorethylene is used when washing with water since
the boiling point of the liquid must be taken about 10 degrees higher than
the temperature required at the surface of the oven. This oven is also
provided with a thin glass plate with a central circular opening of 12-mm
diameter; by placing this plate upon the paper disk, a second circle of
evaporation with a diameter of 12-mm may be obtained and later isolated
by using a die of 15-mm diameter in addition to the one with 10-mm diameter.
The technique of working in paper has the attractions and disadvantages
associated with spot testing. The technique is simple and time saving.
On the other hand, the substances must be finally extracted from the
paper if they shall be obtained or observed in the pure state. The paper
is attacked by strongly acid and alkaline solutions, but this difficulty may
be overcome by using paper made of glass fibers and other suitable materials.
Even high-grade paper contains impurities (Fe, Se) which become
concentrated in the evaporation zones, may interfere with the tests, and
may require washing of the paper previous to use (975).
The special literature should be consulted concerning the various
techniques of chromatography (63-76), which offer unique possibilities
for the sensitive separation of closely related substances.

Experiment 56
Ring-Oven Technique tor Extraction and Evaporation in Paper
L.1., usually fractions of a microgram.
Ring oven with capillary for adding wash liquid and circular die of lO-mm
diameter; apparatus for treating paper with gaseous reagents (916): a glass
Expt.56 Working in Filter Paper 181

tripod for supporting the paper disk in a covered beaker filled with the gas may
be substituted; test tube and atomizer head for spraying reagents; "ash-free"
filter paper, S. & S., No. 589, Black Ribbon, disks of 55-mm diameter. - Ferric
chloride solution, 0.1 mg Fe/ml; copper-iron-nickel solution (0.3 mg of each ion
per ml): a mixture of 3 ml each of the test solutions (10 mg ion per ml) diluted
with O.l-F HOI to 100 ml; potassium ferrocyanide, 2% solution; rubeanic acid,
1% solution in ethanol; dimethylglyoxime, 1% in ethanol.
The following practice experiments have been suggested by WEISZ (916);
the second has been slightly modified.
Concentrating into the Ring Zone for Testing. With a capillary pipet,
transfer 1.5,ug of ferric chloride solution (0.1 mg Fe/ml) to the center of
a disk of paper. Place the disk upon the ring oven so that the moist spot
is located above the center of the oven which is heated to 105 to llO° C.
Place upon the paper the glass or porcelain ring of the oven with 25-mm
inner diameter. Take 10,ul O.l-F HCl into the pipet. Insert the pipet
into the holder of the oven and lower it until the tip touches the paper
in the center of the moist spot. The contents of the capillary are drawn
into the paper. When the capillary is empty, fill it again and apply it as
before. Repeat this 5 to 10 times to wash the ferric chloride completely
into the zone of evaporation. This will require about two minutes. Finally,
remove the paper disk from the ring oven and place it into a drying oven.
Support the dry disk horizontally and spray it with a mixture of equal
volumes of ferrocyanide solution and O.l-F HCI. A sharp circular blue line,
not more than 0.3 mm wide, should be obtained, and the central portion
of the disk should not show a blue coloration.
Separation by Solid-Liquid Extraction. With a capillary pipet, transfer
1.5,ul of the copper-iron-nickel solution (0.3 mg of each ion per ml) to the
center of a paper disk. Place the disk with the wet spot into an atmosphere
of H 2 S and leave it there for 3 minutes. Test for complete precipitation
of the copper by adding to the center of the dark spot from a capillary
pipet just enough of a mixture of equal volumes of ethanol and O.l-F HCl
to get a wet zone, about 1 mm wide around the spot. Return the disk
into the H 2 S atmosphere; the precipitation of the copper was complete
if the color of the moist zone around the sulfide spot does not change.
To extract iron and nickel, place the paper disk upon the ring oven
which has been heated to 105 to 1l0° C. Treat the central spot first with
2,ul ethanol, which seems to change the nature or distribution of the
precipitate so that the extraction may be efficiently performed. Then
use 1O-,ul portions of O.l-F HCl to wash iron and nickel into the zone of
evaporation; this requires 1 to 2 minutes. Finally dry the paper in the
drying oven, and then cut out the central sulfide spot with the 10-mm die.
For the identification of iron and nickel, cut the resulting paper ring
into three sectors. Spray the first with the mixture of equal volumes of
ferrocyanide solution and O.I-F HOI; the second, with rubeanic acid;
182 Techniques of the Submilligram Scale

and the third with dimethylglyoxime solution. Expose the last two sectors
to fumes of ammonia.
For the identification of the copper, moisten the small disk of paper
carrying the sulfide spot with 0.1-F HOI and expose it briefly to bromine
vapors. When the dark color of the sulfide has disappeared, place the
small disk upon the center of a fresh disk of paper mounted upon the ring
oven. Lower the capillary to the center of the small disk, and wash the
copper with 10-,u1 portions of 0.1-F HOI into the evaporation zone of the
large disk. Finally dry the large disk in the drying oven, and cut the dry
circle into four quadrants. Expose one quandrant to fumes of ammonia;
spray the second with rubeanic acid; the third, with ferro cyanide ; and
expose the latter two also to fumes of ammonia.

Particles of Ion Exchange Resins as Reaction Media

A review of this technique, developed for the performance of confirmatory
tests since 1952, has been written by FuJIMOTO (622). In ~ll instances,
a colored compound is formed on the resin, but fluorescence and luminescence
might provide even higher sensitivities. In spite of the fact that the smallest
particles have a diameter of 0.25 mm, 0.05 mms cross-section area, and
several particles are used, the limits of identification compare favorably
with those obtained when collecting the matter upon the tip of textile
fibers (p. 183), which permits concentrating it into an area of only
0.0004 mmS : 0.02-mm length of fiber of 0.02-mm diameter. This is not
strange when considering that resins may be chosen for their special affinity
to the substance to be detected, whereas the collection of test substance
on textile fibers occasionally relies upon adhesion during evaporation
(Expt. 58) and may be far from complete. If either reagent or tested
substance has an affinity to the fiber as in the instance of dyes, the resulting
limit of identification is quite satisfactory (Expt. 57).
FuJIMOTO, who.traces the history back to the experiments of GUNTHER-
SCHULZE (1210) with natural zeolite, lists as advantages offered by the
synthetic resins their uniform reproducibility, their selectivity, and the
improved stability of the obtained colorations.
The performance is as simple as that of spot tests. "On a white (or black)
spot plate, mix a few grains of a colorless or lightly colored ion exchange
resin (cation or anion exchange type, usually of low cross-linkage) with
a drop of the test solution (or of the reagent solution). Adjust the conditions
in the reaction medium and allow to stand until the ions under test (or
the ions of the reagent) are strongly adsorbed in the resin phase. Now add
a drop of reagent solution (or test solution) and, if necessary, adjust the
pH and the ionic strength of the medium. Allow to stand appropriately
and 0 bserve the coloration developed in the resin phase under magnification.
 Working on Textile Fibers and Wires 183

(In some cases the resin grains may be previously impregnated with the
reagent and stored.)" (622)
It is obvious that the limits of identification listed by FuJIMOTO (622)
may be improved in several ways: (a) by the use of small drops and a
single grain or a fraction of it; (b) by concentrating the test solution before
introducing the grain; (c) possibly by removing the exhausted test solution
before adding the reagent or vice versa; and (d) by removing the grain
from the colored reagent solution for inspection. In addition, the grain
might be placed into the taper of a capillary pipet with fine tip through
which it cannot pass so that test and reagent solutions may be added
and removed by the customary pipetting technique. Finally, the grain
may be taken up (speared) with the point of a glass needle and thus
transferred to test drops and reagent drops contained in capillary cones
(p. 198). Both techniques would permit multiple tests (Expt.26) and use
of the exhausted test solution for testing for an ion of the opposite charge .
.AP, a rule, strongly acidic cation exchange resins and strongly basic
anion exchange resins of low cross-linkage are used, which are colorless
or lightly colored to assure high sensitivity. Before use, the commercial
resins must be purified regardless of claims concerning absence of metal
contamination. They are packed into a tiny column, washed with dilute
hydrochloric acid, converted to the desired form in the customary manner,
washed with metal-free water, and dried in air. For examples of tests
see FuJIMOTO (622).

Working on Textile Fibers and Wires

EMleH conceived the idea that the limits of identification of colorations
could be vastly improved by using tiny objects such as yeast cells or even
bacteria for carriers of the colored matter. To-day, one might profitably
generalize by including all tests using a distinctive property: radiation,
fluorescence, absorbtivity for any type of radiation, electric potential,
magnetic susceptibility, catalytic action. Textile fibers were chosen for
the practical utilization of the principle since they may be efficiently
handled without resorting to manipulation under the microscope. Using
fibers dyed with indicators (1000, 1001), adsorption of colloidal matter (1002),
and the deposition of colored precipitates (550), the limits of identification
were pushed into the nanogram range. Logical extension of the reasoning
led to the collecting of matter upon the cross section of thin wires by electro-
deposition, whereby like sensitivities were obtained (858).
Systematic investigation (6S0, 681) lead CHAMOT (US) to the conclusion
that the type of fiber should be selected so that it adsorbs the reagent
strongly in relatively high concentration and does not "bleed", i. e., give
it off too readily when immersed into the test solution. Obviously, the fiber
184 Techniques of the Submilligram Scale

must not react with the reagent in any manner which would decrease
its efficiency. Thus, silk fibers (9/-lm to 21 /-lm diameter) are recommended
for red and blue litmus, Congo red, and the adsorption of colloidal gold;
viscose-rayon (30/-lm) for turmeric, potassium ferrocyanide, potassium
thiocyanate, and gold; and wool (12/-lm to 60/-lm) or gun cotton (10/-lm
to 40/-lm) for carrier of zinc sulfide. Flax (about 20/-lm) is a close second
for use with turmeric.
Very important is that fibers intended for use with aqueous solutions
are prepared and handled so that they do not become water repellent.
The suitability of reactions for fiber tests depends, of course, upon the
possibility of concentrating the characteristic product on or in the fiber.
If the reaction product is a colored precipitate, one might expect that a
test that is succesful when performed as drop reaction on paper should
also be suited for the fiber technique. This need not happen, however,
since single fibers do not possess the same adsorptive power as paper.
Also the particle size may very with the technique and with it, the coloring
power. The coloring power increases with growing size for some substances,
and with others, it decreases (670, 682, 740).
Aside from the performance of confirmatory tests, the fiber technique
may be used for side tests when working upon the submilligram and micro-
gram scales. Only a permissible amount of solution will be lost when
the end of some textile fiber carrying a pH indicator, a redox indicator,
or a precipitating agent is briefly inserted into a minute droplet. Fibers
treated with reagent may also be used to greatly increase the sensitivity
of tests for gases, P. 36. If the material to be investigated may be made
to cling to a fiber, it becomes possible to perform separations upon a very
small scale or to determine the presence or absence of groups of
substances (414).
So far, only simple procedures were tried. To improve the sensitivity
for the detection of very small amounts of matter, EMleR concentrated
it by allowing the droplet of test solution to evaporate while only the very
tip of the fiber is immersed. A relatively small number of reagents have
been tested, and nothing is known about attempts to get separations by
having test solutions rising on fibers treated with selective precipitating
agents. Migration in the electric field, however, has been used for the
separation of nucleic acids (1052, 1053) and proteins (766) on fibers.
Litmus Fiber. L. I., 30 pg hydrogen ion; 200 pg hydroxyl ion. CRAMOT
and COLE (680, 681) recommend the following procedure. Litmus is purified
according to W ARTRA (600) by extracting the commercial "cubes" with
95 % ethanol until the extract has no longer a reddish hue. The dye
remaining in the residue is then extracted with water, whereby air is blown
through the mixture to prevent reduction of the dye. The aqueous extract
js filtered and then concentrated at steam bath temperature to the
 Working on Textile Fibers and Wires 185

consistency of a syrup. The syrup is repeatedly treated with acetic acid

and ethanol and evaporated to dryness for the destruction of carbonates.
The residue is extracted with portions of absolute alcohol (ethanol) until
the light reflected from the extract no longer shows a reddish hue. The
thus purified residue is dissolved in water and evaporated to a thick syrup
which is then treated with absolute alcohol. The paste obtained is stirred
w\th fresh portions of absolute alcohol until the latter does no longer
extract red coloring matter. The final residue is dissolved in water. The
solution is evaporated to a thick syrup which is poured into absolute
alcohol. The gummy precipitate is spread upon a clay tablet and dried
at about 75° C to a hard mass which is readily soluble in water.
Raw silk of good quality is boiled in a very weak soap solution, rinsed
thoroughly, and then digested at room temperature for 2 hours in a solution
of 10 g NaOH in 100 ml water. The silk is then washed thoroughly with
distilled water and transferred into a 10% solution of the purified litmus
dye, which has been acidified by adding 3 drops 3-F H 2 S04 per 100 ml.
The solution with the silk in it is evaporated to a thick syrup, whereupon
the silk is removed and washed in running water. To give the fibers the
middle tint, the silk is stirred with water to which very dilute NaOH is
added until the silk becomes violet. The silk is again washed in running
water and then dried. To get blue litmus fibers, the adding of NaOH is
simply continued until the color changes to blue; in a similar manner,
red litmus silk is obtained by stirring silk of the middle tint with water
and adding dilute acetic acid.
Red Congo Fibers. Silk is purified as described above and then dyed
in a 0.5 % solution of Congo red, made alkaline with NaOH. - Viscose
rayon is boiled with a 2% (alkaline) solution of the dye for 15 minutes,
washed, and dried by pressing between filter paper.
It is obvious that the test solution must not be allowed to evaporate
upon a fiber if the latter shall indicate the pH of the solution as it is given.
Sulfide Fiber. White wool is washed with warm soap solution, rinsed
in water, and dried. The wool (or gun cotton with long fibers) is soaked
in acetone, dried by pressing between filter paper, and then digested over
night in 1 % NaOH of room temperature (ll8). After brief rinsing in water,
the fibers are dipped 5 to 6 times alternately into 10 %zinc acetate and Na 2 S.
Between dippings, the fibers are not washed, but pressed to remove excess
solution. The Na 2 S is prepared by passing H 2 S into 10% NaOH until a
drop of the solution does no longer give a precipitate when mixed with
a drop of MgCl 2 solution. After the final dipping, the impregnated fibers
are washed in water and dried by pressing between filter paper.
Sulfide fibers which have been stored for any length of time should
be tested with a known mercuric chloride solution before using them on
an unknown. When testing for small amounts of heavy metals, the test
186 Techniques of the Submilligram Scale

drop is acidified with HCI, and the fiber is inserted as directed in Expt. 57,
below. If the tip of the fiber turns black, Hg, Ag, Bi, Pb, Cu, or Pt may
be present. If the tip turns yellow, the fiber should be transferred into a
drop of 6-F NaOH; the following may happen: (a) the tip turns black:
Hg, Ag, Bi, Pb, Cu, Pt; (b) the tip remains yellow: Cd; (c) the tip turns
white or colorless: As, Sn. An orange coloration that disappears with
NaOH indicates Sb. The interpretation must take into consideration
that several of these metals may be present, that also some other elements
may react, and that the outcome of the test depends very much upon
the acidity of the solution and the concentration of the elements concerned.
The textile fibers must be replaced by filament of heat resistant material
for tests which require the application of high temperatures. Thus, wires
of suitable metals are used for the performance of fusions and bead tests;
they are the obvious choice also for electrodeposition.
Electrodes. EMICH'S idea of using the cross sections of very thin wires
for carriers of deposits has been systematically investigated by
BRENNEIS (154, 162, 858). The so-called "rod electrode" was obtained by
inserting one half of a thin wire of ll-cm length into a narrow, thin-walled
capillary of 5-cm length and bending the other half of the wire back so
that it runs along the outside of the capillary. The combination was
then pushed, with the bend first, 4 cm deep into the bore of a thick-walled
capillary of 0.5-mm i. d., 5-mm o. d., and lO-cm length. At the point where
the wire inside turned back, the thick capillary was then heated so that
its bore collapsed and the wire became imbedded in the glass. After slow
cooling, the resulting short piece of glass rod was cut below the bend of
the wire, and the circular cross section was polished to get the ends of the
wires in the level of the glass surface. Using wires of platinum or platinum-
iridium of 0.025-mm to O.l-mm diameter, circular electrode areas of 5· 10-4
to 8· 10-3 mm2 could be obtained. Since the distance between the wires
need not exceed 0.1 mm, hemispherical drops of electrolyte of 0.15- to
0.3-mm diameter and 0.8 to 7 nl volume could be used in the performance
of electrolyses. To prevent evaporation of the test drop, a test tube with
an inner lining of moist filter paper was inverted over the rod electrode.
A "needle electrode" of not more than I-mm outer diameter was obtained
by using a thin-walled capillary for the outer glass tube. The needle
electrode is intended for insertion into a small amount of electrolyte which
may be on a slide or in a tube or crucible. For the performance of a side test,
it may be inserted into a solution which is treated in a centrifuge tube
or microcone. The needle electrode as well as the rod electrode is easily
mounted in a plastic handle provided with binding posts for connecting
to the supply of current.
The electrolytic slide, which is commercially available, is obtained by
fusing three insulated wires into a capillary of 5-mm o. d., cutting it to
 Working on Textile Fibers and Wires 187

~ length of 15 mm, and mounting it on a plate of hard rubber, which also

carries three binding posts. The short piece of glass rod containing the
three wires is surrounded by a plastic ring of 12-mm i. d., which can be
raised or lowered by turning it. A closed cell and retarding of evaporation
are obtained by placing a cover slip upon the ring. The area around the
three circular electrodes may be made water repellent by applying paraffin,
and the droplet of electrolyte may be applied to the underside of the
cover slip and then brought into contact with the electrode surfaces by
lowering the ring. The slide has been given standard dimensions so that
it may be readily mounted under the microscope and also fits a mechanical
stage. Focusing through the cover slip permits microscopic examination
during electrolysis. The third electrode is not connected to a source of
current and serves as a blank for comparison. The electrodes are so small
and so close together that all three are included in the field even when
high magnifications are used. A vertical illuminator or equivalent is
naturally required for observation by reflected light with the use of high
magnification.
Electrolyses are performed by the standard procedures. If these require
it or if it is desirable for the better visibility of the deposit, the cathode
(and the blank electrode) may first be coated with a suitable metal; coatings
of gold, silver, and copper have given good results. The needle electrode
is dipped into the properly prepared electrolyte, or the latter is deposited
by means of a loop, hook, capillary pipet, or micropipet upon the electrode
area of the rod electrode or the electrolytic slide. The connections are made
in the customary manner. If much of the sought ion is present, the deposit
may be visible after a few seconds. Near the limit of identification (about
I ng in the instance of Cu, Ag, Au, and Pb), the electrolysis may have to
be continued for 15 minutes to 2 hours; 24 hours in the instance of blanks.
If the electrolyte contains sulfuric acid, this suffices to prevent evaporation.
With other electrolytes, a moist atmosphere must be provided. The deposit
is finally washed without interrupting the flow of current by adding a
droplet of water to the electrolyte, removing it with a capillary pipet or
a piece of filter paper, and repeating the adding and removal of water
as often as seems advisable.
The appearance of the deposit suffices for the recognition of copper,
gold, mercury, lead dioxide, and manganese dioxide. Obviously, the
deposit may be dissolved and the solution used for further tests. A deposit
of mercury may be exposed to vapors of iodine; the resulting HgI 2 should
be soluble in KI solution. A silver deposit turns black when moistened
with Na 2 S solution or exposed to H 2 S, etc. It also should be possible to
·obtain contact prints of the coated electrodes (162, 434). Vice versa,
the tiny electrodes coated with suitable deposits may be used as reagents.
A silver deposit should provide a sensitive reagent for sulfide (Hepar test);
188 Techniques of the Submilligram Scale Expt.57

an electrode covered with tin and connected to a clean platinum electrode

could be used to test for antimony.
For practicing, one may use an experiment described by BRENNEIS.
A few milligrams of fresh beef liver (containing about 40 p. p. m. of copper)
are impaled upon the end of a platinum wire and ashed in the flame of
a Bunsen burner constructed of glass or porcelain. The residual black,
crusty mass is placed into a microcone containing 20 to 30 III 16-F HNO s,
broken up with a glass rod, and heated on the steam bath. After whirling
in the centrifuge, the clear liquid is taken up into a capillary pipet and
in small portions evaporated upon a narrow slide, Expt. 44. The dry
residue is dissolved in 0.2 III 2-F H 2 S04 , and the solution is electrolyzed
with an e. m. f. of 2 volts across the electrodes. The washed deposit may
be transferred to the other electrode by reversing the current after adding
a droplet of fresh 2-F H 2 S04 , A blank test may be performed with a small
tuft of purified asbestos and should give no copper deposit.

Experiment 57
Turmeric Test for Boric Acid (550)
Turmeric linen: dip unbleached linen fibers into acetone, and dry them by
pressing between filter paper. Reflux 5 g turmeric powder for 5 minutes with
10 g 95% ethanol; filter the solution, and evaporate it to dryness. Dissolve
the residue in 4 m} 50% ethanol, and treat the solution with small portions of
solid Na2COa until it becomes clear. Introduce the linen fibers, and heat just
to boiling. With a glass hook, remove the fibers from the bath, and dry them
by pressing between filter paper. Without handling them with the fingers,
immerse the fibers for 30 seconds in l·F H 2 S0 4 , rinse them thoroughly in running
water, press them between filter paper, and then allow them to dry between
two sheets of paper. With forceps, transfer the fibers which should have the
color of egg yolk to a suitable container, glass vial. - Canada balsam or other
cement suitable for attaching the fiber to glass; Plasticine or beeswax for
temporarily fastening glass to metal or glass to glass.
Selection of Fiber. It is essential to handle the fibers only with forceps
and needles so that they will be readily wetted by aqueous solutions.
Above a sheet of paper, cut the fibers with scissors into lengths of 10 to
15 mm. Transfer several of them to a slide for inspection under the micro-
scope with medium magnification and strong transmitted light. Select
an individual fiber (not a yarn or twist of several) which is reasonably
straight, not pointed but cut squarely at the end, and has a distinct yellow
color over its entire length.
Draw a fine thread from one end of a capillary of 5-cm length, and
break the thread 2 cm from the capillary. Moisten the end of the thread
with a trace of Canada balsam or some other suitable cement, and then
touch with it the lesser end of the selected fiber. Finally, attach the capillary
Expt.57 Working on Textile Fibers and Wires 189

to the body tube of a microscope with Plasticine or beeswax as indicated

by Fig. 56.
Concentrating the Test Substance at the End of the Fiber. Take into
a capillary pipet about 0.2,a1 of borate test solution (1 mg BOa per ml)

Fig. 56. Concentrating Solutes upon the End of a Fiber. The thickness of the fiber
is greatly exaggerated

and an equal volume of 6-F HOI. Transfer the liquid to the center of a
slide. Place this slide upon the stage of the microscope, and insert the end
of the fiber into the droplet as shown in Fig. 56. Observe the evaporation
of the liquid with the aid of a magnifying glass; the end of the fiber should
remain in contact with the liquid until the latter has completely evaporated.

Fig. 57. Fiber Mounted for Examination. Its thickness is greatly exaggerated

If the evaporation is nearly complete before the fiber has been properly
inserted, treat the residue on the slide with 0.5,a1 6-F HOI, and again
insert the end of the fiber.
Observation of Test. When the test droplet has completely evaporated,
raise the tube of the microscope, remove the capillary with the fiber on
the end, and place it upon a clean slide, Fig. 57. If desirable, attach the
190 Techniques of the Submilligram Scale Expt.58

capillary to the slide with Plasticine or beeswax. Place a dry cover slip or a
fragment of a cover slip on the fiber to hold it close to the surface of the slide.
With low magnification, focus first upon the point of the glass needle,
and then follow the fiber through its entire length until its free end is found.
If desirable, change to a higher magnification, and again inspect the entire
length of the fiber. For the most part, it should still show the original
yellow color, but the free end should now be brown or reddish brown.
Use strong transmitted light or reflected light and a white background.
If a microscope with polarizer in position is used, rotate the stage
and observe the color of the fiber in the various positions of the latter.
On linen fibers, the red coloration shows very strong pleochroism from
red to nearly colorless, and there is the danger that the test could be
completely missed if the fiber is accidentally kept in the "colorless" position
during the observation.
Treating the Fiber with Reagents. To confirm the presence of borate,
adjust the illumination and focus the free end of the fiber so that the red
or brown coloration is clearly seen. By means of a glass rod or medicine
dropper, place a large drop of I-F NHa upon the slide so that it touches
the edge of the cover slip, and immediately view through the microscope.
The ammonia spreads between slide and cover slip, and the fiber becomes
immersed. The color of its free end changes to blue (or green with small
amounts of boric acid), whereas the rest of the fiber becomes purplish red.
The blue color of the tip fades quickly, which makes it advisable to observe
while the NHa is being added. Also the blue coloration shows pleochroism
to colorless.
Experiment 58
Test for Bismuth, Precipitation of the Sulfide Upon the Fiber and Oonversion
to Sulfate, Ohromate, and Elemental Bismuth
Surgical cotton with long fibers; acetone, pure; Canada balsam; Plasticine
or beeswax; sodium sulfide-hydroxide reagent, Expt. 38; bromine water; O.I-F
K 2Cr20 7 ; stannite reagent, Expt.24.
Preparation of Fiber. Pick up a tuft of cotton with forceps. With
scissors and above a darkly colored paper, cut the fibers to lengths of
10 to 15 mm and bathe them in a few drops of acetone for removal of fat.
Transfer several fibers to a slide for inspection under the microscope.
Select a fiber which is straight and does not have pointed ends, but is squarely
cut. As described in the preceding experiment, attach the lesser end of the fiber
to a glass needle, and attach the latter to the body tube of the microscope.
Concentrating the Test Substance on the End of the Fiber and Precipitating
the Sulfide. Transfer about 0.2 pI of bismuth test solution (10 mg Bi per m1)
to the center of a slide, and insert the end of the cotton fiber as outlined
in the preceding experiment. Make certain that not more than the very
Expt.59 Working on Textile Fibers and Wires 191

end of the fiber is in continuous contact with the evaporating drop. When
evaporation is complete, raise the fiber with the body tube, exchange
the slide for another one carrying a large drop of sodium sulfide-hydroxide
reagent diluted with 25 volumes of water, and lower the end of the fiber
into this solution.
Rinsing the Fiber and Microscopic Examination. Mter a few seconds,
lift the fiber out of the sulfide solution. Remove the capillary carrying
the fiber from the tube of the microscope, and place it upon a slide, Fig. 57.
Place upon the fiber a cover slip with a large drop of water hanging on
the underside of it. Examine the fiber under the microscope. The black
Bi 2Ss should be clearly visible at the end of the fiber when using reflected
light and a white background.
Oxidation of the Sulfide and Precipitation of Chromate. Grasp the
capillary and pull the fiber out from under the cover slip; this will not be
difficult if there is enough water between the two glass plates. Insert
the end of the glass needle with the fiber attached into the gas space of
a bromine water b?ttle, and keep it there for about 3 minutes. Do not
touch the neck of the bottle with the fiber when inserting and withdrawing
it; if the neck is dangerously narrow, pour some bromine vapor into a
small bottle with wide mouth.
Again place the capillary with the fiber upon a clean, dry slide, and
cover the fiber with a dry cover slip. Inspect it under the microscope
with reflected light and a dark background. The end of the fiber should
be white or colorless if the sulfide has been completely oxidized to sulfate.
Deposit a large drop (0.05 ml) of O.I-F K 2Cr20 7 on the slide so that
it touches the cover slip and is drawn into the space between slide and
cover glass. When the fiber has become immersed in the solution, withdraw
it by means of the capillary to which it is attached, and transfer it to a
fresh slide. Place a cover slip upon the fiber, and add a large drop of water
so that the fiber becomes immersed. The yellow (BiO)2Cr207 is very difficult
to recognize. Try reflected light with white, black, and violet background.
Reduction to Metallic Bismuth. Focus the end of the fiber, and use
reflected light and a white background. Add a large drop of freshly prepared
stannite reagent to one edge of the cover slip, and remove water at the
opposite edge by touching to it a piece of filter paper. When the stannite
solution reaches the fiber, black metallic bismuth becomes visible at its end.

Experiment 59
Bead Test tor Cobalt (152)
L. I., 30 ng Co.
Platinum "ire, 0.05- to O.I-mm diameter, 25 mm long: fuse one end of the
straight "\\ire into a wide glass capillary of 4-cm length, which serves for handle. -
Borax, granular; xylene.
192 Techniques of the Submilligram Scale Expt.59

Scatter some borax upon a microscope slide, and select a kernel of

about I-mm diameter (0.5-,u1 volume) for the experiment. By means of
a platinum loop or a capillary pipet, treat it with 0.2,u1 of cobalt test solution
diluted with 9 volumes of water to contain 1 mg Cojml. Then heat the
platinum wire in the edge of a non-luminous Bunsen flame to incandescence,
and take up the treated kernel of borax by touching it with the end of
the hot wire. Adjust the Bunsen flame to a height of about ~ cm, and fuse
the borax by heating the center portion of the wire in the edge of the flame
and about 5 mm above the orifice of the barrel. The bead tends to move
to the cooler parts of the wire. Thus, it may be kept at the point of the
wire or driven back-and-forth over the length of the wire by proper applica-
tion of heat.
When the borax has been fused to a clear glass, drive the bead to the
end of the wire, and then allow it to cool. Put a large drop of xylene or
water near one end of a microscope slide, and lay the capillary with the
wire so upon the slide that the bead becomes immersed in the liquid.
Hold the slide over a brightly illuminated sheet of white paper; the blue
coloration of the bead may be observed with a magnifying glass.
For microscopic inspection, focus first upon the wire with a low magnifica-
tion. Then follow the wire to its end to bring the bead into the field of
vision. Use the front lens of the condenser to concentrate the light on
the bead, and change to a stronger objective if necessary. If the bead
should be too small, heat it again, and add more borax by touching a kernel
with the hot bead.
A blank test may be carried out on a second wire, and the two beads
may be immersed into the same drop and viewed side by side in the field
of the microscope. Very faint colorations may be made visible by drawing
the bead into a fiber (767). The wire with the bead is introduced into a
capillary of soft glass which is then cautiously heated at the point where
the· bead is located. When the borax glass melts, the wire is withdrawn.
The capillary is heated, and when it has collapsed somewhat .so that the
borax glass fills its bore, it is drawn out to a length of a few centimeters
of narrow capillary with a borax glass core. This fine capillary is cut
evenly at both ends and mounted as a coloriscopic capillary, Expt.55.
Immersion oil should be added at the base, and a fragment of a cover
slip should be placed with immersion liquid upon the top of the capillary.
In this manner, the blue color should become visible with 3 ng Co.
For cleaning the wire, again fuse the bead, and throw the molten bead
off the wire by a flip with the hand. Then take up a relatively large amount
of borax, fuse it to a bead, drive it back-and-forth over the wire, and flip
it off. Repeat this once. If this is desired, the borax glass may be
completely removed from the wire by finally placing it for 30 minutes into
dilute HCl.
Expt.60 Work on the Micr:>gram Scale 193

Experiment 60
Luminescence Test for Bismuth, Antimony, and Manganese (604, 1006)
Platinum loop or microspatula: hammer flat the end of a wire of about
0.3-mm diameter; two such wires may be fused side by side into the end of one
glass handle. - Hydrogen, pure: if prepared from zinc and acid, pass it through
a cotton filter; CaCOa, powder, of highest purity.
Lime containing a trace of bismuth, antimony, or manganese gives a
luminescence of characteristic color when touched by a hydrogen flame
so that it does not get hot enough to become incandescent. The test should
be performed in a dark room if small traces of the metals are to be discovered.
Furthermore, the tube supplying the hydrogen should have a vitreous
silica or porcelain tip so that the hydrogen flame does not become luminous.
Perlormanee of Test. In a dark corner, obtain a small hydrogen flame,
not more than 5 mm high, burning from the orifice of a capillary which
may be drawn out from glass tubing of high softening point. Olean the
platinum loop or spatula by dipping into HOI and ignition until it does
no longer color a Bunsen flame. Make a thin paste of OaOOa and water,
and take a small portion of it upon the loop or spatula. First dry and then
ignite in the hydrogen flame. By means of a capillary pipet or a platinum
hook or loop, add about 0.31'1 of diluted bismuth test solution (1 volume
of 10 mg Bi per ml diluted with 10 volumes of 3-F HNO a) to the white
erust of calcined lime upon the tool. Dry and ignite lightly. Allow to
cool, and then approach the lower edge of the flame with the preparation.
A deep sky-blue luminescence will emanate from the parts of the lime
which have received bismuth solution when they first meet the flame.
When the chalk begins to glow, the luminescence is displaced by the yellow
incandescence.
If two spatulas are mounted close together in one handle, one is made
to carry the blank test which may then be carried out and repeated side
by side with the test. If the handle is mounted in a test tube, Fig. 40,
the preparation may be kept for demonstration purposes.
The thermoluminescence is a light greenish blue with antimony and a
deep yellow with manganese. The tests seem to be specific.

Work on the Microgram Scale

The technique described in the following has been developed for work
with solid samples of 0.5 to l,ug (139, 141, 330, 411, 433, 435, 770, 888).
It should be applicable without much change to work on the sub microgram
and possibly the nanogram scale; it has been used on the submilligram scale,
i. e., for the investigation of solid samples of 0.1 mg to several micrograms
weight. In the latter instance, the apparatus is simplified (770, 772).
The capacity of the capillary cones is increased, and since evaporation is
Benedetti-Pichler. Identification 13
194 Work on the Microgram Scale

to be less feared with somewhat larger volumes of solutions, the capillary

cones may be mounted upon simple manipulators without the use of a
moist chamber. The pipet may be directly inserted into a syringe control,
and the latter may be mounted on a second simple manipulator facing
the first. The elaborate microscope with rotating mechanical stage may be
replaced by a binocular microscope of the Greenough type or by a binocular
magnifier (90).
The essential feature of the technique is the use of mechanical and
optical aids for the handling of substances. These aids are a great convenience
on the submilligram scale, but they become a necessity on the microgram
scale. Consequently, it seems proper to describe the technique as it has
been developed for work on the microgram scale and to leave the simplifica-
tions to the ingenuity of the experimenter who desires to apply it to some-
what larger amounts of material. Even on the microgram scale, apparatus
and technique permit various modifications for which the reader may
be referred to the literature (90, 903, 770, 772).
Most of the work is performed in tiny centrifuge cones of l-,ul capacity,
whereby the technique is essentially the same as in working with the
microcone. Because of the small volumes which must be handled, it is
necessary to perform transfers with the aid of mechanical devices while
observing through a low-power microscope. Furthermore, solutions of
considerably less than l,ul volume evaporate quickly when exposed to
air which is not saturated with water vapor. Thus, most of the work is
performed inside a moist chamber mounted upon the stage of the micro-
scope. Small volumes of reagents and wash liquids are held ready within
this chamber, and a mechanically operated pipet which is inserted through
the open side of the chamber serves for the transfer of liquids.
By taking one million times smaller masses and volumes of sample,
reagents, and solutions than on the gram scale, one may follow any tested
procedure or scheme of separation with complete assurance of success.
Because of the mechanical and optical aids, all operations may be performed
with ease and confidence. By projecting the microscopic images upon a
screen, Dr. CEFOLA has repeatedly demonstrated precipitations, filtrations,
and distillations to audiences of sixty to several hundred people on occasion
of meetings of the Metropolitan Microchemical Society of New York (1280),
the New York Section of the American Chemical Society (1281), and the
In-Service Training Course of the City of New York (1282). The usefulness has
been demonstrated in the investigation of transuranium elements (772); the
scale of work renders radiation hazards bearable; explosions become harmless.

Apparatus
Microscope. The stand should be that of microscopes used by biologists
for micromanipulation under high magnification. It should be equipped
 Apparatus 195

with a built-in revolving and centerable mechanical stage without excessive

superstructure and mechanical motions permitting displacements of 6 to
9 cm in two directions. The fine adjustment coming with such stands is
rarely needed; the rack-and-pinion device for the coarse adjustment should
permit tightening so that the body tube will keep its position when it
carries heavy auxiliary apparatus. A revolving nosepiece which provides
for individual centering of the objectives is desirable. The optical equipment
should consist of bright-field condenser which can be focused on a plane
10 mm above the stage, three objectives with magnifications of 5, 10, and
20 diameters, and one micrometer eyepiece with magnification 5 X. The
rulings of the eyepiece micrometer should occupy at least two thirds of
the diameter of the field. A totally reflecting prism to be used above
the eyepiece is converupnt for micro projection.
Manipulator (770, 903). A simple manipulator with three rack-and-
pinion motions, each permitting a displacement of about 10 cm, is needed;
in addition, rotation around the vertical axis of the manipulator is desirable.
A very sturdy construction is essential so that vibrations will not originate
in the manipulator, that would be transmitted to the pipetholder and
magnified at the tip of the pipet.
Base. Microscope and manipulator are placed side by side upon a
board and secured in their correct positions by means of small wooden
blocks as indicated in Fig. 58. To determine the proper distance between
the microscope and the manipulator, the following proc~dure is recommended.
The left-right motion of the manipulator is operated to advance the clamp b
all the way toward the microscope while the two other pinions engage at the
centers of their racks. A capillary of 15-cm length is then inserted into
the pipet holder which is placed into the clamp b so that it is held close
to the capillary. The correct position of the manipulator is found by moving
it until the free end of the capillary is about I cm to the left (referring to
Fig. 58) of the optic axis of the microscope and about I cm above the stage.
To this end, the height of the microscope or manipulator may have to
be adjusted by means of additional boards. When the correct positions
have been obtained, the positions of the apparatus are secured with the
wooden blocks. A cover for the whole assembly should be made to fit
the base board; it will permit to have the equipment always ready for use.
If a strong source is available for illumination with transmitted light,
an image of one to several feet in diameter may be obtained at the most
convenient location. Use of the screen image gives less eye strain than
viewing through the eyepiece; in addition, the height of the table may
be selected to give a maximum of comfort during the operation of the
various contrcl3. If use of a strong source of light is not practical, it may
still be possible to project an image of 5 to 10 cm in diameter on a piece
of white Bristol board which may be mounted inside a black box and a
13*
196 Work on the Microgram Scale

short distance above the eyepiece as indicated in Fig. 58. The box should
be mounted in a manner that it always remains at a fixed distance above
the eyepiece and may be simply swung aside whenever direct observation
becomes necessary to get the image with reflected light.
The eyepiece micrometer is focused upon the screen by adjustment
of the eye lens after focusing the object. If necessary, the positions of

Fig. 58. Assembly for Work on the Microgram Scale

the microscope tube and of the eye lens may be alternatingly adjusted
until object and scale appear simultaneously with sharp outlines. The
value of the scale division of the micrometer scale will be approximately
the same as when looking into the eyepiece. It is preferable, however,
to calibrate the micrometer under the conditions of use. Of course, the
eyepiece micrometer is not essential when projecting since the dimensions
may be measured with a millimeter rule on the screen image.
lliumination. Two lamps are required. One is placed in front of the
microscope to send light in the direction of arrow T to the mirror. The
 Apparatus 197

second lamp is placed to the left of the microscope so that it may send
light in the direction of arrow R horizontally into the moist chamber.
This lamp should permit collecting the light into a narrow pencil of not
more than 6-mm diameter at the focus. Switches for both lamps should
be mounted in a handy location so that observation with transmitted
light and with reflected light may be used in quick succession.
Pipet Holder. It is now possible to obtain micrometer syringes which
may be mounted in the clamp of the manipulator and into which the
micropipet may be directly inserted (903). The separation of the plunger
device from the manipulator will be of advantage, however, if the latter
is not very rigid; in addition, the pipet holder makes also possible to use
the very simple device of regulating the pressure with a levelling bulb (437).

1----~Ir===~~
=L----------_,i----------L-~
Tip S/Ja// ';Jo~ - Sll.7ni: MeniSC(lS

Fig. 59. Micropipet and Pipet Holder. w metal washer; r rubber gasket; bore and
thickness of the tubes are exaggerated

The pipet holder shown in Fig. 59 is a metal tube of 1- to 2-mm bore,

about 4 mm in outer diameter and about 12 cm long, which is fitted with
rubber gaskets r, metal washers w, and screw caps at both ends. The
rubber gasket consists of flexible rubber tubing which is compressed by
the advancing concave face of the metal washer w. If fine-drawn copper
tubing is used to make the connection to the plunger control d, Fig. 58,
it may be inserted into the pipet holder just like the shank of the pipet,
or it may be soldered to it. As an alternative, the connection may be made
with plastic tubing which is stiff enough to maintain the capacity of its
bore during manipulation. Even rubber tubing is suitable if the tube
contains air connected with a larger air space the pressure of which is
adjusted (437).
Obviously, the pipet holder may consist of a glass tubing of suitable
dimensions, and the shaft of the pipet may be inserted with the use of
sealing wax.
Plunger Control. The device indicated in Fig. 58 was fashioned after
the buret control of JOHNSON and SHREWSBURY (876). Any similar device
will serve, and so will a micrometer operated syringe. The control, the
connecting tubing, and the pipet holder are filled with water that has been
198 Work on the Microgram Scale

boiled to remove air. The water is used while still lukewarm, and the filling
is done so that air bubbles are excluded.
Micropipet. The micropipet proper consists of the shaft and the tip,
Fig. 59. The small volumes of solutions will rarely reach the taper. The
shank of 6- to lO-cm length contains the air cushion which separates the
solution in the pipet proper from the water used for transferring the pressure
from the plunger control.
The micropipets are made from soft-glass tubing. Capillaries of 20-cm
length and 0.5- to I-mm outside diameter to fit the pipet holder are drawn
out in the middle to get a quick taper and a shaft that gradually tapers to
an orifice of 30- to 40jUm diameter. The drawing is best done by mechanical
devices as they have been described by Du BOIS (1080) and RACHELE (424).
A microflame the size of a pinhead is needed for drawing micropipets
by hand. The capillary is grasped, between thumb and index finger, at
two points 3 cm left and right of its middle. The hands are steadied by
resting the outer, fleshy parts of the palms on the bench top, and the
middle of the capillary is brought over the flame. A steady horizontal
pull is applied immediately so that the drawing starts when the glass begins
to soften. By rolling the edges of the palms on the bench, the capillary
is removed from the flame and at the same time symmetrically pulled (903),
which takes a fraction of a second. The fine capillary may be 3 to 5 cm long,
and it may snap in the center. The pipet is placed upon a slide and inspected
under the microscope to learn what variations of technique will give a
pipet of proper dimensions. The shaft should be 5 to 8 mm long. The tip
is often too fine and fused shut; it may be snipped off with scissors to
ontain the desired orifice of 30 to 40/-tm.
Moist and Dry Chamber. Fig. 60 shows the top and front views of a
chamber similar to that designed by CEFOLA (433). The bottom is formed
by a glass plate, 60 mm X 68 mm. The two long sides are formed by bars,
6 mm X 67 mm X 11 mm high, which may be made of metal or plastic
and which are cemented to the base plate. A narrow strip of thin glass
plate, II mm X 53 mm, fits into vertical grooves of the bars and forms
the back of the cell. The top of the cell is a thin glass plate, 52 mm X 63 mm,
which is placed upon the bars. To obtain a humid atmosphere, the sides
of the cell are lined with cotton b, Fig. 60, which is kept wet with water.
The dry chamber does not receive the cotton lining.
Capillary Cones. The cone, Fig. 60a, has a capacity of approximately
0.6/-tl and is made of a thin-walled capillary of about 0.8-mm bore. Gloves
are worn so that the capillaries will not be touched by fingers. The capillaries
are freshly drawn out from clean glass tubing, Expt. 19, and cut into pieces
6 to 10 cm long. After the bore has been checked, the middle of a piece
of capillary is heated in a microflame or in the edge of a non-luminous
Bunsen flame until the glass fuses together to form an elongated bead.
 Apparatus 199

The bead is withdrawn from the flame and, after brief delay to permit
some cooling of the thin glass at both ends of the bead, the bead is drawn
out to a rod of about O.3-mm diameter and 4-cm length. This should
give two cones, the taper of the bores of which should be quite blunt as
shown in Fig. 60a. Obviously, the procedure may be repeated at suitable
intervals to give a string of 4-cm rods separated by pieces of the capillary
of about 5- to 10-mm length. Cutting at the proper places gives a number
of capillary cones with handles of about 2-cm length. The cutting should
be done with a sharp tool so that little pressure is needed (903). The length

.J

{j

c
Fig. 60. Moist Chamber Ready for Chemical Work. a capillary cone; b cotton lining;
c carrier slide; d micropipet; e reagent cone

of the cone proper should be 2 mm, and half of this length should be
occupied by the blunt taper. The finished capillary cones are collected
in a screw-cap vial. Like any other apparatus used in the moist chamber,
they should never be touched with the fingers since the fingerprints would
develop into a pattern of small droplets, which greatly interferes with the
microscopical observation of the contents.
Reagent Containers. A large number of reagent containers is prepared
from an assortment of clean capillary tubing of 0.3- to I-mm uniform bore.
The bore of each capillary is determined, Expt.22. The outside of the
capillary is wiped with a moist and then a dry cloth; hereafter it is handled
with gloves. About 2 cm from one end, the capillary is fused to a bead
which is then drawn out to a thin rod of about 4-cm length; no attention
need be paid to the shape of the bore of the taper. The fusing and drawing
is repeated so that a 2-cm length of the original capillary is left between
200 Work on the Microgram Scale

each pair of rods. Cutting at the centers of the rods and of the capillaries
gives reagent capillaries of I-cm length with handles of 2-cm length. They
are stored in a screw-cap vial, the label of which indicates the cross-section
area of the bore and the number of eyepiece micrometer divisions correspond-
ing to the length of the capil~ary holding 1 nl.
Measuring Capillaries, Fig. 609, permit a more accurate measuring of
very small volumes, but are rarely needed if reagent containers of sufficiently
narrow bore are used. They are thin-walled capillaries of 0.05- to 0.2-mm
uniform, known bore and 2- to 3-cm length, which are sealed at one end.
Like reagent containers, they are kept in labelled screw-cap vials.

Fig. 61. Condenser Rod

Carriers. At least two should be available. The carrier c, Fig. 60,

consists of a strip of plate glass, 25 mm X 35 mm X 7 mm thick. It may
be made of several thin plates of glass by cementing them together with
Canada balsam. One half of the top surface is coated with a layer of
petrolatum, about 1 mm thick, as indicated by the dotted area in Fig. 60.
To get a smooth coating, the carrier is slightly heated. EL-BADRY and
WILSON (903) make the carrier of plastic; a deep groove, parallel to the
front edge and 5 mm from it, is cut into the base surface so that a rubber
band placed into it will not touch the surface supporting the carrier and
interfere with the sliding of the latter. The light rubber band is stretched
around the carrier so that the handles of the capillary cones and reagent
containers may be inserted between the band and the top surface of the
carrier.
Condenser Rod. Side and top views of this device are shown in Fig. 61.
It consists of a short piece of glass rod which tapers to a fine thread. The
latter is cut to provide a tiny platform d for the performance of tests.
A strong beam of light is sent into the rod through its base f. Most of the
 Apparatus 201

light is collected in thread c and emerges at d to give efficient illumination

for the observation of the test.
Soft glass is used, which has as little color as possible. A rod of 4- to
5-mm diameter is drawn out to a thinner rod of about 2-mm diameter so
that a quick taper is obtained between a and b,Fig. 6l. About lO mm from
the taper, rod b is drawn out to a thread 0.1- to 0.3-mm in diameter, which
is broken 2 cm from the taper. The thick part of the rod is cut to give a
a length of 10 to 12 mm; the rod is scratched with a file and then broken
while part a is held with pliers. It is important that an even surface is
t.
obtained at The thread c is bent at a right angle by cautiously approaching
it with a micro£lame while the rod is held horizontally; the thread bends
by its own weight. Then, thread c is dipped into molten paraffin nearly
up to the bend and slowly withdrawn. When the paraffin on the thread
has solidified, the thread is scratched 5 mm above the bend and broken

===£22==2======"\
1; a. c d
! I

c:::::-::.-----
lJ

------
Fig. 62. Heating Element

off with the aid of forceps. The break should be clean and at a right angle
to the axis of the thread. Rod a is cemented to a cover slip e so that thread c
assumes a perpendicular position when e is supported horizontally. Finally,
t
a fragment of a cover slip is attached to surface with Canada balsam
to obtain a plane surface; obviously, this is not necessary if the circular
face t is given a high polish.
The film of paraffin on the cylindrical surface of the thread confines
solutions to the glass surface of the cross section. The resulting circular
platforms of 0.1-, 0.2-, or 0.3-mm diameter have areas of 0.008, 0.03, and
0.07 mm2 which would support hemispherical droplets of 0.25-, 2-, and
7-nl volume.
Reservoir of Water for Cleaning Micropipets. The reservoir t, Fig. 58,
has the shape of a short pinchcock (MOHR) buret. A glass bead in the
rubber tubing controls the outflow. The inside of the rubber tubing is
cleaned with brush and soap solution before it is attached to the tube.
The reservoir is filled with distilled water and then covered with an inverted
vial to keep dust out.
Heating Element. A 5-cm length of No. 24 copper-nickel alloy wire
(0.5-mm diameter, 0.7 ohm) is bent in the center to give the shape of a V.
The point of the V is pressed closely together by means of pliers, and it
is then carefully filed down until the cross section of the wire is reduced
202 Work on the Microgram Scale

to one-third of the original (408) as shown in the enlarged drawing of the

point b, Fig. 62. The ends of the wire are soldered to ordinary insulated
copper wire of the type used in radio work. The insulated wire is forced
through a glass tube e of 10- to 15-cm length and of such bore that the
wire fits tightly. Insulating tape is applied at d to prevent the wire from
twisting around in the tube. The leads are connected to a variable trans-
former which is plugged into the a.-c. line; in general, not more than 5 volts
need be supplied to the heating element. The glass tube may be fastened
in the clamp f of the manipulator.
Forceps. Two forceps, preferably of stainless steel and with polished
flat tips, are needed. One of them is provided with gripping surfaces of cork.
Chips of cork, 1 mm thick, are sliced from a stopper with a razor blade.
The tips of the forceps are slightly heated, and some Kronig glass cement
(1 weight white beeswax and 4 weights of rosin) is applied by touching
the stick of cement to the hot metal. The cement melts, and the slices
of cork are placed upon the treated tips, which are squeezed together
until the cement has solidified. The excess of cork is trimmed off with
the razor blade.
Gloves. A pair of thin cotton or silk gloves is kept in a covered jar
or in an envelope so that they remain meticulously clean.

Technique
Manipulation under the microscope does not require skill. Needed are
some practice and the ability of organizing the work so that nothing will
be missing when the cell has been assembled for the performance of an
operation.
Mounting the Micropipet. The plunger of the pressure device is advanced
until a drop of water appears at the opening of the pipet holder. The
shank of the micropipet is inserted into the opening, and the screw cap is
made tight. If necessary, the plunger is advanced until the meniscus of
the hydraulic water in the shank is seen about 3 cm in front of the opening
of the pipet holder, Fig. 59.
Filling Reagent Containers. Clear liquid reagent is taken up into a
capillary pipet having a fine tip of about l-cm length. A reagent container
of a bore assuring the desired precision of measurement (above p. 199)
is selected, and the tip of the capillary pipet is inserted so that the orifice
is about 6 mm inside the container. By blowing with the mouth, liquid
is gradually expelled from the pipet which is at the same time gradually
withdrawn so that the container is filled close to its opening, Fig. 60e.
Assembling the Chamber. All apparatus, reagents, and wash liquids
needed for an operation or a brief series of operations are assembled in
the chamber which takes the role of the laboratory and contains bench
and reagent shelf. If several reagents are needed, a pencil sketch should
 Technique 203

be prepared in which the containers are labelled; the containers are then
arranged according to this plan. The required number of capillary cones 1,
reagent containers 2, and measuring capillaries 9 are assembled upon the
carrier c side by side. The handles of the reagent containers and capillary
cones and the sealed ends of the measuring capillaries are pushed into the
layer of vaseline (or under the rubber band). The openings of all tubes
are brought into a straight line parallel to the edge of the carrier, which
facilitates later manipulations. Without delay, the carrier is then placed
into the chamber, Fig. 60, which may already hold a condenser rod or
other needed devices. A droplet of water is deposited upon the bottom
plate of the chamber before putting down a piece of equipment such as
the carrier or the condenser rod. The water spreads between the glass
surfaces, and the surface tension holds the apparatus in its assigned place.
If necessary, water is added from a washbottle to the cotton lining, and
the chamber is closed by putting on the cover plate.
The chamber is then clamped into the mechanical stage of the micro-
scope so that the opening of the chamber faces the manipulator on the
right of the microscope and the controls of the mechanical stage are on
the left-hand side of the microscope.
Introducing the Micropipet into the Chamber. The manipulator is
swung around so that the micropipet is parallel to the side walls of the
chamber. It is then introduced into the chamber by operating the controls
-of the manipulator while watching with the unaided eye.
Handling Solid Particles. Particles of 1 flg mass and less may be weighed
with suitable balances (156). To this end, the particle would be supported
by a metal foil or a fragment of a cover slip. If the supporting surface is
mounted inclined to the horizontal in the chamber or above the chamber,
the particle may be transferred to a capillary cone by the technique of
Expt. 60, or dissolved in situ by adding a measured amount of solvent
with a micropipet which may then be used to transfer the solution to a
capillary cone for further investigation.
In general, it will suffice for qualitative analysis when the mass of solid
particles is computed from an estimate of its density and measurement
of its dimensions under the microscope with the use of the eyepiece microm-
eter. The material for investigation or solid reagent is scattered upon
the surface of a slide or cover slip, 2.5 cm square. By means of a short
piece of glass rod (with or without the use of cement), the slide is mounted
in an inclined position in or on top of the chamber, Fig. 68c. Obviously,
solids which are not hygroscopic may be mounted inside a moist chamber.
The microscope is focused upon the particles, and a particle of suitable
size is selected with the aid of the eyepiece micrometer, picked up with
the point of a needle or the sealed tip of a micropipet, and transferred
into a capillary cone. Obviously, the needle is inserted into the pipet holder
204 Work on the Microgram Scale

or the clamp of the manipulator. The tip is treated to receive a film of

glycerol, sebum, or other suitable adhesive.
Using the lowest available magnification, the selected particle is brought
into the center of the field of vision and sharply focused. The point of
the needle is brought into the field of vision by using the motions of the
manipulator while observing with the unaided eye from the directions.
2 o'clock and 5 o'clock (the needle coming from 3 o'clock). Without
changing the focus of the microscope, the point of the needle is brought
into sharp focus with the vertical motion of the manipulator. After the
point of the needle has been brought close to the selected particle, one
may change to a higher magnification for the observation of the contact.
As a rule, a loose particle will stick to the needle point that touches it
and may thus be transferred to any desired location.
Fusions. One end of a straight platinum wire of not more than 0.05-mm
diameter and 3-cm length is fused into the end of a glass capillary which
latter is then suitably mounted on a stand (or on a manipulator) so that
the wire is held horizontally and at a right angle to the axis of the micro-
scope.
. The (binocular) microscope is provided with 2.5 X or 5 X objective(s)
and tilted into the horizontal position. If it must be used in the vertical
position, the front lens(es) should be protected by a horizontally mounted
glass pane inserted just below it (them).
The free end of the platinum wire is brought into the field of vision.
A gas or hydrogen microflame, 1 mm in height, Expt.19 (glass tubing
clamped to a manipulator or to a stand that glides upon the top of the bench),
is moved into position so that it heates the wire a short distance (5 to
10 mm) from the free end. The end of the wire should get hot but not
incandescent. Small particles of the selected flux are placed upon the
end of a microspatula (flattened end of a 0.5-mm platinum wire) or upon
a narrow slide. The tool is moved by hand, while watching through the
microscope, so that a particle of flux touches the hot end of the wire.
The particle will adhere to the wire, and it is fused by bringing the flame
gradually closer to the end of the wire. Additional particles of flux are
added to the melt (or to the hot bead) until the bead has the desired size.
The dimensions of the bead are measured with the eyepiece micrometer.
The bead will approach the shape of a rotation ellipsoid with radius a
vertical and axis b parallel to the wire. If r is the radius of the wire
and d is the density of the flux, the weight of the bead is approximately
given by 2 b d (2 a 2 - 3 rI); this is nearly the same as 6 a3 d which is the
weight of a sphere of flux of radius a, Table IV.
When the required amount of flux has been collected in the bead,
various procedures may be used for adding the sample: (a) a solid particle
may be added to the molten bead by using the technique of Expt. 61 and
 Technique 205

a platinum needle operated with a manipulator; (b) the bead may be

allowed to cool (possibly moistened with water or glycerol) and used to
pick up the particle; (c) the particle located upon a slide or spatula may
be transferred by touching it to the bead; or (d) a powder or precipitate
may be made into a slurry and transferred to the cold bead by means
of a micropipet with wide tip.
Table IV. "Equatorial" Diameter 0/ Bead
as Function 0/ Its Weight
Assumed are: density of flux = 2.5 gjml
and diameter of supporting wire = 0.05 mm
Weight of Bead Diameter 2a
pg pm

1 80
2 100
4 130
6 145
8 160
10 175
20 220

Finally the bead is dried, if necessary, by approaching it with the flame

which is first applied to the wire at about 20 mm from the free end with
the bead. By heating the wire closer to the bead, the latter is fused while
observing through the microscope. The advice on pp. 11], 322ff. should be
suitably applied. The heating is continued until the bead becomes clear
or the reaction stops. For dissolving the cold bead, it is transferred into
the solvent held ready in a capillary cone in a moist chamber, see next
section.
An electrically heated loop has been described by KOCH, MALISSA,
and DITGES (570). Finally, it should be possible to perform pyrosulfate
fusions in a capillary cone of vitreous silica.
Inserting a Tool into a Capillary Cone or a Capillary. Using the motions
of the mechanical stage, the capillary cone (capillary) is moved into the
location of the diameter 3 to 90' clock with the opening facing the manipulator.
With transmitted light and a magnification of 40 to 60 diameters, the
opening of the capillary cone is focused so that the outer contours of the
walls of the capillary appear as perfectly sharp straight lines (point of
the taper of the sealed end is in sharp focus). This setting of the microscope
is retained during the following operations, and this assures that the tool
will be located in the axial plane of the capillary in focus.
The capillary cone is withdrawn toward 9 o'clock so that it occupies
only one third of the field, Fig. 63. The manipulator is rotated to get the
tool (needle or micropipet) parallel to the axis of the capillary. The tool
206 Work on the Microgram Scale

is introduced into the chamber, and its tip is brought close to the opening
of the capillary by using the motions of the manipulator and watching
with unaided eye from 2 o'clock and 5 o'clock. At this stage, the microscope
will give a blurred image of the tool. Without touching the adjustment
of the microscope, the tool is brought into focus with the vertical motion
of the manipulator; this brings the tool into the mid plane of the capillary.
To line it up with the axis of the capillary, the stage may have to be rotated
somewhat, and the side motions of the manipulator are used for the final
adjustment. Providing that stage and manipulator are set so that the
side motions follow the diameters 6 to 12 o'clock and 3 to 9 o'clock, the
tool may now be made to enter and leave the capillary along its axis by
operating the 3 to 9 o'clock motion of either the manipulator or the
mechanical stage. Fig. 63 shows capillary and tool not as they are actually

Fig. 63. Micropipet Ready to Enter a Capillary

situated in space, but as they appear when looking into a monocular

microscope.
Measuring Liquids. All measuring is done with the micrometer scale.
The basic measuring devices are the reagent containers and the measuring
capillaries, but the measuring may also be done in the micropipet and in
the capillary cone. Since the volume determinations are based upon
computation of the capacity of the dry containers, the volumes actually
delivered may be about 10% smaller than expected, even when the liquid
is very slowly withdrawn. The fraction of the liquid remaining upon
the walls seems to reach a maximum of 20% with capillaries of 0.2-mm
diameter, but decreases for narrower and wider tubes (437). Obviously,
these uncertainties are avoided by the Teddol treatment of EL-BADRY
and WILSON (423).
Liquid reagents are measured out of the reagent containers, the cross-
section areas of the bore of which are known. The length of the liquid
column which represents the desired volume is readily computed in terms
of the divisions of the eyepiece micrometer for the magnification in use.
The eyepiece is rotated to make the scale appear in the image of the reagent
container, Fig. 64a. The left hand grasps the control of the mechanical
stage, and the right hand that of the plunger device. The plunger is slightly
 Technique 207

advanced to obtain some pressure in the micropipet, which will not dissipate
immediately if the orifice of the pipet is fine enough. Then, without delay,
the reagent container is moved forward by means of the mechanical stage
until the opening of the micropipet is immersed in the liquid somewhat
beyond that length of liquid which is to be taken into the pipet. As a rule,
solution enters the micropipet as soon as its tip touches the liquid. The
entering liquid is completely expelled from the pipet by advancing the
plunger until the tiny meniscus arrives at the orifice of the tip. Then the
position of the reagent container is adjusted until the meniscus in the
reagent container coincides with a convenient division of the micrometer
scale such as 10 in Fig. 64a. Then suction is cautiously applied by with-

1==l=I=I=i=!=H+@+tttti 1111111

20

Fig. 64. Measuring Liquids. a in the reagent conta iner; b in the m easuring capillary

drawing the plunger from the pressure chamber so that the meniscus
recedes slowly into the reagent container. When the meniscus in the
reagent container has travelled through the desired number of scale divisions
and arrived at the predetermined point of the scale, the tip of the pipet
is quickly withdrawn from the liquid, which may be done by moving
either the capillary or the pipet, preferably the former. The pipet now
contains the desired volume of liquid.
The pipet may now be calibrated if the volume of liquid is small enough
so that the meniscus in the shaft appears still on the micrometer scale
when the orifice is lined up with the zero mark or the edge of the field.
Obviously, use of low magnification permits extending the range of applica-
bility. The volume and the position of the meniscus in the shaft are recorded
for future use. If the bore of the shaft is uniform or tapers regularly, it
will be possible to crudely estimate fractions of the determined volume.
A reasonably correct estimate of the volume delivered by the pipet
may now be obtained by transferring the liquid to a dry measuring
capillary (437). The lat~er is moved into the center of the field, and the
micropipet is inserted so that the orifice of the tip moves into the capillary
208 Work on the Microgram Scale

a distance equal to the estimated length of the liquid column. The plunger
control is operated to transfer the liquid from the pipet to the measuring
capillary, and the tip of the pipet is withdrawn as the meniscus in the
capillary moves toward the opening of the latter, Fig.64b. It does not
matter whether or not the tip of the micropipet touches the wall of the
measuring capillary. When the plunger is advanced, a droplet forms at
the orifice of the pipet, which grows and finally fills the bore of the measuring
capillary. The two menisci in the capillary move apart as liquid is added.
Calibration of the pipet would require that its whole contents are
transferred to the measuring capillary. If, however, the latter is used for
the direct measuring of small volumes, then the micropipet is withdrawn

Fig. 65. Transferring Solutions to the Capillary Cone. a delivery of the bulk; b delivery
of the small amount of liquid left in the tip

when a drop of predetermined length has been obtained; the micropipet

is then emptied and hereafter again inserted into the drop in the measuring
capillary to take it up by a reversal of the above procedure for transfer
to the capillary cone in which it is needed.
When finally a measured volume of liquid has been transferred to a
capillary cone, this latter may be calibrated in turn by measuring the
distance from the meniscus to the point of the taper, Fig. 65a.
Transfer of Liquid to the Empty Capillary Cone. The microscope is
kept focused upon the point of the taper of the capillary cone. The tip
of the micropipet is advanced with the manipulator to a point just inside
the capillary cone, and then the cone is moved with the mechanical stage
until the orifice of the micropipet touches the wall of the cone close to
the point of the taper, Fig. 65a. Slight pressure is then cautiously applied
with the plunger device so that the contents of the pipet are slowly delivered
to the point of the cone. The capillary cone is slowly withdrawn so as to
keep the orifice' of the micropipet just below the meniscus. When the
meniscus in the micropipet approaches the orifice, the flow is stopped.
 Technique 209

The cone is withdrawn until the orifice of the micropipet is above the
meniscus of the liquid in the cone. Then the stage of the microscope or
the manipulator is rotated so that the opening of the pipet touches the
wall of the microcone close to the meniscus, Fig. 65b, and the remainder
of the liquid in the pipet tip is expelled gently.
If the micropipet is emptied carelessly so that air bubbles follow the
liquid, this will cause spatt~ring, and it is then advisable to collect all

-
liquid in the point by whirling in the centrifuge.
Emptying and Cleaning the Micropipet. The micropipet is completely
withdrawn from the chamber. A strip of filter paper is grasped with the
fingers by one end. The other end of it is touched to the orifice of the
micropipet while the plunger is slightly advanced. When the liquid has
been expelled, the meniscus of the hydraulic water in the shank of the
pipet is brought back to its proper position if necessary, 3 cm from the
pipet holder.
For rinsing the micropipet, the valve of the water reservoir t, Fig. 58,
is operated so that a large drop of water forms at the tip of the outlet.
The tip of the micropipet is inserted into this drop, and suction is applied
with the plunger device. When the water has advanced to the taper of
the pipet, the latter is withdrawn from the hanging drop, and its contents
are expelled by touching the filter paper with the opening and advancing
the plunger. The rinsing is repeated twice, each time with a fresh drop
of water. Finally the position of the meniscus of the hydraulic water in
the shank is again checked and corrected, if necessary.
Of course, the cleaning may require a solvent different from water,
which maybe supplied hanging at the end of a stirring rod.
Adding Reagents to Solutions in the Capillary Cone. The micropipet
containing the reagent is brought close to the opening of the capillary
cone which is then advanced so that the orifice of the micropipet is close
to the surface of the liquid already in the cone. Slight pressure is now
applied with the plunger device. This assures that the outflow of reagents
starts as soon as the opening of the pipet makes contact with the liquid
in the capillary cone.
The outflow of reagent is stopped when the meniscus in the shaft
of the micropipet gets close to the orifice. This is the time for using the
micropipet for stirring, see below. The tip of the pipet is then withdrawn
from the liquid and touched to the side of the capillary cone for the delivery
of the small remainder of the reagent, Fig. 65b.
The effect of the reagent upon the contents of the capillary cone is
observed with reflected 'as well as transmitted light.
Solid reagents are ground to a fine powder and sprinkled upon a glass
slide which may be mounted inside or on top of the chamber as the hygro-
Benedettl·Plchler, Identification
210 Work on the Microgram Scale

scopicity of the reagent and the atmosphere in the chamber dictate, Fig. 68.
The measuring and transfer of solid particles is described on p. 203.
Stirring in the Capillary Cone. A satisfactory stirring effect may be
obtained by moving the shaft of the micropipet, which is sufficiently sealed
by the presence of the last trace of reagent in the tip, through the contents
of the cone by means of the mechanical stage or the manipulator. Very
efficient stirring is obtained by plucking with the finger the copper tubing
connecting pipet holder and plunger device while the pipet tip is immersed
in the contents of the cone.
Treating the Contents of the Capillary Cone with Gaseous Reagents.
A glass tubing is drawn out to a capillary of about I-mm bore, which is

b c

Fig. 66. Working in the Capillary Cone. a centrifuging; b heating;

c saturating with gas

then provided with a constriction c and bent as shown in Fig. 66c. A plug
of cotton is placed into the tube b. The capillary is cut squarely at a and
the capillary cone with the material to be treated is introduced. To this
end, a piece of hot cloth or lens paper is held ready on a heating block
or in a drying oven. The chamber is opened for a moment, and the capillary
cone is lifted out with cork-tipped forceps. Petrolatum adhering to the
handle is removed with the warm cloth before inserting the cone into the
capillary.
With the capillary cone at d, Fig. 66c, the capillary is drawn out to
a fine tip at a. The wide tube b is connected to the gas supply, and the
gas is allowed to flow through the capillary until the air is displaced.
Then the point of tip a is sealed. The capillary may be left connected to
the gas supply or it may be fused off at the constriction c. In either
instance, it is possible to immerse the capillary into a suitable bath so
that the gas may act at a chosen temperature. To retrieve the capillary
cone with its contents, the capillary is cut at d so that the handle may
be grasped and the cone returned to the carrier in the chamber.
 Technique 211

Heating Solutions in the Capillary Cone. A capillary of 0.7- to I-mm

bore and about lO-cm length is drawn out to a fine tip at one end. A small
volume of water is allowed to enter the tip which is then sealed shut.
The capillary cone is removed from the carrier, and its handle is cleaned
as described in the preceding section. The cone is then introduced, handle
first, into the wide capillary and made to slide down to the sealed end,
Fig. 66b. The open end b is sealed, and the capillary, with the cone inside,
is placed into a bath or heating block having the required temperature.
The presence of water in the wide capillary and the confinement of the
capillary cone into a small space prevent the evaporation of the solution
in the cone. To retrieve the latter, the outer capillary is allowed to cool
to room temperature and then cut at a.
Collecting Precipitates in Capillary Cones. A capillary of 0.6- to 0.7 -mm
bore is cut to a length of 6 to 7 cm. The capillary cone containing the
precipitate is grasped at the handle and introduced, opening of the cone
first, into one end of the capillary which is held horizontally and then
introduced without change of position into a microcone, Fig. 66a. The
latter is then placed into the shell of a centrifuge. Mter whirling, the
microcone is again held horizontally, and the capillary is withdrawn.
The handle of the capillary cone usually protrudes from the opening of
the capillary and may be grasped for the return of the capillary cone to
the carrier.
The wide capillary facilitates the handling of the capillary cone, and
it limits the air space around the capillary cone sufficiently to prevent
excessive evaporation of the solution in the cone.
Separation of Solution and Precipitate. Mter whirling in the centrifuge,
the capillary cone is returned to its former position on the carrier in the
moist chamber and focused in transmitted light. The tip of the micropipet
is inserted into the capillary cone and advanced until its opening touches
the wall of the capillary cone at a short distance in front of the surface
of the precipitate. Suction is cautiously applied by means of the plunger
device so that the clear solution is taken slowly into the micropipet. Either
reflected or transmitted light may be used for observation. The last portion
of the solution is taken up very slowly so that the operation may be stopped
as soon as air begins to enter the tip of the micropipet. The capillary cone
with the precipitate is withdrawn. If needed, the contents of the micropipet
are transferred to another capillary cone; if the centrifugate is to be rejected,
the micropipet is withdrawn from the chamber, and its contents are
discharged on a strip of filter paper. The micropipet may be rinsed once
with water, but this may be omitted if it is used for washing the precipitate.
Washing the Precipitate. The required amount of the proper wash liquid
is taken into the micropipet from the reagent container on the carrier.
The wash liquid is delivered upon the wall of the capillary cone at a short
14"
212 Work on the Microgram Scale

distance from the precipitate. The liquid is slowly expelled from the
micropipet so that the precipitate is not stirred up. If the washing is
to be repeated with a like volume of liquid, the distance from the point
of the capillary cone to the meniscus is measured and recorded for future use.
The wash liquid is left in contact with the precipitate for one minute,
whereafter it is again removed as described above for the centrifugate.
The washing is combined with the centrifugate or rejected depending
upon the requirements of the procedure.
Estimation of the Volume of the Precipitate. The reasoning from the
observed cross-section area of a precipitate to its weight assumes that
precipitates obtained with known and with unknown amounts of estimated
substance are compacted to the same extent by means of the centrifuge.
The use of a high-speed centrifuge is desirable, but whirling for 1 minute

;..-- - /1 - --.;

Fig. 67. Estimation of the Volume of a Precipitate

with 2000 to 3000 r. p. m. is usually satisfactory. Adherence to a standard

procedure is also necessary during precipitation since a variation of the
particle size affects the volume of the compacted precipitate.
Transmitted light is used for observation, and the microscope is focused
on the point of the taper of the capillary cone. The precipitate often
occupies an area A similar to that indicated by shading in Fig. 67. The
volume of the corresponding ideal truncated cone of diameter B at the
base and height H - h is given by 0.26 (J32 H - 62 h). A less accurate
but simpler procedure is to imagine a sphere of the same volume as the
cone of precipitate. A circle representing the cross section .of this sphere
is shown in Fig. 67 . Its diameter D may be estimated and expressed in
micrometers rather than divisions of the micrometer scale, the value of
which depends upon the magnification used. The mass of the precipitate
and the estimated constituent is in direct proportion to the volume 0.52 D3
and is directly proportional to the cube of the diameter of the imagined
sphere. Obviously, a mistake of 10% in selecting the diameter of the
sphere will cause an error of 30 % in the derived volume and mass.
Evaporation in the Capillary Cone. Placing the cone into a desiccator
or under a small bell jar together with some drying agent will suffice for
the evaporation of most aqueous solutions. Hydrogen chloride may be
 Technique 213

absorbed by a bead of sodium hydroxide; ammonia, by potassium acid

sulfate or a drop of sulfuric acid placed near the capillary cone. Obviously,
it is not permissible that liquid in the capillary cone is made to boil, but
evaporation may be hastened by laying the capillary cone upon a watch
glass or into a dish heated to a suitable temperature. EL-BADRY and
WILSON (903) use a heating block. Creeping may be prevented by applying
the "hot point" (see under Distillation) in front of the opening of the
capillary cone. If the evaporation is performed in a dry chamber, the
process may be observed with the microscope.
Expelling Dissolved Gases. Because of the small expenditure of time
involved, evaporation to dryness and dissolution of the residue in solvent
of the desired composition is recommended for the elimination of gases
and other volatile constituents.
Distillation from Capillary Cone to Capillary Cone. A distiHing capillary
is obtained by giving the cylindrical part of a wide capillary cone a length

Fig. 68. Distillation from Capillary Cone to Capillary Cone; approx. nat. size

of 5 to 6 mm; it must be wide enough to permit insertion of the capillary

cone containing the charge to be distilled. The capillary cone has to be
calibrated (p. 208) if the distillation shall be stopped or a fraction shall
be removed when the charge has evaporated down to a certain volume.
Suitable allowance for the meniscus must be made when estimating the
volume equal h3 tg2 (ex/2) from the angle ex and the height h of the cone
filled by the liquid.
The charge to be distilled is placed into a capillary cone which is then
spun in the centrifuge to assure that all material is collected in the point.
The distilling capillary a is mounted in a dry chamber, and the capillary
cone b with the charge is inserted into its opening as shown in Figs. 68
and 69, whereupon the chamber is closed and mounted on the mechanical
stage of the microscope.
The heating element, Fig. 62, is inserted into the clamp of the manipulator
and introduced into the dry chamber. As a rule, it is impossible to get
the whole distilling capillary into the field of vision as shown in Fig. 69.
Thus, the capillary cone in end a is focused in transmitted light, and the
mechanical stage and the eyepiece scale are adjusted until the point of
the taper of the capillary cone coincides with a convenient scale division
of the micrometer. The "hot point" h of the heating element is brought
214 Work on the Microgram Scale

into the level of the distilling capillary and moved to a point about 1 mm
to one side of the capillary cone d containing the charge.
The current is turned on, and the voltage is slowly stepped up until
the meniscus in the capillary cone begins to recede slowly toward the
point of the taper. Before this happens, small droplets may be seen to
form along the walls of the distilling capillary. The rate of movement of
the meniscus in the capillary cone indicates the rate of evaporation which
must be quite slow and may be controlled by either changing the position
of the hot point or by regulating the voltage supplied to the heating element.
It may happen that a gas bubble forms at the point of the taper, begins

.Fig. 69. Distillation from Capillary Cone to Capillary Cone as Seen with the
Microscope; schematic

to grow, and pushes the liquid contents of the capillary cone toward its
opening. It then becomes necessary to reduce the rate of heating to keep
the bubble to a small size in the point of the taper.
When the receding meniscus reaches the predetermined mark (15 nl in
Fig. 69), the hot point is quickly withdrawn from the chamber and the
current is turned off. The whole length of the distilling capillary is inspected
under the microscope, and it will be found that the distillate collects at
the end b, Fig. 69. The capillary cone with the distillation residue may
be transferred to another distilling capillary for the collecting of another
fraction or it may be transferred to the carrier of a moist chamber for
other treatment of its contents. The distilling capillary is spun in the
centrifuge to collect all distillate at the sealed end. It is then scratched
at Ii. distance of 3 mm from the taper, and the unwanted portion of the
capillary is snapped off by means of cork-tipped forceps. The part containing
the, distillate is mounted on the carrier of a moist chamber, and the liquid
 Technique 215

is transferred to a capillary cone of standard dimensions by means of the

micropipet.
Determination of Boiling Point and Boiling Range (902). A uniformly
tapering cone is drawn at the end of a wide capillary of several centimeter
length. The liquid is collected in the point of the cone and confined. there
by a drop of mercury as shown in Fig. 70a. The capillary is then heated
upon a microscope hot stage which has been calibrated by melting point
(or still better, boiling point) determinations. The boiling point is taken

Cone, 2.0,at

-~
IX - 10·

~
'I, (Jlmm
a,

~ b

.a.'.J.o.·~\G~I '#' - L____________________ S c

... kz -

.a.-M··-c~~I·__wn__ t _______________
$
~
d

° .fmm

Fig. 70. Determination of Boiling Point and Boiling Range

as the temperature at which the bulk of the liquid suddenly vaporizes or

condenses. Boiling ranges may be determined by plotting the temperature
against the decrease of the volume of the liquid, which is computed as
a fraction or percentage of the initial volume of the liquid.
Capillary tubing of soft glass, 1- to 2-mm outer diameter, and lO-cm
length is heated near one end over a microflame so that the glass fuses
to a solid bead which is then drawn out to a rod. The bore at the sealed
end of the capillary should assume the shape of a uniform taper with
an angle IX of 10 to 45 degrees at the apex. It is suggested to sort the
capillaries according to the angle IX and to store them in stoppered test tubes.
The rods are fused off about 5 mm from the taper, and the cylindrical
portion of the capillaries should remain 8 cm long. Capillaries with blunt
tapers will be satisfactory for boiling point determinations. A taper of
216 Work on the Microgram Scale

20 degrees will permit more precise measurement of the decrease of volume

if a boiling range shall be determined. Lengths and capacities of the
tapered portions of a capillary of 1.2-mm bore are shown and listed in
Fig. 70. Table V will aid in selecting the taper according to the volume
of liquid available. It should be hardly necessary to add that the capillaries
must be clean and dry.

Table V. Height of Circular Cone for a Given Volume and Angle of Taper
Volume of Cone Height h of Cone in Millimeter if the Angle IX at the Apex is
,uI 45 degrees 30 degrees i 20 degrees 10 degrees

0.001 0.18 0.24 0.32 0.51

0.01 0.39 0.52 0.69 1.1
0.1 0.84 1.1 1.5 2.4

The liquid sample is taken into a micropipet which is then completely

withdrawn from the dry chamber. Without delay, the selected capillary
is pushed over shaft and shank of the micropipet until the tip touches
the inside wall of the taper of the capillary. The liquid is expelled from
the pipet, while watching through a magnifier, and the pipet is simultaneously
withdrawn. The liquid is collected in the point of the capillary by whirling
in the centrifuge.
A micropipet with a relatively wide opening is filled with mercury.
The capillary is pushed over the micropipet until the tip of the latter is
close to the liquid in the taper of the capillary. While watching through
a magnifier, a mercury column of a few millimeter length is transferred to
the capillary so that only a small air space is left between the liquid and
the mercury, Fig. 70a.
The capillary is bent at the middle to give it a U shape to fit the micro-
scope hot stage. Care is taken that the capillary is not sealed during
bending and that the closed end containing the sample is not heated.
The capillary must be about 8 cm long so that the drop of mercury is not
expelled from it when the sample vaporizes. About 6 cm of a capillary
of I-mm inner diameter will suffice to hold the vapor of 100/kg of liquid,
but a capillary of 1.6-mm bore should be taken if 200/kg shall be vaporized.
The capillary is placed upon the hot stage so that the liquid sample
is in the field of vision. An aluminum plate, 15 mm X 50 mm X 2 mm thick,
having a 3-mm slot cut lengthwise is placed around the sealed end of
the capillary. The glass baffle of the KOFLER (98) hot stage is placed
crosswise over the capillary, and the top plate is put in place as for melting
point determinations. The microscope is focused upon the meniscus of
the liquid, which is observed while the stage is heated. The meniscus of
the mercury usually moves out of the microscopic field when the boiling
 Technique 217

point is approached, Fig. 70b, but its position is of no consequence since

the volume of the vapor is not measured. The volume of the liquid decreases
slightly before the boiling point is reached. The boiling point of a pure
substance is then characterized by the complete evaporation within 1 or
2 degrees and the complete disappearance of the liquid phase, Fig. 70d.
On cooling, the boiling point is indicated by the appearance of the meniscus;
in addition, the mercury drop may be seen to snap back into the field
of vision.
If a boiling range is determined, the magnification should be chosen
so that the meniscus of the liquid sample appears near the end of the microm-
eter scale when the apex of the taper is at the zero division. In the
instance of fine tapers, the micrometer scale may not be able to cover
the length of the liquid cone. It is then advisable to measure distances
from a suitable reference point (particle of dust upon the outside of the
capillary) as suggested in Fig. 70a. The temperature of the stage is raised
very slowly, and the distance of the meniscus of the sample from the apex
of the cone (or from the reference mark) is measured at the start and at
frequent intervals. The volume of liquid vaporized in each given temperature
interval would have to be computed as the volume of the frustum of a
cone, 1.05 (tg2 1)(.[2) • (h I3 - h 23 ) = 0.26· (hI DI2 - h2 D 22), which would re-
quire also knowledge of either the angle I)(. or measurement of the diameters
of base and top of the frustum, Fig. 70c. The measurement of 1)(., D I , and D2
is not necessary, however, since the percentage of the total volume is given
by 100 (h I 3 - h 23 )/H3, where H is the length of the cone of liquid at the
start of the distillation.
Especially when known mixtures are treated for comparison, the
boiling curves obtained with 50 to 200 nl of liquid should make it possible
to recognize, for example, characteristic petroleum fractions and products.
Crystal Precipitation Upon the Condenser Rod. The limits of identification
of sEde tests may be improved by reducing the volume of the test drops
to 10 or 1 nl. Whenever, for lack of material, the crystals of the precipitate
remain so small that their shape can no longer be discerned under the
microscope, one is simply limited to the criteria used with test tube tests:
appearance or disappearance of phases and color phenomena. Under such
conditions, it will be advisable to give preference to tests giving characteristi-
cally and intensely colored precipitates which will assure a high degree
of specificity and sensitivity.
A small volume of test solution, which will cover the area of the platform
of the condenser rod, is measured off and taken into the micropipet. The
latter is then withdrawn to a distance of 1 cm from the axis of the micro-
scope, and the condenser rod is brought into the field while observing
with the unaided eye. After focusing upon the platform, one turns off
the lamp sending light to the mirror of the microscope and concentrates
218 Work on the Microgram Scale

a strong beam of light from the lateral lamp upon the base of the condenser
rod. The lamp is adjusted until the platform is brightly illuminated while
the rest of the field of vision remains dark. This illumination facilitates
the observation of colors and precipitates, but it is desirable to add the
general illumination with transmitted light during manipulations.
By means of the manipulator and while observing with the unaided
eye from the side, the tip of the micropipet is moved close to the platform
of the condenser rod, which is in the focus of the microscope. While viewing
through the eyepiece, the tip of the micropipet is brought into sharp focus
and moved close to the platform with the manipulator. Using the controls
of the manipulator only, the tip of the micropipet is slightly raised so that
its image becomes somewhat blurred; it is then advanced horizontally
so that the opening of the tip appears at 3 o'clock above the side of the
platform. The pipet is then lowered so that the opening touches the top
surface of the platform and its outlines become sharp. The plunger is
slowly advanced to expel the contents of the micropipet upon the platform.
When the first bubble of air appears at the opening of the pipet, the latter
is first raised and then completely withdrawn from the moist chamber
for cleaning.
Solid reagent may be transferred to the test drop upon the platform
by the standard procedure, p. 203; a particle of about 1 p,m in diameter
will represent a suitable quantity. Liquid reagents are measured in and
added with the micropipet. The transfer to the platform is performed
as described for the test solution, but it is advisable to apply slight pressure
by means of the plunger device and to cut off the general illumination
with transmitted light just before lowering the tip of the micropipet into
the test solution upon the platform. The pressure upon the solution in
the micropipet causes it to flow out of the pipet as soon as the tip touches
the test solution, "and separation of precipitate inside the tip is prevented.
Absence of the general illumination improves the conditions for the
observation of the effect of the reagent upon the test solution.
When a bubble of air appears at the orifice of the pipet, the micropipet
is removed and completely withdrawn from the moist chamber for cleaning.
The test is observed with the illumination furnished by the condenser rod.
The platform may finally be cleaned, after removing the condenser rod
from the moist chamber, by dipping it into suitable solvents and rinsing
with distilled water. If this does not give satisfactory results, a new platform
is readily obtained by cutting the glass thread close below the old platform.
Performance of Spot Tests (903). EL-BADRY and WILSON provide
upon the carrier a thin glass rod bent into the form of aU, 30 mm long
and 6 mm wide. A cotton thread is placed across the open end of the U
and attached with a suitable cement (gum or starch paste) to the glass.
The thread is impregnated with reagent by applying to the middle portion
Expt.61 Technique 219

of the thread the reagent solution hanging as a drop from the end of a
"glass rod or capillary. By means of the glass U, the thread is then mounted
on the carrier, and a measured volume of solution to be tested is added
from the micropipet. The latter is manipulated so that the opening of
its tip is in contact with the middle of the thread. If necessary, slight
pressure is then applied to transfer the solution to the thread at the rate
at which it is absorbed.
Obviously, the technique might be refined by using a single fiber in
place of the thread. The acro technique, too, appears promising, p. 130.
Preserving Solutions and Precipitates. Solutions and precipitates may
be preserved for days by sealing the capillary cone containing them into
a capillary as shown in Fig. 66b.
Starting a New Series of Procedures. It is not practical to plan too
far ahead and to crowd the canier in: the chamber with a large number
of apparatus and reagent containers. Thus it becomes necessary, from
time to time, to clear the chamber of used apparatus and reagents no
longer needed and to assemble the material needed in the next step.
Obviously, it is convenient to have at least two chambers available so that
the capillary cones with the material under investigation may remain
undisturbed in one of them while the other is prepared for the continuation
of the investigation.

Experiment 61
Mechanical Separation of the Components of a Powder (154)
Preferably a Greenough-type binocular microscope; needle mounted in a
handle, 12 em long. - Mixture of 1% bone black and 99% A120 3 • The particle
size of the two ingredients should be checked with a total magnification of about
20 diameters before they are mixed. If the particles are too small, it will be
impossible to separate them as told below. - Glycerol.
The isolation of a material by mechanical collecting assumes that
the particles may be recognized by their color, shape, fluorescence, or
behavior in polarized light. It is also necessary that the boundaries of
the individual grains are clearly recognizable and that aggregates are either
absent or readily separable into their components.
With a camel's-hair brush, dust a small amount of the carbon-alumina
mixture onto a clean microscope slide. Place the slide under the microscope
and focus the particles of the powder with a low-power objective which
has a working distance of at least 20 mm. Use reflected light and a colored
background.
For collecting the particles, place a droplet of about 0.5,tt1 upon the
center of a clean slide, and cover it with a I-inch (25 mm) watch glass
to prevent its evaporation.
220 Work on the Microgram Scale Expt.61

Examine the powder under the microscope and select a black particle
which is not closely surrounded by white ones. By moving the slide, get
this particle into the center of the field. Moisten the point of a sewing needle
(glass thread, platinum wire of 0.05- to O.I-mm diameter) mounted in a
suitable long handle by rubbing some glycerol on the back of the hand
and drawing the point of the needle across the treated area. With the
hand resting on the stage of the microscope, hold the needle 45 degrees
inclined to the horizontal and insert its point half way between the slide
and the front lens of the objective. Look into the microscope and move
the needle in small horizontal circles until a blurred image of its point
appears in the field of vision. Then bring the point of the needle straight
down upon the particle to be removed. Touch the particle and lift it out
of the preparation.
While the hand which holds the needle remains resting upon the stage,
use the other hand to exchange the slide with the drop of water for that
with the powder mixture. Remove the watch glass and focus the edge
of the drop of water. While observing through the microscope, insert
the point of the needle into the space above the drop and then bring it
straight down into the drop. The particle floats off when the needle touches
the water.
Cover the droplet of water with the watch glass, bring the powder
mixture under the microscope, and remove another black particle, etc.
Repeat the procedure until 8 to 10 black particles are collected in the drop.
If some white particles have been carried along, allow the drop to evaporate
and remove the white particles to another drop of water. The black particles
will finally be located in an area which is small enough so that one may
proceed to chemical treatment without further preliminary work.
Some advice may be added. The needle should always be held as nearly
vertical as possible. In this manner it is possible to avoid touching other
particles in addition to the one selected. If the working distance of the
objective is shorter than 15 mm, it becomes necessary to bend the end of
the needle. Any visible amount of glycerol on the needle will defeat its
purpose; when the particle is touched, the glycerol flows down over the
particle and spreads to a drop on the slide. It is then impossible to pick
up the particle floating in the glycerol. Of course, the selection of glycerol
and water is arbitrary. As a general rule, the adhesive used for the treatment
of the needle must be readily soluble in the liquid in which the particles
are collected.
Du FREsNE (978) mentions the possibility of developing on the particles
an electric charge by warming them gently under an incandescent lamp
for 15 minutes, whereafter they will adhere to a metallic needle. For
the removal df particles from oil (immersion liquids used for the determina-
tion of the refractive index), he uses a sewing needle with a heat sink
Expt.62 Technique 221

(triangular sheet of copper of about 5-g weight) soldered to it near the

wooden handle. The tool is chilled by immersion in liquid nitrogen or
dry ice-acetone mixture. When the point of the needle is then brought
close to the immersed particle, the oil freezes around the needle, and the
occluded particle may be lifted out and transferred to another location
where the oil is allowed to melt. The oil may be removed with a capillary
pipet, and the particle may be washed with a suitable solvent like acetone.
The practicability of the described techniques depends upon the working
distance of the objective rather than upon the total magnification. Using
a 10 X objective and projection, one should be able to apply it with total
magnifications of several hundred diameters. The use of simple manipulators
for tasks of this sort (409) renders the work less tedious, removes most of
the fear that some untoward accident might prevent a successful conclusion
of the task, and permits the experimenter to assume a more detached
attitude which is very helpful in arriving at sound decisions.

Experiment 62
Precipitation of Silver Dichromate Upon the Platform of the Oondenser Rod
L. I., 1 ng Ag or less.
Equipment for working on the microgram scale. - Mixture of 8 volumes
of silver test solution (10 mg Ag per ml) with 1 volume 16-F HN0 3 ; saturated
solution of ~Cr207.
Assemble in a moist chamber a condenser rod (0.2- to 0.3-mm diameter)
and a carrier with a measuring capillary and two rather narrow reagent
containers, one containing the acidified silver test solution and the other,
the dichromate solution. With the micropipet, take up somewhat more
than 1 nl of the silver solution and transfer just 1.1 nl of it to the measuring
capillary; 10% of this volume will probably be left behind when the solution
is then transferred to the platform of the condenser rod. For this task,
empty the micropipet on a strip of filter paper; do not rinse it, but keep
it wet with the solution. Take the solution from the measuring capillary
into the micropipet and record the distance from the tip to the meniscus
in the shaft, which corresponds to a volume of 1 nl. Transfer the solution
to the platform of the condenser rod.
Rinse the micropipet, and then take into it about 1 nl dichromate
solution and add it to the silver solution on the platform. As a rule, some
crystals will be observed, which have the shape and color characteristic
()f the silver dichromate. Clean the micropipet.
The test may be repeated with a solution of 1 mg Ag per ml, which
is 2-F in HNOs. Evaporation of the test drop upon the platform may
be obtained by removing the cover of the moist chamber for a short time.
222 Work on the Microgram Scale Expt.63

Experiment 63
Estimation of the Quantities of Arsenic and Antimony in a Solution of
Unknown Concentration
Sample: about 0.1 pI of a mixture of 2 to 30 ml antimony stock solution
(50 mg Sb per ml), 50 ml 12-F HCI, 2 to 20 ml arsenate stock solution (50 mg As
per ml), and water to make 100 ml.
Equipment for working on the microgram scale; buzzer, p. 73; heating
element. - Antimony test solution (10 mg Sb per ml); supply of hydrogen
sulfide; KBr0 3 , finely powdered solid; 9-F HBr; 3-F H 3 P03 ; buffered silver
nitrate (1 g AgN0 3 and 7.7 g ammonium acetate in 6 ml glacial acetic acid and
200 ml water); quinine-iodide reagent: before use mix equal volumes of solutions A
(1 g quinine hydrochloride dissolved in 50 ml warm water; the cold solution
treated with 0.2 ml 6-F HCI) and B (2 g KI in 50 ml water).
A dry chamber is needed in addition to a moist chamber. In the latter.
assemble a condenser rod and a carrier with 4 capillary cones and 3 reagent
containers containing the sample solution, antimony test solution, and
water.
Measure out of the reagent containers 10 nl sample solution, 2 nl antimony
test solution, and 10 nl antimony test solution; transfer each portion to
a separate capillary cone. To correct for the liquid remaining behind,
first wet the shaft of the micropipet with the solution to be taken, and
then remove 10% more from the reagent container than indicated above:
11, 2.2, and 11 n1.
Add 100 nl water to the solution in each capillary cone, and then
saturate each with hydrogen sulfide. Heat the mixtures of solution and
precipitate for half a minute at 60° to 80° C, and then allow to stand for
one hour at room temperature. Centrifuge simultaneously all three
precipitates and estimate their volumes both from the dimensions of the
cone or frustum occupied and the diameter of the equivalent sphere.
Remove and reject the centrifugate in the instance of the sample.
In the dry chamber assemble a capillary cone which fits into a distilling
capillary held ready in a vial and 3 reagent containers with 12-F HCl,
9-F HBr, and 3-F H aP0 3 • Transfer the capillary cone containing the
sulfides of arsenic and antimony to the dry chamber. Treat the sulfides
with 30 nl 12-F HCl; seal the capillary cone into a dry capillary and
immerse for 15 seconds into a water bath of 70° C. Agitate the mixture
with a buzzer and inspect under the microscope to observe that only part
of the precipitate has dissolved. Return the capillary cone to the dry
chamber and treat the contents with KBr0 3.
Grind some KBr03 to a fine powder and sprinkle it on a small slide
which is mounted upon the cover of the dry chamber, Fig. 68. Transfer
individual small particles of the salt to the mixture in the capillary cone,
mixing after each addition, until all sulfide is dissolved and only sulfur
Expt.63 Technique 223

remains behind. Centrifuge and transfer the clear solution to the capillary
cone which fits into the distilling capillary. Wash the residue of sulfur
with one 40-nl portion of 12-F HCI, and transfer the washing to the solution
which is then treated with 10 nl HBr and 20 nl HsPO s' Seal the capillary
cone into a dry capillary and immerse for 5 to 10 seconds into a bath of
80 to 90 C. Cool with tap water, centrifuge, remove the capillary cone
0 0

from the capillary, and insert it into the distilling capillary which is then
mounted in the dry chamber.
Using the graduated scale of the rotating stage, measure angle lX of
the taper of the bore of the capillary cone containing the solution to be
distilled, and compute the lengths h of the taper corresponding to 10-nl and
15-nl volume, hS = v[(tg lX[2)2. Express h in divisions of the micrometer
scale. Focus the meniscus in the capillary cone and heat cautiously until
the volume of the liquid has been reduced to 15 nl. Then stop heating.
Withdraw the capillary cone from the distilling capillary and place
it upon the carrier. Add 10 nl 12-F HCI to the distillation residue, return
the capillary cone to the distilling capillary, and continue the distillation
until 10 nlliquid is left in the capillary cone. Stop the heating, and transfer
the capillary cone with the distillation residue to the moist chamber.
Centrifuge the distilling capillary to collect its contents at the sealed end;
return it to the dry chamber and transfer its contents to a standard capillary
cone, which is then transferred to the moist chamber.
Treat the residue of the distillation with 50 nl water, and dilute the
distillate with water to a volume of approximately 100 nl. Precipitate
the antimony with H 2 S as before. Treat the distillate containing the
arsenic with 40 nl 12-F HCI before saturating with H 2 S. The arsenic sulfide
has a tendency to become colloidal. To avoid this, heat the mixture of
solution and precipitate by immersion for 30 to 45 seconds into a bath
of 60 0 to 70 0 C, and then agitate by means of the buzzer. Without opening
the sealed capillary, inspect the contents of the capillary cone under the
microscope. If the arsenic sulfide is properly flocculated, collect it into
the point of the capillary cone by centrifuging simultaneously with the
antimony sulfide. Mount the capillary cones with the precipitates in the
moist chamber and estimate their volumes from the dimensions of cone
or frustum and equivalent sphere. Using the volumes of the known quantities
of Sb 2 S a (28 ng and 140 ng Sb 2 Ss), compute the weights of the mixture
of the sulfides, the antimony sulfide, and the arsenic sulfide from the sample.
Finally, compute the weights of arsenic and of antimony in 10 nl of the
sample solution.
The colors of the sulfide precipitates are observed with reflected light.
To confirm the antimony, introduce into the mo:st chamber reagent
containers with 12-F HCI and with quinine-iodide reagent. Remove
the centrifugate and treat the remaining antimony sulfide with 10 nl
224 Work on the Microgram Scale Expt.64

12-F HCl. Seal the capillary cone into a capillary containing some 12-F HCI,
and heat at 70° C until only a white residue of sulfur remains. Spin in
the centrifuge, return the capillary cone to the moist chamber, and transfer
1 to 2 nl of the clear solution to the platform of the condenser rod. Add a
like volume of quinine-iodide reagent. A yellow precipitate confirms the
presence of antimony; the corresponding bismuth precipitate is orange
or brown.
To confirm the presence of arsenic, transfer the capillary cone with
the precipitate to the dry chamber which has been cleared of other apparatus.
Add a reagent container with 6-F NH a. Close the chamber and transfer
10 nl of the NHa to the arsenic sulfide. Bring the hot point of the heating
device in front of the opening of the capillary cone and evaporate just
to dryness. Open the chamber, remove the reagent container with ammonia
and replace it by containers with 16-F HNOa and 0.5-F HNOa.
Treat the residue in the capillary cone with 5 nl 16-F HNOs and again
evaporate. Treat the residue with 5 nl 0.5-F HNO s, centrifuge, and transfer
the capillary cone to the moist chamber. Clean the condenser rod, and
place upon the carrier a reagent container with buffered silver solution.
Transfer 1 to 2 nl of the clear arsenate solution to the platform of the
condenser rod and treat with 1 nl of the buffered silver solution. A brown
precipitate of AgsAsO, confirms the presence of arsenic.

Additional Practice Experiments for the Chosen Scale

of Work
Experiment 64
Study of Ohemical Behavior
Test solutions of silver nitrate, mercuric nitrate, mercurous nitrate, lead
nitrate, bismuth nitrate, cupric nitrate, cadmium nitrate, stannous chloride,
stannic chloride, antimony trichloride, and arsenate containing 1 mg of the
metal per 1 ml solution; see Appendix. Reagent solutions: NaOH, Na2S, KI,
and (NH4)2S5.
The purpose of the experiment is the collecting of information for the
efficient identification of the cations in Expts. 65 and 66. Perform the
task with the apparatus and technique chosen for confirmatory tests.
Most convenient is the use of a spot plate or glass slide on darkly colored
paper and stirring rods, capillaries, capillary pipets, or loops for the adding
of reagent. Depending upon the amount of material available for study
and upon preference, the tests may be carried out in test tubes, micro
cones, or capillary cones.
a) Test a portion, droplet, of the solution of each metal ion as follows.
Mixing after each addition, add small increments of 2-F NaOH until the
mixture is just alkaline, test with litmus or pH paper, then add 4-F NaOH
Expt. 60, 66 Additional Practice Experiments for the Chosen Scale of Work 225

to bring the NaOH concentration of the mixture to about 1 formal. Record

the observations in table form. Add to the mixture 4-F Na 2S to bring
the concentration of Na 2S to about 2 formal and mix. If a clear solution
results, add 4-F H 2S0 4 until the mixture is slightly acid. Record the
observations and, if they do not agree with your expectations, repeat the
experiments and compare with the literature.
b) Test a portion of the solution of each metal ion by adding 2-F NHs
until just alkaline, then 6-F NHs to bring its concentration in the mixture
to 2 formal. Add 6-F (NH4)2S5 to bring its concentration to 1 formal,
warm, and stir. If a clear solution results, add 4-F H 2 S0 4 to make it
slightly acid. Record as under (a).
c) Treat all but the chloride solutions with 2-F HOI until they are
acid and have a chloride concentration of 0.5 formal or more. Record as
under (a).
d) Treat all t'3st solutions first with small increments of O.l-F KI and
finally with an excess bringing the iodide concentration to 0.05 formal.
Acidify the arsenate solution with HOI if it is alkaline. Record as under (a).

Experiment 65
Analysis of Two Unknown Solutions
Samples: Each solution contains only one metal. ion of those studied in
Expt. 64 in the concentration of 1 to lO mg per ml. D~pending upon the chosen
sc::tle of work, 0.5 ml, 0.1 ml, lO ,ul, or less solution is given. Recommended
is work on the submilligram scale with about 2,u1 of each unknown solution
handed over in a capillary pipet.
Apparatus and reagents as in Expt. 64.
Determine the approximate pH of the unknown solution. Use a scheme
of testing, which will permit recognition of the ion present with the use
of two, at the most three, portions of the unknown solution so that enough
material is left for three or four confirmatory tests. Oonfirm the finding
with the use of at least one slide test and one spot test and try to collect
some evidence (preparations, drawings, or photographs) that may be kept
for the purpose of demonstration. If necessary, refresh the memory by
first performing the confirmatory tests with a like volume of known test
solution of similar concentration and composition before risking the loss
of a portion of the unknown.

Experiment 66
Identification of Simple Compounds of the Common Metals of the Hydrogen
Sulfide Gronp
Two different Samples: Dapending upon the chosen scale of work a suitably
small amount (lOO mg, 10 mg, I mg, or less) of a solid compound of one of the
metals considered in Expts.64 and 65. Work upon the submicrogram scale is
Benedetti· Pichler, Identification 15
226 Additional Practice Experiments for the Chosen Scale of Work Expt. 67

recommended, and this suggests samples of a total mass of about 0.3 to 1 mg,
most conveniently about 5 or 6 particles of 0.2- to 0.5-mm diameter, which
may be placed upon a microscope slide and covered with a small watch glass.
Apparatus and reagents as in Expt. 65. In addition, provision has to be
made for heating in various gases or upon charcoal, pp. 78 and 276, etc. A hand-
book of chemistry should be readily accessible.
Inspect with magnifying glass or microscope and make use of criteria
and tables given with the procedure of systematic analysis, P.9, etc.
Use color, shape, and behavior in polarized light. Depending upon amount
of material available, test by heating in open and closed tube, upon "char-
coal", or in a stream of air, hydrogen, etc., P. 23 to 30.
The information collected should exclude all but one or two metals
and leave a choice of a small number of possible compounds. Use the
residue and condensates obtained in heating tests to get a solution suited
for the orientation tests used in Expt. 65 and confirmatory tests. The
collected evidence should make it possible to select a suitable solvent and
to decide whether it should be applied to the original or to a certain residue.
Make it a habit to proceed as in Expt. 65; always collect material to be
used as evidence in future demonstrations.
Using tables on the properties of compounds and minerals in a handbook,
decide the identity of the sample. There should be material left for additional
tests.
If time permits, it may appear desirable to insert experiments with the
third analytical group (Fe-AI-Ni) corresponding to Expts. 64 to 66. In this
instance, testing with a magnet should be added and, if the amount of material
permits, bead tests.

Experiment 67
Identification of Simple Inorganic Compounds
Two different solid samples, each consisting of one inorganic compound, are
given in the amount specified in Expt. 66.
Apparatus as in Expt. 66 with the possible addition of wire for bead tests.
Reagents needed for the confirmatory tests for all common ions.
Proceed as in Expt. 66, but add testing with the magnet and the
performance of bead tests if working on the gram or centigram scale.
Try to guard against being misled by the effects of incidental impurities
by always making certain that the intensity of tests corresponds with
the amount of unknown taken. The latter may be estimated by measuring
under the microscope the dimensions of the particle taken and computing
the mass with a crude approximation of the density (about 4 glml for
compounds of heavy metals). In case of doubt, compare with a test obtained
with the appropriate quantity of the substance in question.
Expt. 68, 69, 70 Additional Practice Experiments for the Chosen Scale of Work 227

Experiment 68
Identification of Simple Oompounds
Two different solid samples, each consisting of one substance, are given in
the amount specified in Expt. 66.
Equipment as for Expt. 66 with the addition of apparatus for the observation
of transition temperatures and, when work is done on the gram or centigram
scale, wire for flame tests and possibly a simple spectroscope.
After studying the appearance and the behavior in polarized light,
heat in the melting point apparatus and determine the transition temper-
atures. Performance of the heating test with a very small amount of
material and observation under the microscope is recommended. An
ignition test with a small amount of material upon a wire must be carried
out before heating any appreciable quantity of an unknown substance
if there is no definite knowledge that the substance is not an explosive.
The residue may be used for the following tests.
If the collected evidence does not suggest a different approach, proceed
to heating in a current of air or oxygen (heating in closed and open tube)
and pass the gas coming from the reaction zone through Ba(OH)2' Continue
as suggested in the systematic procedure. Testing the solubility in water,
HCI, and organic solvents may give useful information.

Experiment 69
Identification of Simple Oompounds
Two different solid samples, each consisting of one substance, are given in
the amount specified in Expt. 66.
Equipment as for Expt. 68.
Proceed as in Expt.68, but after studying the appearance and the
behavior in polarized light, do not fail to test the hardness.

Experiment 70
Identification of Materials as They Occur in Nature, Industry, and Research
Any desired number of solid or liquid unknowns representing simple or
complex materials and mixtures of such.
Proceed as suggested in the systematic procedure.

15*
 Part II

Systematic Analysis

Choice of Materials and Cleaning

It is important to use reagents and apparatus which will not introduce
detectable amounts of the substances that are considered in the search.
In the search for majors and minors, distilled water and reagents satisfying
ACS specifications (24) will be suitable. Special precautions are required,
as a rule, when testing for traces.
The label of the manufacturer guarantees the quality of the reagent
only up to the time of the breaking of the original seal; from there on,
it is the responsibility of the user to prevent contamination (13). Organio
reagents are best kept in a dark closet since many of them deteriorate,
some quite rapidly, when exposed to light.
Reagent solutions should not be kept longer than for one year; they
should be discarded immediately if their appearance changes or a precipitate
separates. Small glass bottles with glass stoppers serve for acids, solutions
giving off corrosive vapors, strongly acid solutions, and organic solvents.
Glass bottles with rubber stoppers, which have been carefully freed of
coatings, are recommended for neutral and mildly acid solutions. Poly-
ethylene containers are suggested for distilled water, ammonia, ammoniacal
and strongly alkaline solutions, hydrofluoric acid and fluorides, and for
metallic mercury.
The material of apparatus must be considered in every phase of the
work including the preparation of reagent solutions and the storing o~
samples. Ordinary bottle glass is quite strongly attacked by water and
harmless aqueous solutions and may introduce unpredictable impurities
in addition to silica, alkalies, aluminum, iron, and calcium. The black
. glaze on porcelain may contain chromium, lead, cobalt, copper, manganese,
and iron. Pyrex glass is recommended because of its simple composition:
silica, B 2 0 a, alumina, sodium, 0.4 % potassium, and only traces of arsenic
and antimony. The safest materials are those which contain essentially
only one element: vitreous silica, platinum, silver, and nickel; the last
two are used for alkaline fusions. All of them may be slightly attacked,
 Choice of Materials and Cleaning 229

even when properly used, and may get into the reagents or the material
under investigation. New platinum ware may give off iron which should
be removed by heating with strong hydrochloric acid. The surface of old
platinum ware may be alloyed with other metals taken up as a consequence
of improper use. The permeability of platinum at high temperature
may permit flame gases (H2' S02) to react with the contents of the
apparatus.
Attention must also be paid to the cleaning of apparatus. Common
sense suggests cleaning immediately after use. The nature of the contamina-
tion is still known, and the proper solvents may be sele<;lted for its efficient
removal. The cleaning may require several steps which are in logical
order: (a) disposal of dangerous contents and removal of corrosive agents
by rinsing with solvent, usually water; (b) removal of greasy residues
by wiping with absorbent paper or rags; (c) cleaning with brush and water;
(d) use of solvents to remove residues escaping mechanical treatment:
acid for metallic, sulfidic, etc. mirrors; thiosulfate for silver halides, etc.;
(e) brushing with soap or detergent and water; (f) heating on the steam
bath with chromic-sulfuric acid for the removal of organic residues;
(g) thorough rinsing with tap water and distilled water; (h) wiping dry
the outside and possibly drying the interior by either draining, or a stream
of clean air and possibly heating.
Phosphates of detergent solutions and chromic ion from the chromic-
sulfuric acid may be adsorbed tenaciously on glass. In addition, apparatus
having been in general use may have adsorbed other substances or may
be contaminated by invisible residues of insoluble substances. Cleaning
may be tried by a succession of treatments: digestion with bromine water
for the oxidation of sulfur and sulfides; heating with sodium carbonate
solution for the conversion of insoluble compounds into carbonates; rinsing
for the removal of anions; followed by digestion with 6-F nitric acid and
rinsing. In the search for traces and in work with very dilute solutions,
however, it may be best to use new apparatus; even this should be rinsed
with 8-F nitric acid and water (13).
Especially when working on microscope slides or watch glasses, it
should be remembered that fingerprints furnish sodium chloride, substances
strongly absorbing in the ultraviolet region (13), oil (sebum), and
cholesterin (1226).
Plastics also may contribute to contamination as has been recently
summarized by DELHEZ (623). Water and solutions stored in polyethylene
bottles may take up some organic substance. It has been reported that
Tygon tubing introduced some plasticizer into a gas stream. Furthermore,
the history of plastic apparatus should be considered; it may absorb and
give off hydrocarbons and other organic substances, and it has been observed
that it retains fluoride, nitrate, and sulfide ion.
230 Systematic Analysis

Sampling for Analysis

Analysis presupposes some object of interest and a specimen of it,
a sample, for investigation. It requires that some matter has been collected,
the chemical identity of which is of interest per se or because of the deductions
to be based upon it.
Sampling and analysis are related as question and answer are (463).
Just as a foolish question will rarely lead to a useful answer, an analysis
will not be able to solve the problem in the mind of the person suggesting
it if the sample has not been selected so that knowledge of its composition
will shed light upon the situation on hand. At times, a very thorough
understanding of the general problem may be required to arrive at an
intelligent decision concerning the objects that shall be analyzed and the
questions that shall be answered by the analyses. The decision may require
conferences with all persons present, who are interested in the problem
or are able to contribute useful information. It is desirable that the analyst
participate in the deliberations; he may gain knowledge to aid him in his
task, and he may be able to suggest approaches that facilitate the analytical
work and improve the probability of its success.
Qualitative ~icro aimlysis is rarely concemed with the average composi-
tion of mixtures; for the reasoning used in such instances, the reader is
referred to the literature (13, 14).
As a rule, the sample for micro analysis will have to be separated and
freed from extraneous material and collected in a more or less "pure"
condition. Frequently, this task must be performed by the analyst himself,
and it occasionally turns out the be the most difficult part of the work.
Not only patience and skill, but also considerable ingenuity may be required
since only very general advice can be given and the procedure must be
adapted to the requirements of the special instance and the nature of the
particular material. The apparatus and technique depend essentially upon
the size of the specimen to be isolated and upon the amount of force required
to separate it from its surroundings. Complications may arise if the specimen
is either poorly defined 01' requires special methods of observation to render
it visible. Only a few of the most obvious examples shall be mentioned
in the following discussion.
In the instance of loose particles, the problem reduces to the collection
of a number of them sufficient for the investigation. Considerable ingenuity
has been expended upon the collection of dust particles. Filter mats made
of carbon dioxide snow, potassium nitrate, benzoic acid, naphthalene,
anthracene, and salicylic acid have been used since they allow isolating
the collected particles free from extraneous matter by simple evaporation
or dissolution of the filter (910, nOl, nOS). For the investigation of
individual particles floating in air as well as for the picking up of individual
 Sampling for Analysis 231

particles from surfaces, it is convenient to use the impinger principle which

has been widely applied in recent times. An impactor which collects the
particles upon a microscope slide for investigation by variations of the
microgram procedure described is used by CADLE (450). The identification
may be simplified by coating the slide with a film containing the reagent (456),
and even continuous counting of particles of a certain kind is possible
by collecting them on a strip of cellulose acetate carrying a film with the
reagent, that moves through the field of the microscope after having
passed the impacting chamber (450). LODGE (455, 470, 471) collects upon
Millipore filters (Lovell Chemical Co., Watertown, Mass.) which become
transparent on impregnation with immersion oil so' that there is no inter-
ference with the microscopic investigation of the collected particles.
Obviously, the individual particles of conglomerates may be set free
by crushing and grinding until the aggregates are resolved into their
components. Suitable t@ols are the wellknown Plattner diamond mortar
and the Ellis' mortar if contamination with steel is not objectionable.
Mortar and pestle for fine grinding are now available made of glass, porcelain,
agate, Mullite which is synthetic 3 Al 20 a • 2 Si0 2 , Coors U. S. A. alumina
and Diamonite consisting essentially of corundum (Mohs hardness, 9),
Kennametal (tungsten carbide with a binder of metallic c~balt or nickel),
and pure boron carbide (468) next in hardness to diamond (13). The micro-
mortar of ALBER (431), obtainable made of Mullite, uses interior proportions
approaching the tall form in order to avoid loss of material; the inside
dimensions are 15 mm high, 7 mm in diameter at the bottom, and 20 mm
at the rim. Micromortar and pestle of similar shape but fashioned of a
single crystal of synthetic corundum, "x-mono" (Al 2 0 a), by cutting and
polishing with industrial diamonds are commercially available (1290). The
material is resistant to common acids and bases, has a Mohs' scratching
hardness of 9, can stand pressure up to 12000 kg!cm2 , and is more abrasion
resistant than Kennametal.
Manipulators and micromanipulators (85, 86, 90, 103) may be used
for dissection and the removal of particles from objects if not much force
is required. Efflorescences, deposits, coatings, or pigments may be scraped
loose with a steel needle and then collected by the technique of Expt. 61 (409).
The simple microchemical manipulator of ALBER (862) which permits
magnifications up to 200 diameters and the application of considerable
force was used for the removal of small inclusions from soap bars (414).
A preliminary treatment with solvents (892, 913) is applied in the inv.estiga-
tion of paintings for loosening the protective surface layer; and solvent
extraction is used to remove the binder from chips of paint (853, 860, 882).
The principle may find other applications and it has been widely used
for the isolation of inclusions in steel by anodic dissolution (162, 188).
A microsectioner for paint films has been described by GETTENS (1092).
232 Systematic Analysis

An interesting detailed account of the technique of dissecting metallic

objects of various sizes under the microscope with the use of simple tools
has been given by CHAMOT (87). The technique has been developed for
the investigation of small arms' primers, but it appears generally useful.
CLARKE and HERMANCE (407) used a rugged manipulator in conjunction
with a dental motor with flexible shaft and a set of drills, burrs, stones,
and cutting wheels for attacking complex structures in a systematic manner
under the binocular microscope. Upon the object was placed a drop of
oil, large enough so that the tiny tool is completely immersed; the particles
removed by the burr are retained in the oil and may be collected by
centrifuging and washing with a suitable solvent.
Microdrills (90) for the removal of grains and inclusions from (polished)
specimens have been described by GRANIGG (768), MORITZ (1230), HAY-
COCK (660), and RUSSANOW (162, 871); the rotating shaft of the drill is
mounted in one tube of a binocular microscope .or magnifier so that the
point of the tool is in the focus of the objective carried by the other tube.
Jimmying of the tool may be arrested by first coating the specimen with
a 0.2-mm layer of celluloid which is applied as a solution in amyl acetate.
KOCH, MALISSA, and DITGES (570) developed a simple manipulator for
the isolation of tiny objects (mass of the order of 1 pg) from the surface
of metallic objects. The specimen is mounted in special holders on the
stage of a Zeiss microscope with binocular eyepiece attachment and special
objectives giving magnifications of up to 80 diameters with working
distances of 92 mm (more than 80 diameters with 27 mm). The tool is held in
a clamp with ball-in-socket joint, but operated directly by the hand which
rests upon a gliding support. This gives smooth motions and permits the
application of considerable force.
Inclusions of 0.1- to 3-mm diameter are isolated by removing the
surrounding material with suitable dental tools operated by a motor
with flexible shaft. When the surrounding metal has been sufficiently
removed, the inclusion is lifted out by means of a lancet or needle. Inclusions
close to the surface may be squeezed out by the forces created when drilling
a hole close by. Corrosion spots may be planed off for separate investigation
of successive layers, and it is also possible to polish small areas for micro-
scopic investigation. Needles, forceps, and a suction tool are described
for the collection of the detritus.
Particles more than 0.2 mm in diameter are picked up and transferred
by me~ns of delicate forceps which are attached to the manipulator and
may be opened and closed by turning a milled screw head at the end or
the tool shaft. Smaller particles stick to surfaces by adhesion forces and
may be lifted after touching them with a needle or fiber. Likewise, they
are deposited by touching them to the collecting surface; this, however,
becomes difficult when the diameter of the tool is large compared with
 Sampling for Analysis 233

that of the particle. For this reason, sharply pointed needles of steel or
platinum are used for particles of about O.I-mm diameter, but textile
fibers (fine vitreous silica fibers) for smaller particles. Fibers are cemented
to the end of a glass thread (Expt. 57) which is inserted into the tool holder
of the manipulator. For the rapid collection of many particles of a kind
from a mixture, an aspirator operated suction device may be used. It is
obtained by placing a small suction tip with narrow orifice, by means of
a ground joint, upon a small disk of fritted glass fused into a tiny funnel,
the stem of which is connected to the glass tubing held by the clamp of
the manipulator. The particles collect upon the filter disk and may be
removed with suitable tools after the disk has been mounted on the stage
of the microscope.
Ferromagnetic particles are collected by applying first a permanent
magnet and then a small electromagnet which picks up weakly magnetic
particles that have been left behind. A cover slip is attached with glycerol
to the pole end of the magnet which is then moved with the manipulator
over the detritus so that a gap of I to 2 mm remains. The magnetic particles
are pulled to the surface of the cover glass and remain there when the
magnet is turned upside down, whereafter the cover slip may be lifted
up vertically. Most of the magnetic material may be col1ected by repetition
of the procedure. The electromagnet is operated with alternating current
and may be given the shape of a pointed needle which may be brought
close to the particle that may then be dropped wherever desired by gradually
reducing the current to zero. Since the wire of the coil has only 0.05-mm
diameter, the current must be limited to about 20 milliamperes.
It is understood that small specimens may be imbedded in resin or
plastic and thus mounted in a ring fitting into a specimen holder.
The glaze of pottery may be separated from the burned clay (141).
The fragment of pottery is split through the middle by means of chisel
and hammer so that only one side of each half has a glaze coating. A
fragment is placed, with the glazed surface down, into molten paraffin
which is allowed to solidify. While running tap water over the specimen,
the clay may then be removed with a dental drill until only the glaze is
left imbedded in the paraffin, which is then transferred with knife and
forceps to filter paper. Pressing between filter paper while heating upon
a block to 80° C, will remove most of the paraffin.
The "ultrasonic jack hammer" (90, 765) uses styli made by electrolytically
pointing a drill rod of 1.1 mm in diameter. The specimen is covered with
a small glass hood which is supplied wit}1 argon to prevent air oxidation as a
consequence of the created frictional heat. The inclusion is focused under
the microscope, whereupon the stylus is lowered to make contact with it.
When this happens, the inclusion is shattered around the point of contact,
and a shower of debris is scattered over the vicinity of the inclusion. The
234 Systematic Procedure of Analysis P.I

extent of destruction is readily controlled by manipulation of the micro-

positioner which guides the tool. The debris is finally collected in suitable
manner. Inclusions of diameters down to 1O!lm may be isolated.

Systematic Procedure of Analysis

It is understood that the technique of working must be adjusted to
the amount of material available for investigation. It should be kept in
mind that also the general approach is determined by the size of sample.
If enough material is available and more may be obtained without
trouble, it is only reasonable to proceed without hesitation to destructive
methods if they promise a quick solution of the problem.
On the other hand, if little material is available and cannot be replaced
or cannot be increased without excessive expenditure of time and effort,
it becomes imperative to derive a maximum of information from careful
reasoning based upon the history of the specimen, its appearance, and
non-destructive testing. Destructive testing must be delayed until careful
deliberation has given a plan assuring solution of the problem with as small
a part of the available material as possible. No part of the sample should
be sacrificed without assurance that some vital information will be gained.
If the sample cannot be replaced and the procedure is not based upon
reliable routine experience, it will become necessary to first test each step
with knowns representing possible compositions of the unknown and to
make any changes that will improve the hope for success before proceeding
with the analysis. Depending upon the nature of the specimen, work of
this kind may require a considerable amount of analytical research.
In work connected with the investigation of crime and in all instances
where there is no knowledge whatsoever concerning the nature of the
unknown, the experimenter is advised to protect himself by first testing
for thil presence of dangerous radiation (P. 14) and of explosives. In addition
to testing for stability upon heating (P. 21), a small sample, not more
than 0.5 mg, may be treated-with the necessary precautions: goggles
and heavy gloves-in a porcelain mortar to test the stability on shearing
and mechanical shock.

P.I The History of the Sample

Extremely small quantities of sample which cannot be replaced may
have to be analyzed in connection with the investigation of crime, accidents,
material failures, and objects of art, archeology, basic and industrial
research. In all these instances, it is desirable to sacrifice only small part
of the sample for analysis and to keep most of it as evidence and for additional
tests, the desire for which may occur at a later time. Under such circum-
P. 2, 3 Preliminary Inspection 235

stances, one should try to get so much information from inspection, non-
destructive testing, and a careful study of all factors surrounding the origin
of the material that a few chemical tests with a small fraction of the sample
will establish the identity.
As a rule, the story of the origin is given by the agency submitting the
sample, but an interview may give additional data. Inspection of the
sample may reveal the need for additional information. It may become
desirable to inspect the milieu from which the sample comes, the tools
used in isolating it, and the wrapping materials used for shipping. It may
become necessary to go through a careful study of a manufacturing process
to get a list of the substances used. When dealing with objects of archeology,
art, etc., the advice of experts is usually available, who should be able
to point out literature and museum collections for additional information.
It is the analyst's responsibility to properly select and evaluate the
information and to decide what the method of isolating and shipping
the sample may have done to it.

P.2 Description of Sample and Record of Investigation

The report should begin with the date of receipt and a description
of the sample as received or of the method of its collection and the circum-
stances connected with it. Especially when the amount of material is small,
an estimate of the quantity of sample should be given and this should be
followed by statements concerning the amounts sacrificed for the various
destructive tests and the quantity left for evidence and future testing.
The report on the investigation should contain all facts leading to identifica-
tion and consequently explained by the identification and also those
observations which seem incidental and are not necessarily connected with
the identity. Depending upon the importance of the problem and the
scarcity of the sample material, efforts should be made to preserve fractions
and tests for future reference or to obtain photographs or drawings of the
specimens and significant tests. The record should finally disclose where
the left-over sample is kept or for what is was used.

P.3 Preliminary Inspection

The aid of a magnifying glass will not be needed, as a rule, in deciding
whether the sample is solid or liquid. Use of centrifugalforceisrecommended
for separating the phases of a slurry; the washing and analysis of the solid
may have to be postponed until investigation of the liquid has furnished
the information needed for efficient performance.
Solids should be inspected with the magnifying glass or (and) under
the microscope (478). To start with, this will reveal obvious heterogeneity
.and may suggest mechanical separation into several fractions for separate
236 Preliminary Inspection P.4:

investigation or mechanical removal of obviously incidental contaminants.

Often it may be advantageous to repeat the inspection in ultraviolet light.
The sample may be placed into a dark cabinet illuminated with ultraviolet
radiation and observed with or without optical aid while shielding the
eyes against the radiation. If available, a fluorescence microscope may
be used (88). The light emitted by the sample or parts of it may reveal
even very slight contamination or heterogeneity (95). Use of or contamina-
tion by lubricant will be revealed and minute stains by biological fluids
will be discovered.
The presence or absence of repetitive external shape or internal structure
will reveal whether or not the material is organized, i. e., derived from
plant or animal. If not fossilized or incinerated, all organized material
will later (P.22) show the presence of organic matter. Since species and
type of tissue frequently give striking differences in shape and structure,
whereas the subtle differences in chemical composition are still mostly
unknown, identification must rely to a large part upon the recognition of
form and microscopic structure, P.5.
Matter which is not organized may again be separated into three classes
be mere inspection with the use of magnifying glass or microscope. The
material may have a shape or structure indicating that it is the product
of some human activity, an artifact (see P. 5); it may consist of more or
less well developed crystals (see P. 6), and it may not be distinctly crystallized
and present shapes which are more or less incidental, P. 7.

P.4: The Sample is a Liquid

Liquids derived from organized matter are, as a rule, either oils (essential
or fatty) or aqueous solutions. The surrounding tissue retards the evapora-
tion of small droplets of these fluids. Water, aqueous solutions, and mineral
oil form large bodies in nature. Outside of living matter, small samples
of water and other volatile liquids will persist only if sealed into cavities;
water, liquid carbon dioxide, and mineral oil may be found included in
minerals and rocks. In all these natural occurrences, the history of the
sample will be a good guide to recognition of the general nature of the liquid.
Likewise, the circumstances connected with the origin of industrial and
research samples should narrow the area of search so that identification
becomes possible with a very small sample. Even when a contain~r is
found without identifying label, the location (place in a systematic collection),
and the type of container should give a very good idea of the possibilities
involved.
Carbon dioxide or water included in minerals is identified according
to W. N. HARTLEY (320, 690, 691) by determination of the critical temper-
ature under the microscope.
P.4 The Sample is a Liquid 237

It is understood that goggles should be worn and the work performed

behind a protective shield as long as there is no assurance that the unknown
has not an explosive character. If sufficient sample is available, a test for
stability on heating should be carried out before exposing any sizable
amount of sample to elevated temperatures. To this end, take up about
1 mg of the liquid through the firepolished end of a thin-walled capillary.
Mount the capillary in slightly inclined position behind a safety shield
and push a Bunsen burner under the capillary so that the just non-luminous
flame envelops the part of the capillary adjacent to its opening holding
the drop, while the opening itself is in or just barely outside the seam of
the flame. Observe whether the substance burns, the appearance of the
flame (luminous, non-luminous, color, sooty), and the nature of decomposi-
tion products; save for analysis any ash left. If explosion or deflagration
does not occur, it may be assumed that it probably is safe to heat the unknown
material.
Assuming that a small sample of a liquid shall be investigated and no
clues to its nature are given, it will be best to determine first its volatility
so that the necessary precautions against loss may be taken in the sub-
sequent testing. To this end, all or part of the sample is taken into a
capillary pipet (p. 100 or 103) with the necessary precautions. The rate
of flow through the fine intake capillary may provide a rough indication
of the viscosity. The color of the liquid may become perceptible, before
sealing, when viewing the opening of the fine capillary containing the
liquid, p. 177. If the whole sample is taken for the test, also the appearance
in ultraviolet light should be observed, P. 9, before heating. The attempt
of determining the boiling point may indicate the presence of a more or
less pure substance and give its boiling point; it may indicate a mixture
of liquids and give its approximate boiling range; finally, it may indicate
the presence of dissolved solids (or gases). By all means, the temperature
should not be raised much above 300 0 C, and heating should be stopped
at the slightest sign of decomposition. After cooling, the centrifuge will
serve in collecting all material in the tip of the capillary.
If a sharp boiling point is obtained, a study of the tables in handbooks
together with additional observations (color, fluorescence, fluidity, odor
or fuming perceived during transfer) may suffice to get identification
by means of a few additional tests. The recognition of a particular batch
of the liquid may be possible by means of a schlieren test, Expt. 5, which
will also allow to identify a particular batch of mixture.
If boiling starts above 200 C, one may assume that vaporization losses
0

will not be excessive at room temperature. If it starts below 200 0 C, the

liquid should be classed as volatile and special precautions should be
taken when working with small samples; most tests are easily modified
for performance in capillaries.
238 Preliminary Inspection P.4
Testing the odor by P. 16 becomes practical with small samples if the
volatility is low. Furthermore, a droplet may be transferred to the surface
of a slide for inspection under ultraviolet light; this may be desirable since
viewing the sample in glass, which absorbs most of the ultraviolet, may
not reveal slight fluorescence. The refractive index too, may be determined
with the glass powder scale upon the microscope slide (159). In the instance
of volatile liquids, a few particles of a glass powder may be brought with
the centrifuge into the taper of a capillary pipet, whereupon the tip of
the pipet is inserted into the liquid and the latter permitted to enter until
the particles are immersed (653). For microscopic inspection, the tip
may be sealed, and the capillary may be placed into a suitable liquid,
Fig. 50. After observation of the test and cleaning of the outside of the
capillary, its tip may be cut open, and the contents may be transferred
with the centrifuge to a pipet with another sample of glass powder or
returned to the storage vessel.
The melting point may be determined in the same capillary in which
the boiling point has been observed.
Nonaqueous liquids that fume in contact with air may react with water.
A small sample may be treated with water, and the resulting aqueous
solution may then be tested for inorganic ions; if a gas is liberated during
the reaction with water, it should be collected for identification (153, 157,
1090).
If the liquid is a mixture, the procedure should be selected with due
consideration of the collected evidence and the origin of the sample.
Separation may be tried by distillation, gas chromatography (477),
fractional crystallization (Expt. 54), or a combination; if properly performed,
significant loss of material need not occur. A mixture of organic liquids
may be extracted with water, dilute NaOH, or (and) dilute acid; the same
treatment may be applied to solutions of solids in organic solvents, but it
will be preferable to evaporate the solvent in such a manner that it and
the residue may be collected for the determination of their relative masses.
In the instance of an aqueous solution, it will suffice to evaporate a measured
part of it so that the resulting solid residue may be weighed; neither
measurement of volume nor determination of weight have to be very
accurate. The residual solid is investigated as suggested in P. 6 or P. 7.
The identification of isolated liquids may be tried by infrared and
ultraviolet spectroscopy (P.34), determination of the refractive index
(p.238 above and P. 18), determination of the density (P. 19), of optical
rotation, of surface tension, of viscosity, of melting point, of boiling point,
of critical temperature, of the critical solution temperature, by ultimate
analysis (P.35), tests for functional groups, specific confirmatory tests,
and methods of quantitative analysis.
P.4 The Sample is a Liquid 239

Plotting one physical constant as a function of another reveals simple,

nearly straight-line relationships for the points of homologous series.
Charts showing the plotted .data on refractive index, density, and boiling
point permit classification, limit the possibilities to be considered very
sharply, and indicate what tests or derivatives, if any, are needed for final
identification (449).
MARION (427) has shown how to convert at little expense a polarizing
microscope to a polarimeter with a cell of 0.15- to 0.18-ml capacity. For
other suggestions see SMITH and EHRHARDT (439).
The surface tension may be determined with 2 ,ulliquid by the method
of FERGUSON and KENNEDY (1068). The liquid is placed into the open
end of a narrow capillary which is connected to a manometer system
in which the air pressure is varied until the meniscus at the mouth of
the tube is plane.
The capillary microviscosimeter of BOWMAN (423, 735) requires only
30,ul of sample and gives a precision of ± 0.001 in the range from 2 to
10000 centistokes.
The critical temperature may be readily determined by heating, on
the microscope hot stage, a sealed capillary of about 20-mm length, one
third of which is filled with the liquid (653, 1019).
The critical solution temperatures have been listed (I) for 6000 pairs
of liquids, about 70 % of which are hydrocarbon compounds. The determina-
tion with the use of the microscope hot stage has been described by
R. FISCHER (886, 889, 897). It requires from 0.2 to 2,ul of sample depending
upon the bore of the capillary used, which may be varied from 0.2 to
0.9 mm. Needed is a test liquid which is pure or closely reproducible, is
not completely miscible with the sample at room temperature, but becomes
miscible at a higher temperature before a critical temperature is reached
or decomposition starts. Suggested are: paraffin oils for low molecular
weight alcohols, aldehydes, ketones, and phenols; glycol for essential oils,
ethers, esters, aldehydes, ketones, and benzene derivatives; glycerol for
aldehydes and ketones; ethylenecyanide for benzene, its homologous series,
and halogen derivatives; benzyl alcohol for aliphatic hydrocarbons; methanol
and aniline for aliphatic hydrocarbons and naphthenes.
Take into a capillary of 30- to 40-mm length from 5 to 7 mm of the
sample and of the test liquid selected. The two liquids must be in contact.
Seal the capillary at both ends to give it a length of about 30 mm.
It is assumed that the hot stage has been equipped with a metal frame
which carries an aluminum sheet, 26 mm X 37 mm X 1.5 mm thick, so that
the latter may be moved by means of milled heads outside the hot chambel'.
The aluminum sheet has a slot, 32 mm long and either 0.5 mm or 1.1 mm
wide, into which the capillary fits.
240 Preliminary Inspection P.l)

Place the sealed capillary into the slot of the aluminum slide, and cover
the latter with a glass slide, 26 mm X 34 mm, which is held in place in some
suitable manner. Tilt the microscope to a position 45 degrees to the
horizontal, and move the frame so that the meniscus appears in the field
of vision with the capillary in the diameter from 6 to 12 o'clock. Raise
the temperature at any desired rate and keep the interface of the two
liquids in the field of vision by means of the motions of the frame. When
the critical solution temperature is approached, the meniscus flattens out,
and its outlines become faint and disappear when the critical temperature
is reached. Allow the stage to cool very slowly (1 to 2 degrees per minute)
and watch the region in the capillary, in which the meniscus was last seen.
Read the temperature when the critical point is indicated by the appearance
of droplets and a fine line, the meniscus.
The critical solution temperature can be reproduced within 0.20 C.
As a rule, the effect of the pressure seems to be negligible. Because of the
lack of convection currents, the two liquids mix mainly by slow diffusion
which automatically produces the critical mixture somewhere in the
narrow diffusion zone of about 0.5-mm height. The possibility should
be kept in mind, however, that the meniscus may disappear if both liquids
acquire the same refractive index at a certain temperature. Mistakes are
easily avoided by the use of oblique illumination or of the Becke test;
the meniscus never disappears altogether, and it becomes again more
distinct if the temperature is raised further. Finally it must be pointed
out that the critical solution temperature may be very strongly affected
by impurities of the sample; in such instances, it may be used for a rather
accurate determination of the amount of impurity.

P. I) Identification of Organized Matter

Identification by shape and structure requires familiarity with these.
The chemist with little or no training in biology and general microscopy
may, with the use of introductory books and by the study of the micro-
scopic appearance of various samples of known origin, nevertheless be
able to recognize at least the class of matter to which an unknown belongs.
If samples of this kind will have to be handled frequently, it becomes
certainly advisable to study the specimens or materials which have to be
considered, and to collect samples, drawings, photomicrographs, and
microscopic mounts for comparison. Study of different samples of the
same kind will show the characteristic features which are preserved in
spite of incidental variations. The execution of drawings is strongly
recommended; it is not only an aid to the memory, it helps developing
the ability of recognizing distinctive features.
Whereas inspection under the microscope may immediately reveal
the identity of a sample in some instances, it would be a mistake to
Plate I. Investigation of a Solder with an Electron Probe Microanalyzer, Magnifi-
cation 400x. a Light photomicrograph showing the darkened area analyzed with the
electron probe; the darkening is caused by contamination with oil from the diffusion
pump, which polymerizes where the electron beam strikes the surface; b Image
produced by backscattered electrons when the specimen is scanned with the electron
beam; the dark areas indicate regions which are lower in average atomic number
then the light areas; c, d, and e Images obtained with the characteristic X-rays
excited by the scanning electron beam and showing the distribution of arsenic (c),
tin (d), and antimony (e). i
Micrographs by the courtesy of Research Laboratories, General Motors Corporation,
Warren, Michigan

Benedetti-Pichler, Identification Springer-Verlag in Wien

Plate II. Examples of Organized Mat.ter. 1 Amoeba Proteus, 40X; 2 Volvox, 40x ;
3 Euglena, 700X; 4 Thalassicalla Nucleata, 15 X ; 5 Diatoms, 450x; 6 starch grains,
SOX; 7 bacteria, lOOOx; 8 pollen, 120X ; 9 filings from finger nails, SOX; 10gro1JJld.
black pepper in cedar wood oil, 400X; 11 cat's hairs in cedar wood oil, 400X;
12 ground black pepper, dry in reflected light, SO X ; 13 particle of human skin, in
air and reflected light, SOX; 14 legs of insects, in air, SOX; 15 wood filings in cedar
wood oil, 400x

Benedetti-Pichler, Identification Springer.Verlag in Wien

P.5 Identification of Organized Matter 241

expect this to occur as a rule. The shape of starch grains (Plate II),
to mention one example, does appear also among pollen, bacteria, and
occasionally even in crystalline material, Expt. 36. The very variable
size may not provide a decisive clue. Soil, ground black pepper, tobacco
dust, and many others may look very much alike under low magnification;
filings from finger nails (Plate II) resemble chips of white paint film
or plastic. In all such instances, immersion in water or cedar wood
oil and change to medium or high magnification will reveal the cellular
structure of organized matter. Origin and history of the sample provide
most useful leads. Protozoa occur only in essentially aquatic media
including living tissue and body fluids. Soil is of such variable com-
position that no general characteristic can be given; the presence of
hard particles (quartz) is readily recognized by the scratch marks
produced on moving the cover slip under slight pressure, P. 17. The
recognition of the soil of a particular locality requires close comparison
with a sample taken from the same spot (dirt on shoe and soil of
foot print); the cooperation of a biologist .tnay be essential.
The beginner needs advice concerning sources of materials for study
and the preparation of microscopic mounts. To this end, CORRINGTON'S,
"Exploring with your Microscope" (175), and other elementary introductions
to general microscopy (Ill, 190) are valuable. The following list gives an
idea of the kinds of organized matter, that may be met, and supplies
references to the special Jiterature.
Micro Organisms
Algae (175) Molds (lSI, IS2)
Bacteria ( IS 9) Protozoa (175, lSI)
Diatoms (175) Yeasts (IS2)
Seeds (ISO)
Parts or Tissue of Plants (177, ISO, IS5, 194, 196, 197, 202, 205)
Fibers, see below Starch (ISO, 190, 197)
Wood (179, ISO, 190, 197, 1100) Charcoal (1l00)
Spices (195) Vegetable Drugs (200)
Parts or Tissue of Animals (ISO)
Insects (175, lSI) Skin and Leather (672)
Excreta (lSI) Hair and Feathers (lSI, IS6, 675,
Blood (175, IS6) 1130), see also Fibers, below
Fibers (ISO, IS3, IS4, ISS, 190, 191, 197, 204)
Artificial Fibers (180, IS4, ISS, 204) may also be identified by the
melting point (459)
Food (lSI, 201)
Water (207)
Benedetti-Pichler, Identification 16
242 Preliminary Inspection P.6
A few chemical tests may be mentioned, which may serve for general
orientation.
1. If not completely fossilized or already ashed, the material will char
and in general show the presence of carbon compounds when tested in P. 26.
If the ashing of tissue is carried out with the necessary precautions, the
structure is preserved in the ash and may be more clearly recognized there
than in the original (194, 310, 641, 642, 852).
2. Moisten with O.l-F 12 in O.4-F KI solution: starch turns blue; fat,
oil, protein, wool, and silk become yellow or brown; artificial cellulose
fibers become brown at first and turn blue after some time. The material
may be recovered by allowing the test to stand until the iodine has evaporated.
3. Moisten with 16-F HNO s and heat upon the steam bath until dry.
Proteins assume a yellow color which darkens upon adding (6-F) NH s.
Many resins, alkaloids, and tyrosine behave in a similar manner.
4. Moisten with a solution of 1 g phloroglucinol in 50 ml ethanol,
which has been treated with 25 ml 12-F Hel: lignin containing, woody
tissue assumes a bright red color which slowly changes to violet (650).

P.6 Identification of Artifacts

Man-made materials are recognized by shapes or textures which must
have been purposely brought about and cannot be imagined as the result
of a chance happening. Even when the external shape is completely
preserved, the criterion fails in the instance of fibers. When fragments
occur, that preserve little or none of the outside shape, the structure may
not afford a clue to the origin. Thus, splinters of glass and obsidian, or
fragments of industrial slag and pumice cannot be differentiated by their
structure. As in the instance of organized matter, a check list of the objects
which may be met, appears useful.
No Outstanding Dimension, Irregular Shape
Fragments of structural parts: metal, alloy, concrete, brick, mortar,
caulking, ceramics, plastics, rubber, sponge of cellulose,
plastic, glass
Filings, shavings, chips of metal, alloy, plastic, wood (llOO), cork,
pulp (179)
Beads of metal, alloy, glass, or plastics
Powders obtained by grinding or precipitation, see P.8
Food Products: candy, baked and unbaked dough, bread, cake,
noodles
One Outstanding Direction,
Spun fibers, thread, wire, needles
P.7 Well-Developed Crystals 243

Two Outstanding Directions,

Metal foil, laminates, paper (179), inorganic paper, cloth, paint and
lacquer films, coatings, chips of paint, enamel, glaze, plaster,
and other structural materials
Since very different materials are put to identical uses and are given
like shapes, chemical analysis gains in importance. As far as organized
matter is involved (wood chips, paper pulp, fibers), identification by
structure as outlined in P_ 5 will have to be used_ In general, it will be
advisable to obtain good photomicrographs or (and) drawings of the
sample material before it is put to destructive tests since the appearance
may assume increased importance when the results of chemical analysis
are known. A scale should be included into the likeness. The particle
size distribution of powders may be decisive in the identification of a
particular product or brand.
Inspection in ultraviolet light may produce valuable evidence concerning
origin or characteristic contamination and should not be omitted with
artifacts. The selection and order of the tests of the systematic procedure
will be determined by the history of the sample and its appearance (shape,
texture, transparency, luster, body color, surface color) in white and
ultraviolet light. The use of polarized light, testing with a magnet, or a
hardness test may be indicated before reviewing the evidence with the use
of Table I, handbooks, and special literature. One may conclude that
mechanical separations or the removal of binding matter by solvent
extraction should precede the chemical investigation of parts of special
interest. In general, the whole systematic testing procedure, starting
with P. 9, should be considered.
In drawing the final conclusions at the close of the investigation,
the question of origin or batch identity may have to be solved on the
basis of appearance or characteristic contamination and comparison with
samples of known provenience.

P.7 Well-Developed Crystals

If the sample consists of more or less well-developed crystals, proceed
with the systematic procedure starting with P.9. Photomicrographs or
drawings should be made. An attempt may be made to determine the
crystal class, and observation of transition temperatures while heating to
300 C should not be omitted. Especially when dealing with minerals,
0

identification by their optical properties may be considered. Under all

conditions, it will pay to determine characteristic angles and to study the
behavior in polarized light.
Crystals which are not clear have undergone some change, and their
outward shape is no longer related to their internal structure. The outward
16*
244 Non-Destructive Testing P. 8, 9

appearance may give a clue to the chemical composition of the originally

formed crystals and to the nature of the change (transition to another
modification, change to a different state of solvation, dehydration, oxidation,
reduction, metathesis) which has taken place.

P.8 Solids of Random Shape and Structure

This group shall comprise crystalline and amorphous matter in the
form of powder or irregular lumps or pieces; it is also assumed that the
out[nes of crystals cannot be clearly discerned.
Microscopic examination should be tried with transmitted and reflected
light and varicolored backgrounds; observation in ultraviolet light is
recommended, and the magnetic behavior should be tested, P. 15. Photo-
micrographs may be used to record the general appearance since type of
agglomeration and luster of surface are difficult to describe. If this holds
any promise, a Geiger test and an autoradiograph of the sample, P.14,
may be made before proceeding with the investigation. In general, the
latter will have to follow the systematic procedure starting with P.9;
testing the solubility, P. 32, may be rewarding if recrystallization furnishes
crystals suited for optical investigation and angular measurements. The
hardness should be tested if abrasives may be present, and the determination
of transition points below 300 0 C should not be omitted, if carbon compounds
are not excluded.

Non-Destructive Testing
P.9 Action Upon Light, Color
By color of an object is meant its transmission or body color, i. e.,
the sensory effect of the change in composition imparted to daylight which
has passed through a selectively absorbing object. It is implied that
illumination with artificial light may give entirely misleading color
impressions, and it is assumed that the observer has "normal" color
perception.
The body color is greatly affected by the thickness of the specimen.
Thin layers of lightly colored substances appear colorless in transmitted
light, and thin layers of strongly colored matter or highly opaque substances
may show unexpected hues which do not seem related to the color of bulk.
The body color of fine powders is most sensitively observed with darkfield
illu:rp.ination or with reflected light before a black background; a white
background will serve if the particles are darkly colored. Use of all three:
low-power microscope, magnifying lens, and of the unaided eye is
recommended; if the particles are very fine and appear only lightly colored,
one may also try to gather them into a pile for the observation of the
color of the reflected light. Immersion in media of like refractive index
P.9 Action Upon Light, Color 245

for the elimination of disturbing refraction phenomena is not practical

at this stage of the investigation if only little material is available.
Any form of darkfield illumination (reflected light and black back-
ground) may also bring forth surface and fluorescence colors. Surface or
reflection colors are caused by selective reflection from the surface. As in
the instances of crystals of KMn0 4 or indigo, they occur only with strongly
absorbing substances and are approximately complementary to the body color .
Finally, body color should not be confused with color resulting from
structural peculiarities of the object, which may produce:
a) white which is always caused by repeated reflection and refraction
of white light from the large "internal" surface of conglomerates of small,
colorless, transparent objects (flour, marble, porcelain, milk, foam, smoke);
such matter prevents passage of light and, in transmitted light, may
consequently show:
b) black which as a true body color would be the consequence of an
absorption band taking in all of the visible region;
c) interference colors characteristic of thin plates or films or brought
about by the grating effect of repetitive microscopic structures (88);
d) color resulting from difference in dispersion when the refractive
indices of object and surrounding medium are approximately matched,
Christiansen effect (88);
e) Tyndall colors resulting from the selective scattering of particles
of about 300 nm diameter, which makes them appear blue In reflected
light and orange or brown in transmitted light (88);
f) transmitted blue (88) and colors derived from the chromatic aberration
of the optical system, which will disappear or change when the illuminating
or (and) image-forming system is altered.
Interference colors and the Christiansen effect are characteristic of
certain substances, but are not directly related to their chemical identity.
An accurate description of color requires statement of the spectral
composition or reference to a chart of standard hues. No such device is
being used in Table I since color and hue of many substances varies with
their history. The Table must be in many ways incomplete and inaccurate;
it should be taken as an aid to the memory, trying to make certain that at
least the most obvious possibilities are recognized. Among others it does
not mention lakes of organic dyes and tungsten bronzes, both of which
may exhibit any color of the rainbow. The former char and leave an
inorganic residue when ignited. Tungsten bronzes are characterized by
their metallic luster, high density, and semimetallic properties. They are
of variable composition and obtained by the reduction of alkali and alkaline
earth tungstates. They precipitate metallic silver from ammoniacal
solutions, are resistant to HCI or HN0 3 (but not always to a mixture
of the two), and are oxidized to tungstates when ignited in air (16).
246 Non-Destructive Testing P.9
Luminescence is the emission of light after absorption of radiation
of other frequency. In this connection, emission of visible light is wanted.
If it occurs during excitation, one speaks of fluorescence, but if it continues
after the exciting radiation has been cut off, it is called phosphorescence.
Ultraviolet light is most convenient for excitation of fluorescence.
If it is alltogether free from visible radiation, the light emitted by the
object cannot be confused with reflected or refracted visible light coming
from the source.
Lenses and prisms of vitreous silica are needed for concentrating ultra-
violet light into the small field of a high-power microscope (88). On the
other hand, illuminating with ultraviolet for the observation of fluorescence
(phosphorescence after the light has been turned off) with the unaided eye,
a magnifying lens, or a low-power microscope is relatively simple. Argon
glow lamps may be mounted in a black cabinet so that they irradiate
objects placed into the center of the box, but cannot be seen through
openings provided for observation. A cleverly designed cabinet should
permit observation with the low-power microscope as well as inspection
with the unaided eye of large objects held in the hand. Ultraviolet light
is absorbed by the lenses of a microscope or a glass window, but stray
light is difficult to avoid. Thus the observer should, by all means, wear
goggles with ultraviolet absorbing glass, that enclose the eyes on all sides.
The samples to be inspected are best directly exposed to the source
of ultraviolet without intervening cover glass. Containers of vitreous
silica or of Corex glass are best for liquids and gases, but apparatus of thin
ordinary glass may serve.
A plate of strongly fluorescing uranium glass or a thin-walled beaker
with alkaline fluorescein solution is used for determining the path of the
ultraviolet radiation. A smear of petrolatum or a fragment of a boric acid
crystal may be used as test objects; they should be luminous when brought
in the position intended for the object. In the same position, unglazed
porcelain should be black if visible radiation and stray light are absent.
The outstanding feature of inspection in ultraviolet light is the extra-
ordinary sensitivity with which fluorescent substances, contamination,
and heterogeneity are shown. This recommends it for the preliminary
inspection of samples and the isolation and collection of the parts of special
interest. The usefulness for identification of substances is seriously impaired
by the fact that trace impurities in a substance as well as its history greatly
affect its behavior; they determine whether or not there is fluorescence or
phosphorescence, and they determine the color of the fluorescent light.
This peculiarity, however, renders observation in fluorescent light a useful
tool for batch identitication after the nature of the substance has been
recognized by other means, and there are many applications connected
P. 10,11 Investigation of Crystals and Crystal Fragments 247

with determining the origin and history of materials and with the authenticity
of documents, objects of art, etc. (50).
Pure inorganic substances show little or no fluorescence, but technical
grades fluoresce in various colors. Outstanding and useful for identification
are the strong yellow-green fluorescence of uranium compounds, the
strong phosphorescence (yellow-orange) of impure zinc sulfide and luminous
paints, and of Zr0 2 (white). Strong fluorescence is also characteristic and
indicative for some minerals: calcite, CaCOa (red to violet); fluorite, CaF 2
(red); celestite, SrS04 ; ruby, AI 2 0 a ; spinel, MgO· Al 2 0 3 (red); hydro-
zincite, 5 ZnO· 2 CO 2 ' 3 H 2 0 (blue-white); and zircon, 7.r0 2 • Si0 2 (10).
It is claimed that most organic compounds fluoresce (10, 621), tut
there is still doubt whether this holds for the pure compounds. Strong
fluorescence is exhibited by (solutions of) derivatives of diphenylmethane,
triphenylmethane, quinoline, acridine, as well as by leuco compounds of
dyes (indigo), chlorophyll, porphyrins, body liquids, petroleum and its
fractions.

P.10 Investigation of Crystals and Crystal Fragments

Most of the follo"'ing observations assume that material is transparent.
Obviously, angular measurements may be performed on opaque objects,
but the data obtained may belong to a form or compound which has
undergone metamorphosis and is no longer present.
The measurement of angles requires reasonably well-developed crystals
and fragments which retain characteristic facets. In the instance of isotropic
fragments, the lack of crystal form makes it impossible to decide whether
the material has a regular (cubic) lattice or is glassy. If a solvent which
does not react with the material can be found, the result of recrystallization
will readily solve this problem and will also produce crystals for measuring
angles in the instance of anisotropic materials.
The determination of precise crystallographic data will assure most
positive identification but requires use of special equipment; in addition,
considerable difficulty may be met in the preparation of crystals suitable
for microscopical measurements. The effort spent, naturally, will not lead
to identification if the data are not listed in the literature or the indexing
system fails, but the effort gives the data which may be listed after the
material has been identified by other means.

P.ll
Interfacial Angles of pyramidal forms and of prism and dome faces of
monoclinic and triclinic crystals are characteristic for each substance and
permit its immediate identification; In 1912, E. S. VON FEDoRow submitted
an index of intedacial angles of some ten thousand substances to the
248 Non-Destructive Testing P.12

Russian Academy of Sciences, but the organization of the data was not
practical. Simple classification rules for selecting the key angles were
developed by T. V. BARKER (83) and applied in the Barker Index of
Crystals (104). The latter lists in three volumes about ten thousand of
the more common inorganic and organic substances arranged in the order
of their characteristic interfacial angles; in addition, other useful data are
given such as melting points, refractive indices, and cross references to
the X-ray powder index.
The relatively inexpensive and efficient two-circle optical goniometer
of L. W. CODD and W. T. MOORE (1105) is claimed to give the interfacial
angles with an accuracy of three minutes of the arc, but its usefulness in
micro analysis is somewhat limited by the size required for the crystals
to permit proper mounting, at least 0.1 mm XO.1 mm XO.2 mm = 0.002mm3
or 6 flg for a density of 3 g/ml. The angular measurements, the precision
of which depends mostly upon the perfection of the crystal facets, are
used to construct a gnomonic proJeetion of the faces which are identHied
by reference to orthogonal or (and) perspective drawings of the erystal (665).
The revealed symmetry gives the crystal class, and application of the
Barker rules leads to the recognition of the classification angles and
identification.
If crystals are too small for mounting on the goniometer or such instru-
ment is not available, it is possible to derive by graphic methods the inter-
facial angles from angles observed and measured under the microscope (438).
The precision of angular measurements under the microscope, however,
will rarely exceed ± 0.5 degree, but the graphically derived interfacial
(Barker) angles "will be serviceable, if allowance is made for a possible
error of 2°" (438). Naturally, the interfacial angles of cubic crystals or
combinations of pinacoid and prism of the hexagonal, tetragonal, and
orthorhombic classes are determined by the implied symmetry and not
characteristic of a particular substance.

P.12
Profile Angles. SHEAD (412) recommends measuring the angles of the
outline of very thin crystal plates which, by necessity, will always settle
with the large faces parallel to the surface of the microscope slide. Thus,
the profile angles observed under the microscope are readily reproducible
and characteristic for certain substances which have a tendency to separate
in tabular form.
The profile angles are related to the interfacial angles, and have like
the latter little or no diagnostic value in the instances of regular hexagons
and rectangles observed on cubic crystals or pinacoids and unmodified
prisms of the hexagonal, tetragonal, and orthorhombic classes. Angles
derived from pyramids and domes and their combinations with prisms
P.12 Investigation of Crystals and Crystal Fragments 249

and pinacoids are useful for identification purposes. But for the triclinic
system and clinopinacoid faces of the monoclinic, the angles in the outline
of thin tablets are related to one another in a simple manner.
An asymmetric octagon is ordinarily the most complex form met.
If the characteristic profile angle of the basic parallelogram cannot be
derived, it is still possible to use the angles and the order of their occurrence
for comparison and identification.
SHEAD (958) suggests determining the profile angles on selected perfect
thin tablets of the simplest geometric form available, preferably parallelo-
grams, the whole outline of which i.s simultaneously in focus. If only one
kind of parallelogram is found, the two different angles add up to 180 degrees,
and either one of them may be taken as the characteristic profile angle.
If other, more complicated forms appear in addition to the simple
(basic) parallelogram, the derived forms may be identified by simple
reasoning which may also be used for recognizing the basic parallelogram
and its characteristic profile angle (956). Five- to eight-sided forms originate
when one to four corners of the basic parallelogram are cut off parallel
to a diagonal of the basic parallelogram. Depending upon whether the
basic parallelogram is equilateral or not, from one to two or from two to
six new angles, respectively, may appear.
The equilateral square or rhombus is readily recognized by testing
whether or not the cross hairs of the eyepiece can form the diagonals.
Truncation of a corner or two opposite corners by a parallel to the diagonal
gives a symmetrical pentagon or hexagon; there is only one new angle B
which is related to the characteristic Frofile angle A of the basic parallelo-
gram by 4 B = 720 - 2 A. This condition holds also when the equilateral
nature of the basic parallelogram is hidden by unequal growth rate which
changes the square to a rectangle and the rhombus to a rhomboid.
If the basic parallelogram is a rectangle or rhomboid, truncation parallel
to one diagonal gives two new angles, Band G, so that 2 A + 2 B + 2 G =
= 720, and truncation parallel to both diagonals of a rhomboid gives
two more new angles, D and E, so that D + E = 180 + A and 2 B + 2 G +
+ 2 D + 2 E = 1080.
If the truncation is so severe that nothing is left of the sides of the
basic parallelogram, the derived form is a parallelogram that preserves
the angle of the diagonals of the basic parallelogram.
The desired simple shapes may be obtained by recrystallization (412, 970)
or sublimation (412, 958, 971). In general, an intermediate rate of growth
seems to faver the formation of thin tablets of simple outline. If well
developed, three-dimensional crystals are obtained, the rata is too slow.
On the other hand, finely divided granular or formless as well as feathery
and dendritic deposits indicate that they grow too fast. Shead sublimes
from one slide to another which is separated from the first by a glass ring
250 Non-Destructive Testing P.12

__
Table VI. List of Profile Angles (412, 958)

Parallelogram I~ ~~~:gon ___ ~

Substance I 4 angles
degrees I
acute angle 2 angles
degrees degrees

KCIO a ......•..•....••..•.... 79.8

KNO a •••.•••••.••.••.••.•••• 79.8
CaS0 4 • 2 H 20 ............... . 52.5
CaC H 0 6 • 4 H 20 ............ . 57.5
AgCHaCOO ................. . 90
Ag 2C2 0 4 • • • • • • • • • • • • • • • • • • • • • 58
Ag 2Cr2 0 7 • • • • • • • • • • • • • • • • • • • • • 44.5 87.4 136.3
HgBr2 ••••••••••••••••••••••• 69
HgI 2 •••••••••••••••••••••••• •
64.4
HgCHaCOO ................. . 83.5 97 131.5
Acetanilide .................. . 99.5 130.2
Anthracene .................. . 109.7
Antipyrene .................. . 128 116
Asparagin ................... . 50.7
Aspirin ..................... . 119.7 120.3
Bromanil ................... . 109.2
Chloranil .................... . 113.3
p-Dichlorobenzene ........... . 59
Morphine .................... . 59.1 118.2 120.9
Naphthalene ................ . 108.2
o-Nitrobenzoic acid ........... . 39.5
o-Nitrophenol ................ . 80
m-Nitrophenol ............... . 58
p-Nitrophenol. ............... . 77
p-Nitrotoluene ............... . 80
Phenobarbital ............... . 56.7 II:H 123.3
122.1 119.5
Picric acid .............. ; . . . . 87 87 136.7
108.5 (?) 126.3 (?)
Sulfonal. . . . . . . . . . . . . . . . . . . . . . 85 95 132.5
Tribromphenol bromide ....... 69.7 110.3 124.5
Trional ...................... 86.5
Urea nitrate ................. 81.8 81.8 139.4
49.8 99.5 130.2

of 4-mm height and 16-mm inside diameter. The condensing slide is slightly
greased with sebum from the fingers or face, wiped to leave only an invisible
film of oil, and then heated to the sublimation temperature before placing
it on top of the ring. The material to be sublimed should be evenly distributed
over the floor of the cell.
P.13 Investigation of Crystals and Crystal Fragments 251

One estimate of the precision of the profit::: angles listed in Table VI

is ± 0.25°.

P.13
Behavior in Polarized Light. The following data may be determined:
a) Test for anisotropism: Expts. 7, 9, and A. N. WINCHELL, p. 106 (113).
b) Determine the vibration directions and the angl~ of extinction:
Expts.8, 9, and WINCHELL, p. 106. Prepare drawings and enter the
vibration directions and the angle of extincticn.
c) Determine the relative velocities, and mark the vibration direction
of the slower component with 8: Expt. 9.
d) URing one nicol, determine pleochroism: Expt. 11. In the drawing,
record the colors observed with the vibration directions.
e) Using one nicol prism, observe the intensity of shading (heavy or
light outline) and possibly behavior of Becke line in both extinction p08itions.
Record the difference found and the C,)llc]usion made concerning the
stl'ength of double refraction (magnitude of the difference between the
refractive indices for the two components of light). The strength of bi-
refringency may also be estimated from the interference color and the
estimated thickness of the specimen (88).
f) Possibly try to observe axial figures, Expt. 12, and if such are
obtained, try to determine the sign of double refraction; the location of
the optic axial plane OAP; the magnitude of the optic axial angle 2 E in air,
2 V in the crystal, and 2 H in an immersion medium of n = 1.515; and
the direction of the acute bisectrix Bx a •
Concerning the interpretation of the findings, some caution is indicat.::d.
Obviously, solid substances which are isotropic are either a glass or
crystallized in the cubic system, but these materials become anisotropic
when exposed to strain. Internal strain may result from twinning or
external influences during growth, and diamond is usually aniE.otropic.
Thus, whereas isotropism indicates the cubic or glassy states, anisotropism
does not exclude them.
The cubic or isometric class is established when an obviously crystallized
material is isotropic. In the instance of irregularly shaped bodies,
recrystallization or X-ray diffraction may be used to differentiate between
glass and cubic crystal.
Hexagonal crystals are recognized by the fact that they always show
parallel or diagonal extinction and that "end views" (or cross sections)
are isotropic and have the shape of regular hexagons (profile angle, 120°),
equilateral triangles (60°), or dodecagons with alternating profile angles
equal to (120 + 21)° and (180 - 2 ft, respectively. By isotropic end
views is meant that these remain completely dark between crossed nicols
while the stage is given a rotation through 360°.
252 Non-Destructive Testing P. 13

Tetragonal crystals and orthorhombic crystals also have always parallel

or diagonal extinction. If an end view may be obtained which is either
a square (profile angle, 90°) or an octagon (angles of 90 + 2 f alternating
with such of 180 - 2 f degrees), remains completely dark between crossed
nicols during a complete revolution of the stage, and (or) gives a uniaxial
interference figure, tetragonal structure may be safely assumed. As a rule,
orthorhombic crystals will not present an "end view" with a simple rec-
tangular profile when they approach the behavior of an isotropic substance
between crossed nicols. In addition, even when viewed parallel to an axis
of isotropism, they never appear as dark as a uniaxial (hexagonal or
tetragonal) object, and the interference figure is of the biaxial type.
Theoretically, the biaxial classes could be differentiated by the fact
that orthorhombic crystals show only parallel (diagonal) extinction,
monoclinic crystals show parallel and oblique extinction, and triclinic
crystals exhibit only oblique extinction. In reality, measurement under
the polarizing microscope is not precise enough to assure recognition of
a slight degree of obliqueness, and the danger exists that monoclinic crystals
may be classified as orthorhombic or triclinic ones as monoclinic. This
does not prevent, however, identifying triclinic crystals with certainty if
all views give clearly oblique extinction; it will be necessary to "roll" a
crystal under the microscope to get all possible views.
All classification based upon crystal profile is impossible with fragments.
The positions which give darkness (uniaxial) or a more 01' less dark gray
(biaxial) when rotating the stage with crossed nicols inserted into the
path of light, are suited for starting the observation of interference figures
(with biaxial objects, ODe wil1 then try to view along the acute bisectrix) (88).
Use of the interference figures will permit distinguishing between uniaxial
and biaxial materials, but it will not be possible to assign a crystal class
unless suitably developed forms are obtained by recrystallization.
Since cubic and hexagonal crystals are readily recognized and do not
occur too frequently, lists of the more common of these substances have
been abstracted from Lange's Handbook of Chemistry which in 1952
described about 800 minerals, 2600 inorganic substances, and 6800 organic
compounds. The lists of Tables 2 and 3 (Appendix) are long and obviously
incomplete, but they may be helpful since circumstances may rule out
whole groups. Thus, metals are excluded if the particles are transparent,
and rare minerals or exhibition specimens of synthetic chemicals need,
as a rule, not be considered.
In the instance of organic compounds, the six cubic and 18 hexagonal
substances listed seem to narrow the field of search close to final identifica-
tion. It may be expected that there are far more cubic and hexagonal
carbon compounds. Even in the group of the 6800 most common of them,
many solids are described as crystalline, needles, plates, leaves, powder,
P.14 Testing for Radioactive Decay 253

prisms, scales, rhombs without specifying the crystal class. Lattices of

low symmetry seem to predominate, however, and the lack of definite
statements indicates reluctance to form well-developed crystals which
could be recognized under the microscope. If this interpretation is correct,
the following lists are fit for practical use since they give the substances
which may be recognized as cubic or hexagonal with the use of the microscope.
Inorganic substances are listed by formula, and names or remarks
are placed in parentheses. Minerals are indicated by giving first the name.
Abundant and common minerals are emphasized by the use of bold face;
radioactive minerals by italics. Most of the minerals are either uncommon,
or rare, or very rare. Carbon compounds are listed by name and (in some
instances) formula. The temperature of melting or decomposition is added,
if known.

P.14 Testing for Radioactive Decay (770)

A test for the radioactivity of the sample is generally advisable in legal
investigations. Otherwise, if induced radioactivity or the presence of
artificial radioactive isotopes need not be considered, a test for radio-
activity is indicated if the substance is a mineral or an industrial product
which may contain uranium (yellow-orange glass or glaze, luminous dials)
or thorium (gas mantles) ; it is not necessarily without purpose if the sample
is a carbon compound.
Natural radioactivity is shown by the elements belonging to the three
series of radioactive decay: U 238 - Th 234, 230 - Ra 226 - Rn 222 -
Po 218, 214, 2lO - Bi 214, 210 - Tl210 - Pb 214, 210; 'Ih 232, 228 -
Ra 228, 224 - Ac 228 - Rn 220 - Po 216, 212 - Pb 212 - Bi 212 -
T1208; and U 235 - Th 231,227 - Pa 231 - Ac 227 - Fr 223 - Ra 223
- Rn 219 - Po 215, 211 - Pb 211 - Bi 211 - TI 207 and by the elements
K 40, Rb 87, In 115, La 138, Nd 150, Sm 147, Lu 176, and Re 187. (Italics
indicate a half-life of less than lO days.) The decay of the last-named
eight elements is very slow so that the radioactivity is low even with
Rb (P-), In (P-), Nd (P), Sm (IX), and Re (P), which contain a high ratio
of radioactive isotope. The naturally occurring elements K (P+ y) and
La (y) contain only a minute fraction of radioactive isotope. Lutetium (P)
need not be considered since it occurs in such small quantity that it will
not be detected unless a special search is undertaken. Concerning the
series of natural radioactive decay, the presence of anyone element of
long life will assure the presence of an equilibrium mixture of daughter
elements and strong radioactivity (probably IX-, p-, and y-radiation).
Any available apparatus may be used for the detection of radioactivity .
.Alpha radiation is readily recognized in a spinthariscope or with an electro-
scope; the Geiger-Miiller counter is most convenient for beta radiation
but must have a very thin window for recording low-energy beta rays or
254 Non-Destructive Testing P.ll)

must be charged with heavy gas to make it sensitive for gamma rays;
oscillation counters are able to handle all kinds of radiation including
X-rays. Shields are generally useful in classifying radiation according to
penetration power, and they are used in autoradiography also.
Autoradiography. Place the sample or part of it upon a thin sheet of
black paper or upon a perfectly flat sheet of the thinnest aluminum foil
available. In the darkroom, lay the sheet with the sample upon the emulsion
side of X-ray film as used by dentists, which rests on the bottom of a light-
tight can. Close the can, and set it aside for at least 12 hours, best at a
temperature of 0° to 5° C. Finally save the sample and develop the film
in a solution consisting of 1 volume Eastman print developer D-72 and
two volumes of water (434). The film will show a dark area where the
radioactive matter was located; inspection under the microscope may
reveal radiation tracks if the radioactivity was too low to produce general
grayness.
The sensitivity of the detection of alpha and weak beta radiation is
improved by placing the sample directly upon the photographic emulsion,
but this may produce darkening also because of phosphorescence or chemical
action. Use of special emulsions (Eastman Kodak NTB 2 for alpha particles,
NTB 3 for beta radiation) is needed to get records of the radiation tracks.
Prints showing the location of radioactive matter may be obtained by
placing thin sections or the polished faces of rock specimens upon the
emnlsion. The autoradiographs may be magnified by the customary
photographic enlarging procedure and compared with corresponding photo-
micrographs of the object.

P.15 Testing for Ferromagnetism

Testing for ferromagnetism may be useful if metallic or darkly colored
;particles are present. It must not be forgotten that, to-day, organic
materials also may appear magnetic as a consequence of incorporation
of magnetic oxides or metals (rubber, recording tape).
Spread granular material upon a sheet of glazed paper of contrasting:
color, and while watching the material, move the pole of a permanent
magnet along the underside of the paper. Ferromagnetic matter is recognized
by its motion, orientation along the lines of magnetic force, and attraction
to the pole of the magnet. If the sample is very small, observe it through_
a magnifying glass or microscope while approaching it with the magnet ..
It may be left in a container of thin glass; if it is transferred to a paper or
thin glass plate, which is placed under the microscope, it is practical to,
use a strong Alnico magnet in the shape of a knitting needle (about 1.2-mm
diameter and 12 cm long), which is easily used below the stage. A slurry
is tested by moving the pole of the magnet close to the surface of the-
liquid; a drop of it may be placed upon a slide or watch cover of thin glass,.
P.15 Testing for Ferromagnetism 255

which is placed upon the stage of the microscope and approached from
below with one end of the Alnico needle.
If ferromagnetic particles are present, they are best separated from
the unmagnetic material. The procedure has been described on p. 233.
Individual particles may be lifted out of the microscopic field with the
point of a magnetized needle and collected upon a square of glazed paper
or upon a slide, which has been placed over the poles of a strong magnet,
Fig. 71 d; glycerol may be used to prevent the glass plate from sliding
off the metal support. For alternatives, a fine wire of soft iron (reagent
wire meeting A. C. S. specifications, 0.23-mm diameter) may be used and
temporarily magnetized with an electric coil or by contact with a permanent
magnet. To the latter end, prepare a capillary pipet of the shape indicated
by Fig. 71 a and dimensions permitting insertion of an Alnico needle into

Fig.71. Use of Needle-Shaped Permanent Magnet for the Transfer of Particles

the wide tube. Insert a 0.5- to 0.7-mm length of the soft iron wire through
the tip which must be short so that the wire may contact the Alnico needle.
Using a flame of about I-mm diameter, heat a portion of the tip so that
the glass tube shrinks on the wire and holds it in place. Depending upon
choice, either seal the end of the tip to enclose the metal, Fig. 71 b, or
leave the end of the wire exposed, Fig. 7Ia. For the transfer of particles,
insert the Alnico needle so that it touches the wire in the tube. Make
certain that the tip of the tool is thoroughly clean and free from matter
that may act as adhesive. Pick up the particle as directed in Expt. 61,
and then move the tip of the tool close above the spot where the magnetic
particles are to be collected. On withdrawing the Alnico needle from the
tube, the particle should drop off the point of the tool. If this does not
happen, try collecting with the use of a strong magnet, Fig. 71 d. Obviously,
the technique may also be used for removing individual particles from a
slurry and depositing them in a drop of liquid for collection.
A more crude technique may be used if sufficient material is available.
One end of the Alnico needle may be pointed by dipping it into acid until
its diameter is reduced to the desired extent. The needle is then, pointed
end first, inserted into a glass capillary which has been sealed at the finely
drawn out end. A square, 4 mm X 4 mm, of glazed paper, plastic, or thin
256 Non-Destructive Testing P.16

sheet rubber is punctured at the center and pushed up upon the capillary
to a point about 2 cm from the sealed end, Fig.71c. After collecting
the particle at the sealed point of the capillary, the latter is then moved
so that the lower edge of the (paper) square is above the point of collection.
On withdrawing the Alnico needle from the capillary, the collected magnetic
material follows the Alnico magnet up to the square and then drops off.
The list of ferromagnetic substances is small: the metals iron, cobalt,
and nickel and some of their compounds; alloys of Fe, Co, Ni, Mn with Al,
Cu, Ag, Cr, Ti, W; the oxides Fe 3 0 4 , Fe 20 3 , C0 20 3 and their mixtures;
finally any other substances suitably compounded with ferromagnetic ones
to produce magnetic materials.

P.IS Odor
The sense of smell is about 25000 times as sensitive as that of taste.
One may be hesitant about using the latter, and tasting small samples
is completely out of' question because of the loss of material. One should,
however, not fail to note the odor of the sample for investigation, which may
indicate the presence of a substance that other methods cannot detect
because of lack of sensitivity.
All identification by smell is naturally based upon familiarity with
the odor of the substance concerned or upon comparison with known
samples. This fact is clearly indicated by the adjectives describing odors,
which nearly always refer to things or conditions: fresh, sweet, fragrant,
balmy, spicy, aromatic, flowery, sour, acrid, burnt, oily, rancid, earthy,
stale, moldy, musty, foul, fetid, putrid, etc. as well as by the more close
descriptions with reference to the origin: odor of rose, geranium, mint,
pine, bitter almonds, onions, garlic, vinegar, old sherry, stale beer, rotten
egg, burnt rubber, burnt flesh, decay, and untold other things more or
less closely specified.
Some obvious disadvantages of identification by odor are derived
from the high sensitivity of the olfactory organs. One may be mislead
by the odor of incidentally present trace impurities which may have been
introduced by the handling of the sample. The reasoning may be speciously
based upon the perception of trace impurities responsible for the "odor
of illuminating gas" (odor of intentionally added additive), "odor of
acetylene" obtained from calcium carbide and water (od<?r of phosphine),
or "odor of coumarin" (odor of unknown contaminant ~). Furthermore,
the response of the sensing organ shows individual differences and depends
greatly upon the health and condition of the mucous membranes. Fatigue
may be explained by a lack of response of the olfactory nerves when the
membranes have become loaded with absorbed vapors. It may be connected
with the well-known facts that the odor of some substances is not perceived
when they are given in high concentration, that the odor changes with
P.l6 Odor 257

concentration, and that the odor is modified by the presence of other

vapors as well as by preceding exposure of the membranes. To properly
function, the olfactory organs must be used with caution and be given a
chance to recover after each severe assault with strong odors by a rest
in fresh air until the mucuos membranes have lost the absorbed vapors.
It should hardly be necessary to add that testing for odor should be done
in clean air (and in the absence of deodorants).
As a rule, the observation of odor will be a casual one and happen
during the first inspection of the material under investigation. Any odor
of the packaging material, paper or plastic sheet containing the sample
should be noted. When opening a container, the odor of the interior should
be observed before its atmosphere is replaced by fresh air. A stopper or
cap may have absorbed vapors, and also its odor should be noted.
A deliberate test may not be practical when only a small amount of
substance is available and its odor is not very strong so that its perception
requires an atmosphere saturated with the vapor.
A test may be carried out by placing some of the material into a perfectly
clean and odorless vial which is closed with a glass stopper. After the
lapse of time sufficient to allow saturation of the gas space with the vapor
of the substance (standing over night at room temperature), the vial is
then opened in clean air for smelling its contents. A simple calculation
shows that the procedure may not be advisable for the investigation of
very small samples.
If the vial is given a capacity of only 2 ml, the mass m of evaporated
substance will approach
m = 10-7 P W gram,
where W is the molecular weight of the substance and p its vapor pressure
at room temperature in millimeter mercury.
In the instances of chloroform, acetone, and esters (p ~ 100 mm),
this would cause evaporation of more than 1 mg, a loss which would not
be tolerable on the milligram scale. When working with microgram samples,
the procedure would not be permissible with substances like naphthalene
and aniline (p ~ 0.1 mm, loss of 1 to 2 flg), or iodine, phenol, cresol, benzoyl
chloride, and camphor (p ~ 1 mm and evaporation of about 12 flg) or
acetic acid (p ~ 20 mm and evaporation of about 0.1 mg).
These limitations apply also to chemical tests based upon the appearance
of characteristic odors. The identification of acids or alcohols by the odor
of their esters does not appear inviting for the small scale. On the other
hand, not more than a few milligrams of the ester should be prepared since
the characteristic odor may not be recognized if too much of the vapor is
produced. The fragrance of acacia or orange blossom will not be perceived
in the laboratory where a milliliter of the oil is prepared, but it will be
noticed (by other people) on the aired clothing of the personnel.
Benedetti-Pichler, Identification 17
258 Non-Destructive Testing P.17

P.17 Hardness
Whereas the cohesion force in a properly annealed glass is the same in
all directions, the resistance offered by crystallized matter to mechanical
penetration by a foreign body depends significantly upon the direction
of the applied force. Crystals of various substances (mica, gypsum, etc.)
are quite readily split with a knife along certain cleavage planes which
are always parallel to a possible face which mayor may not be shown
by the particular crystal (usually prism, pinacoid, dome, cube or octahedron).
The perfection, smoothness, and gloss of the resulting facet is determined
by the ease of cleaving and also characteristic for the substance.
Hardness may be variously interpreted to mean ability to withstand
pressure, shear, wear, etc. and the data obtained depend upon the choice
of the test method. Good reproducibility is given by methods which
employ a standard point attached to a lever so that it may rest under
a known (and adjustable) load upon a smooth, horizontal surface of the
test object. The depth and width of the scratch obtained when the latter
is moved sideways are measured under the microscope. The test is performed
with a very fine point under the microscope to show the hardness of the
individual grains visible in polished surfaces of rocks, ores, alloys, etc. (90).
UYTENBOGAARDT (109) lists the hardness obtained in this manner, crystal
system, reflectivity, color, reflection pleochroism, behavior on etching,
and various other characteristics for the microscopic identification of ore
minerals, and BOWIE and TAYLOR (1054) claim that the task may be
accomplished in minutes. Experience must be assumed since observations
with polarized light on opaque materials do require it (169).
MURDOCH (101) recommends performance of scratch tests on polished
surfaces (thin sections of minerals) by means of the point of a No_ 10
Sharp's needle which is attached at an angle of 30 degrees at one end of
a handle that is 12.5 cm long and weighs 7 g. Materials are classified as
soft if the mere weight of the handle, held near the middle, suffices to
scratch them. If pressure has to be added to obtain scratching, the material
is called medium hard; hard materials are not scratched by the steel needle.
In his tables, the minerals are classified first according to color (colored,
white, and gray) and secondly according to hardness; reagents applied
with a platinum loop to individual grains (169) lead finally to identification.
For classification tests serve 8-F HN0 3 , 6-F HOI, 20 % KCN, 1.5-F FeCI3 ,
10-F KOH, 0.2-F HgCI 2, aqua regia, and 3 % H 20 2 0

SHORT (169) finds too many overlaps between Murdoch's medium

and soft grades of hardness; a needle point loses its sharpness very quickly,
and some minerals show considerable variation of hardness depending
upon the crystallographic direction of the scratch. Consequently, he
divides the minerals into soft ones, which are readily scratched by the
P.17 Hardness 259

needle, and hard ones, which are not scratched or only with difficulty.
His determinative tables use the classes "soft" and "hard" and a sub-
division into isotropic and anisotropic; etch tests lead to the final decision
as in Murdoch's scheme.
If a large specimen is available for investigation, one may use the
hardness scale of MORS: (1) talc, soft to the finger nail and greasy to the
touch; (2) gypsum, readily scratched with the finger nail; (3) calcite,
scratched by a brass pin; (4) fluorite, easily scratched by a knife; (5) apatite,
scratched with difficulty by a knife; (6) orthoclase, easily scratched by
a file; (7) quartz, with difficulty scratched by a file; Nos. (7), (8) topaz,
(9) corundum, and (10) diamond scratch "window" glass.
Especially when the type minerals are being used, one will try to scratch
the mineral with the type and vice versa. A scratch must not be confused
with a "chalk line" or streak produced by material rubbed off the scratching
tool; the latter may be removed by wiping without leaving a mark. The
color of the streak, which is the color of a fine powder of the softer material,
is also used in the identification of minerals.
Obviously, the distinction between soft and hard, below or above
Mohs' hardness number 5, is not sufficiently helpful if the material under
investigation may be any inorganic or organic substance. In addition,
there is no method of testing without risk of loss or contamination of a
small specimen. Even with very small samples, a test for high hardness
is advisable, however, if the material has a dark color, might be an abrasive,
might come from some grinding or cutting tool, or has a glassy appearance
combined with a high refractive index. The test is performed as follows.
Transfer a particle that may be spared to the center of a perfectly
clean microscope slide and place upon it a clean square of glass (25 mm X
X25 mm) cut from a second slide. Pressing down with the raser end of
a pencil, move the top square about 5 mm along the surface of the supporting
slide. Inspection under the microscope will show a scratch on the glass
surface which was sliding over the particle if the latter has a hardness
of 7 or more. The scratch may be surrounded by glass splinters. If the
material is soft, it may crumble under the pressure; if it is plastic, it may
show deformation. For recovery of the specimen, lift off the glass square
and place it, upside down, upon one end of the microscope slide. Inspect
the slide as well as the square, and when the specimen is found, collect it
with a needle as told in Expt. 61. If the specimen is transparent and
colorless, it may not be recognizable in the debris of glass splinters; the
latter are isotropic, however, and an anisotropic specimen will be recognized
between crossed nicols. After removal of the test material, wipe the glass
surfaces and confirm the presence of the scratch by inspection under the
microscope.
17·
260 Non-Destructive Testing P.IS

If the material under investigation scratches glass, it must have a

hardness of 7 or more, and the choice is reduced to the small list of Table VII,
rocks containing some of the listed substances, and certain of the newer
structural materials for high-temperature use such as Pyroceram (micro-
crystalline borosilicate glass of Corning containing Na, K, Mg, Zn, and Pb)
and cermets (metals in ceramic matrix and vice versa).

Table VII. List of Very Hard Materials

Hardness
Number of Material
MORS

10 Borazon, cubic boron nitride, hardest material known; diamond

above 9 WC, B C, SiC
9 Bromellite, BeO; corundum, AI20 a ; TiB; ZrB; WC
8.5 Chrysoberyl, BeAIO,; Burundum, nonporous high-alumina
ceramics containing 85% and more Al20 a ; TiC; ZrC; TiN
8 Picotite, (Fe, Mg) . (Al, Cr, Fe)204; pleonast, (Mg, Fe) . (Al, Fe)204;
spinel, MgAlzO,; topaz, Alz(F,OH)SiO,; ZrN
7.0 to 8 Beryl, 3 BeO . Al20 a • 6 Si0 2; gahnite, Zn(Fe )AIO,; hercynite,
FeO· Al20 a ; phenakite, Be2SiO,
7.5 Andalusite, Al 20 a • Si0 2; laurite, Ru(Os)Sz; zircon, ZrOz' Si0 2 ;
Mullite, synthetic 3 AI 20 a • 2 SiO z
7.0 to 7.5 Various rare silicate and borate minerals
above 7 Si, ferro-silicon, tool steels containing Si, Ti, Mo, W, Cr, Ni, Co
7 Quartz; vitreous silica; "hard" glass

For the detection of very soft materials, the test might be repeated
with two polished plates or sheets of brass or copper, which are not scratched
by materials of hardness less than 4. Finally, the unknown might be
tested between two glass plates that have been coated with a suitable,
soft (and colored) film such as collodion or gelatine.

P.18 Refractive Index

The ability to match refraction within 0.00002 by the use of phase
contrast or interference microscopy (476) may serve for batch identification,
but such precision is useless for the recognition of substances, where the
presence of incidental impurities will give far greater variations of refractive
index. In general, the determination of refractive index under the micro-
scope requires availability of an Abbe or a dipping refIactometer for
the checking of liquid standards and is only conditionally recommended
to the chemist.
If a liquid substance is given, which may be expected to be reasonably
pure and which is not noticeably volatile, the determination of refractive
index may be perfortned with a purpose and little loss. If 10 to 25,ul are
P.18 Refractive Index 261

available, the method of the Due DE CHAULNES (88, 422) may be used
with cells of 3- to 4-mm diameter and 1- to 2-mm height and give the
refractive index within ± 0.005 after calibration of the cell with standard
liquids as suggested by F. E. WRIGHT and described by CHAMOT and MASON.
The general idea applied in the cells of A. MOHRING (153) and L. NICHOLS (425)
should permit further refinement to reduce the required volume of liquid
substantially below 10 Itl. Both cells have been commercially available.
The immersion method, too, should allow to work with less than 10 It]
of sample, but will not always give three decimals for n as the cell of
NICHOLS is able to do; since variation of temperature is not advisable,
as a rule, the availability of a solid of closely similar refraction determines
the attainable accuracy. If necessary, Kofler's scale of glass powders
may be extended by adding MnSi0 3 glass (n = 1.700), spinel (1.718),
periclase, MgO (1.736), garnet (1.735), and arsenolite, AS 2 0 3 (1.755). The
final matching may be done with the use of a microscope hot stage (P. 21)
with the liquid containing a fragment of the solid of slightly lower refraction
confined in a sealed capillary. The refractive indices of the more common
organic liquids may be found collected in table form (10).
For the determination of the refractive index of very small solid objects,
only the immersion method is available. For lack of suitable immersion
liquids, this excludes the determination of refractive indices above about 1.8,
but still permits the study of organic solids and of approximately 70 per cent
of the minerals and inorganic solids, the refractive indices of which range
from 1.3 to 3.6. If need arises, a liquid with a refraction from 1.74 to 1.83
may be obtained by dissolving sulfur in methylene iodide. It is understood
that the solid must be transparent.
The mixture which perfectly matches the refraction of the solid should
be prepared upon a not too small scale so that evaporation of some of the
more volatile constituent cannot significantly affect the refractive index.
The solid particle should be washed with this mixture before it is immersed
in it for the observation in monochromatic light. Obviously, the immersion
media must not dissolve the solid or react with it in any way, and this
fact may have to be established by preliminary trials (P. 31, 32).
The application of the immersion method is relatively simple if the
solid is either isotropic or uniaxially birefringent (hexagonal or tetragonal)
so that a view vertical to the principal axis is readily obtained. Systematic
compilations of refractive index data are available for minerals (99),
inorganic substances (113), and some organic substances (112). In the
instance of organic solids, the effort is least rewarding, and it seems preferable
to determine the melting point and the refractive index of the melt as
outlined in P. 21.
The determination of the outstanding values for the refractive index
in the instance of anisotropic solids which are not obviously hexagonal
262 Non-Destructive Testing P.19
or tetragonal involves a study of the orientation of the optic axes and is
best undertaken by a crystallographer. The effort may be worth while
if the test material is of mineral origin, if well developed crystals are
available, or if destructive testing must be avoided by all means. Very
small crystals are best studied with the use of a universal stage. The axial
rotation stage (464) made by Kenneth A. Dawson Co. (Belmont, Mass.)
may serve as a substitute, and minute crystals might be mounted in a
droplet of Canada balsam at the end of a glass needle.
If a large number of approximately equidimensional crystals or crystal
fragments may be mounted in a suitable medium between slide and cover
slip, the axial figure may be studied on several specimens to find whether
the solid is uniaxial or biaxial, positive or negative. After suitable change
of immersion medium, the axial figures may again be used for determining
the optical orientation previous to use of the immersion method. With
uniaxial solids, no may be determined on specimens that remain dark
between crossed nicols when the stage is rotated, and ne is then that
refraction observed, which differs most from no. Similarly, {J of biaxial
solids is exhibited by specimens that remain gray during the rotation of
the stage, and IX as well as r are the extreme (lowest and highest) values
that may be found. If the crystals or fragments are small, use of an elaborate
petrographic stand and availability of special high-power objectives for
the observation of axial figures become desirable. Additional advice may
be found in the literature (88, 93, 96, 97, 110).

P.19 Density
Liquids. Provided that suitable balances are available, accurate
determination of the density of small amounts of liquid may be performed
with pycnometers or specific gravity pipets, and no loss of material need
be involved. Data on the density of liquids may be found in handbooks
and systematic compilations (54).
ANDERSON (444) described semi-self-filling Ostwald pycnometers of
I-ml capacity giving precisions from ± 0.0001 to 0.000025. The commercially
available specific gravity pipets of .ALBER (426) weigh about 5 g and are
provided with tight-fitting ground caps to prevent evaporation or uptake
of moisture or carbon dioxide during weighing. The decigram pipet has
a capacity of 0.1 ml and gives a relative precision of ± 0.0005. The centi-
gram and milligram pipets are made of heavy-walled capillary tubing of
uniform (I-mm and O.5-mm) bore and are provided with a millimeter scale
to allow use of amounts of liquid varying from 20 to 80,a1 and from 6 to
I6,a1, respectively; a relative precision of ± 0.005 may be attained.
The principle of the specific gravity pipet may be used with very small
volumes of liquid .. The success will depend upon vapor tension, stability,
P.19 Density 263

and viscosity of the liquid. The pipet of VON W ARTENBERG (153, 602)
will require less than l,ul if made from capillary tubing of 0.6-mm diameter
of bore so that the length of the bulb does not exceed 1 to 2 mm. The fine
capillaries, Fig. 72a, may be 15 to 20 mm long and have a bore of about
0.03 mm. The microbalance may be provided with a wire rack for holding
the pipet. Essential is that liquid and pipet have, at all times, the temperature
of the room; to avoid warming the apparatus, it should be handled with
forceps having flat tips covered with a soft sheet that does not leave material
on the glass surface. Small pieces of soft rubber or plush are cemented
upon the tips and then trimmed with scissors; a quickly drying cement

----~~bi:lj=::::
~ c

------<~ if d

Fig. 72. Determination of Density. The bore is greatly exaggerated in e

which does not penetrate the sheet should be used. To hasten the intake
of liquid, the pipet may be connected through a drying tube to the suction
line so that the bulb rests on the opening of a stopper or rubber tubing
mounted on the end of a glass tube. When the liquid reaches the upper
capillary, the connection to the suction is broken so that the tube may
fill by capillary attraction. The adjustment of the volume is automatical,
but the liquid adhering to the outside of the intake capillary must be
removed by touching to the edge of spot test paper. Even this task may
be avoided or simplified by treating the outside of the apparatus with
Dri-Film or Desicote before cutting the fine capillaries to proper lengths.
Evaporation during weighing may be suppressed by inserting the pipet
into a somewhat wider capillary, Fig. 72b. Water may serve for calibration.
Obviously, all rinsing and drying must be done with use of suction.
Von Wartenberg used a Nernst balance for a pipet of 4-,u1 capacity and
obtained a relative precision of ± 0.003. The arbitrary scale of the balance
was calibrated by weighing the pipet filled with several different liquids
of known density, but calibration by customary procedure (142) should suffice.
264 Non-Destructive Testing P.19
The density of down to 0.02,tt1 of liquid could be estimated by introducing
drops of about I-mm length into capillary cones of 0.2-mm bore by the
standard procedure, p. 208, and by taking special care that the tip of the
micropipet will not touch the cone anywhere but at the point where the
drop is to be deposited. The fine capillary from which the cone is to be
made should be inspected for uniformity and roundness of bore by cutting
out sections of 20-mm length and measuring the diameter of bore at both
ends, Expt. 22. To check on the circular shape of the bore, four diameters
are measured by rotating the stage through 45 degrees after each measure-
ment. A satisfactory section of capillary is used for the preparation of
the capillary cone. After introducing the liquid, the capillary cone may
be placed inside a somewhat wider cone, Fig. 72c, to reduce evaporation
during measuring the length h of the liquid column and weighing. If the
menisci are spherical or nearly so, the volume, v = 'Tt r2 (h +: r), may
be computed as that of the cylinder of height h increased by that of a
cylinder of height 2 r minus that of a sphere of radius r; Fig. 72e. The
radius may be checked at the conclusion of the experiment by cutting
the capillary cone where the drop of the liquid had been located. A relative
error of 0.01 in measuring the diameter of bore will give an error of 0.02 in
volume and density.
Solids. Density data are available for minerals and common inorganic
solids, and the larger number of these substances has densities between
1 and 4 gjml so that the available heavy liquids permit accurate determina-
tions. In addition, the densities range from about 0.5 (Li) to about 22 glml
(platinum metals) so that even crude estimations will permit to distinguish
between light metals (Be 1.8, Mg 1.74, Al 2.7), the common metals (Fe 7.86,
Co 8.9, Ni 8.9, Cu 8.9, Ag 10.5, Zn 7.14, Cd 8.6, Sn 7.3, Pb 11.3, Sb 6.68,
Bi 9.8), and the heavy metals (Ta 16.6, Au 19.3, Pt 21.4, Ir 22.4 - W 19.3,
U 18.9). On the other hand, the density of organic solids varies only from
about 0.8 to 4.3 glml, and few data are listed so that knowledge of density
will rarely permit identification.
The density of large objects is conveniently and accurately determined
by weighing in air and in a suitable liquid of known density. Determination
of volume by this method is not applicable to small objects, however,
because the buoyant effect becomes too small as compared with the forces
developing where the suspending filament enters the surface of the liquid.
The density may be found by weighing the specimen and computing
its volume from its linear dimensions. BRILL and EVANS (720) obtained
a relative precision from ± 0.015 (bead of tin) to ± 0.001 (crystals) with
specimens of 0.06 to 1.5,ttl volume (lineal dimensions of 0.5 to 2 mm).
Far smaller specimens will suffice if a suitable microbalance is available
and magnifications are used permitting a relative precision of ± 0.01 or
P.19 Density 265

better in the lineal measurements which may be carried out on screen

images or enlarged photomicrographs. The precision of volume and
consequently density will greatly depend upon the shape of the specimen.
If it has a simple form with known exact relation between dimensions
and volume, the latter may be accurately computed from a few simple
measurements of length; this holds for sphere, ellipsoid of rotation, cube,
prism, pyramid, tetrahedron, octahedron, cylinder, and cone. Well developed
crystals may represent a simple geometry. Noble metals may be fused
to a spherical bead by fusing them suspended in a bead of boric acid,
carried on the end of a fiber of vitreous silica (1200). A flux of KCN or
NaCN or heating in hydrogen may be tried with metals that readily oxidize.
The diameter of metal wire may be measured in the side view, but as in
the instance of fibers and transparent filaments, the cross section should
be inspected to be certain of its nature. Deviation from sphericity is serious
with beads since the relative error committed in estimating the diameter
appears tripled in the volume of the sphere.
HABER and JAENICKE (1200) tried to obtain the density of small metal
beads from the rate of fall through paraffin oil, but difficulties arose when
the diameter was less than 30 flm, volume less than 1 pI or 10-12 liter.
The rate of fall through liquid as well as the buoyant effect or the
displacement of liquid (volumenometric method) are affected by adsorption
of gas on the surface of particles or its presence in cavities or cracks.
Consequently in all these methods, care must be taken to eliminate gas
adhering to the specimens by the use of wetting agents or application of
the vacuum impregnating technique. The surface tension of water may be
reduced by adding some ethanol, or about three drops of Triton 100 per
50 ml fluid (300), or "Anti-Creep" according to the directions of the
Schleicher and Schuell Co.
Volumenometric methods require a balance having sufficient precision
for the weighing of the specimen which is then immersed in a fluid contained
in a calibrated tube; the volume indicated by the displacement of the
meniscus is the volume of the specimen. CALEY (400) describes a calibrated
cylindrical centrifuge tube which permits attaining a relative precision
of about ± 0.005 with 25 to 100 mg of powdered solid. Coarse powders
are preferable since fine particles tend to float on the surface. Caley
recommends using diethylether and to keep the tube stoppered to prevent
evaporation. It is understood that a liquid must be chosen, which does
not dissolve the solid or react with it, and this may require a preliminary
study of the solubility, P. 31, 32. KIRK (157) has applied the method to
small particles which could have a diameter as small as 0.1 mm (0.5 flg
weight) and got a relative precision of ± 0.1 or better. A 15- to 20-mm
section of thin-walled capillary of uniform circular bore (see above p. 264),
slightly larger than the diameter of the particle, is selected for preparing
266 Non-Destructive Testing P.19

a capillary cone, Fig. 72d. To obtain a fixed mark, the outside of the
capillary cone may receive a very fine scratch with a diamond pencil,
or a short length of a straight textile fiber may be cemented to the outside
so that it lies parallel to the axis of the tube and either end may serve as
reference. Using the standard procedure with the micropipet, p. 208,
water, butyl phthalate, or any other suitable liquid is introduced into
the capillary cone until the meniscus arrives close below the mark. If it
seems desirable, the tube is centrifuged before determining the distance
between meniscus and reference mark with the eyepiece micrometer. The
air of the "moist" chamber may be saturated with the vapor of the liquid
used, and the customary precaution may be taken to prevent evaporation
during centrifuging, p.211. The weighed solid is introduced by means
of a glass fiber or microforceps, p. 232, operated by a mechanical manipulator,
while the capillary cone is resting upon the carrier slide in the chamber.
The particle is deposited in the opening of the capillary cone and then
centrifuged into the liquid, whereafter the position of the meniscus is again
determined against the reference mark. The volume is computed from
the displacement of the meniscus and the known diameter of the bore.
The accuracy may be improved by multiple performance of all measurements
and use of the averages.
In the buoyancy method, the composition of the fluid is changed until
the solid specimen remains suspended without rising or falling; hereafter,
the density of fluid is determined by a standard method. Since a large
volume of liquid may be used, its density and thus the density of the
floating particle, no matter how small, may be determined with high
accuracy. Obviously, the temperature must be controlled within the limits
required by the accuracy of the determination of density. The range of
applicability, however, is limited by the density of the available heavy
liquids, a good list of which may be found in Lange's Handbook of Chemistry.
Mixtures of benzene (d, 0.88), nitrobenzene (1.20), carbon tetrachloride
(1.59), bromoform (2.89), and methylene iodide (3.33) cover a range from
0.88 to 3.33 giml, and this may be extended to 3.65 giml by dissolving
iodine and iodoform in methylene iodide. The solution of ROHRBACH is
prepared by dissolving 20 g BaI 2 and 26 g HgI2 in less than 6 ml water;
it has a density of 3.58 and is diluted first with a 20 % solution of BaI2 •
When an equal volume of the BaI 2 solution has been added to prevent
separation of HgI2' further diluting may be done with water.
For accurate determination of density, a coarse powder should be used
in preference to large pieces. ANDRAE (153, 1215) selected under the
microscope perfect crystals of I-mm diameter and somewhat less. These
were transferred into a test tube where the mixture of methylene iodide
and benzene was prepared in which the crystals floated at room temperature.
The floating of the heaviest particles should be taken as criterion since
P.19 Density 267

the lighter ones may be carried up by adhering air; at this stage, the
mixture also might be exposed to reduced pressure in the hope of displacing
adsorbed gas. ANDRAE transferred the crystals and liquid of equal density
to a dilatometer with a calibrated capillary of 2-mm bore. When the
dilatometer was nearly full, the small remaining air bubble was used for
thoroughly mixing the contents, and then the filling of the apparatus was
completed. The dilatometer was mounted in a large beaker with water,
the temperature of which was regulated until the crystals remained
stationary in the liquid as judged by the immobility with respect to reference
marks on the outside surface of the beaker. When this condition was
reached, the temperature was read and the volume of the contents was
determined by reading the position of the meniscus in the capillary. For
getting the weight of the contents, the capillary may be capped, but the
apparatus must be wiped dry on the outside and allowed to acquire the
temperature of the balance room. To obtain accurate results, it is necessary
to mount the dilatometer in the water bath so that it may be rotated
around its axis and that the temperature may be changed so slowly that
gradients will not be established. Obviously, the weight must be corrected
for the buoyant effect, and the volume of the dilatometer for the expansion
of the glass.
KIRK (157) prepares the mixture of like density in a test tube of about
6-mm bore and 5 cm length. If the particle to be tested is somewhat
porous, only little fluid is added and the pressure above the liquid is reduced
so that boiling occurs which will aid in eliminating adsorbed gas. After
each addition of heavier or lighter liquid, the contents of the tube are
mixed by twirling in it a glass "zigzag stirring thread containing a number
of successive sharp bends". When the specimen, which may also be a
drop of immiscible liquid, remains suspended without motion, some of
the mixture is taken into a specific gravity pipet for the determination
of the density, p.262-3.
HUTCHISON and JOHNSTON (701) refined the buoyancy method so that
the density could be determined within ± 0.000005 g/ml with fragments
of 3- to 4-mm edge. The precision was determined by the temperature
control. Other factors-particle size, viscosity of liquid, length of observa-
tion, heat diffusibility of system-become dominant with smaller particles,
and PRIMAK and DAY (454) accepted relative precisions from ± 0.0001 to
± 0.001 with crystal fragments 0.25 mm on edge, i. e., about 4000 times
smaller than those used by HUTCHISON and JOHNSTON.
Because of the high toxicity, special precautions should be taken for
the prevention and collection of spillage when working with "heavy liquids"
and especially when using thallium salts.
HENDRICKS and JEFFERSON (745) as well as BERNAL and CROW-
FOOT (1050, 1051) centrifuge the fluid with the immersed specimen to find
268 Classification Tests P.2c}

the mixture of like density, and it is claimed that a particle of 50 f-lg will
suffice for the determination. Finally, it should be mentioned that separa-
tion of the constituents of a mixture (rock) may be carried out by sedimenta-
tion from a liquid. The mixture is ground to a size so that individual
particles, for the most part, contain one component only. The powder>
which should be as coarse as possible, is then transferred to a centrifuge tube
and treated with a liquid that is so dense that only one component may
settle out. After centrifuging, the liquid with the light components is
transferred to another tube and mixed with a lighter fluid until the next
component drops out, etc. Use of acetylene tetrabromide (d, 2.96) is
recommended for the separation of ore and gangue; among others, it will
float quartz, d = 2.65.

P.20 Classification Tests

As indicated by the name, the following procedures will permit recognizing
the general class or group of materials to which the unknown belongs;
they also allow recognizing a number of the more common substances.
Some of the tests are destructive, and the others may lead to destruction
in certain instances. Thus, if very little material is available for investigation
and cannot be replaced, it becomes advisable to consider once more whether
or not identification is possible by non-destructive means. A review of
the collected evidence may indicate a way. H X-ray diffraction and electron
probe are either not available or do not give the needed information, one
may decide to try identification via the optical constants of the crystal.
This is promising in the instances of minerals and inorganic materials (83, 94,
99, 113, 114, 921) and possible with organic substances (112). One may
decide on first determining the solubility (P. 32) on a very small fragment
of the specimen and to use the knowledge gained for the preparation of a
medium for spectrophotometric testing (compare P. 34) or for re-
crystallizing (448) the specimen for crystallographic investigation. Slow
evaporation of the solution from a flat dish, 30-mm diameter at the bottom,
40-mm diameter at the top, and 20 mm high, may give well-developed
crystals since the undesirable concentration gradient toward the edge of
drops on a slide is missing (924). The systematic procedure for identifying
mineral particles is outlined also by FRY (93) and by KERR (97).
A technique for the performance of a series of orientation tests on a
particle, 0.1 mm or less in diameter, is described by PRAZAK (969). The
substance may be heated in various gases (P. 27 to 30) and finally tested
for its solubility (P. 32).
The apparatus, Fig. 73, consists of a short length of glass tubing,
about 6 mm in bore, which is drawn out so that a bulb is left between
two short pieces of capillary. If ignition is contemplated, the apparatus
P.21 Observation of Transition Points Below 350 0 C 269

may be made of combustion tubing or vitreous silica. Capillary a may be

0.5 mm in bore or less; capillary c should have about I-mm inner diameter
-or more.
The particle to be investigated is introduced into bulb b through the
wide tube d by means of a glass thread or a fine capillary.
"For a simple ignition test, the tube is tilted so that capillary a points
up and a microflame is applied to the bulb. If water or another volatile
substance is liberated, it will condense near the outlet a of the bulb and,
by cautious application of heat, it may be driven into capillary a for
collection and (or) testing. Any desired gas may be supplied through d
and gaseous products of the reaction may be collected by inserting capillary a
into a droplet of absorbent."
"For solubility tests, a small amount of solvent is allowed to enter
capillary a, whereafter it is sucked into the bulb and brought into contact
with the particle to be tested. The action may be observed under the

----------~~=====-
ti c b
Fig. 73. Tube for Classification Tests and Dissolution of Sample; approx. nat. size.
PRAZAK, G., in Mikrochimica Acta 1961, 899

microscope. If the particle dissolves, the solution may be blown out

through capillary a and analyzed. If dissolution does not seem to occur,
the solvent is evaporated by heating the bulb which should be inclined
so that d points up. A suitable gas may be supplied through a by connecting
d to a suction line. Thus it is possible to try a series of solvents: water,
dilute and concentrated hydrochloric acid, dilute and concentrated nitric
acid, aqua regia, etc. Agitation may be obtained by tapping the bulb."
"An insoluble residue may finally be heated in a current of oxygen
(formation of CO 2 indicating carbon, organic substance, or carbide) or
in a current of hydrogen (reduction of insoluble sulfates and of insoluble
compounds of heavy metals: AgCI, Sn0 2 , etc.). Formation of water may
be simply demonstrated by directing the gas escaping from a onto a small
crystal of copper sulfate which has been heated until it turned white
(anhydrous salt)."
If material is available for destructive testing, the following "chemical"
procedure, which does not require specialized instrumentation, will be
found quite efficient. It may be used with very little material without
becoming cumbersome.

P.21 Observation of Transition Points Below 350 C 0

Test for Stability Upon Heating. The possibility of the presence of

~xplosive substances in the unknown must be excluded with certainty
270 Classification Tests P.21

before an amount of the unknown exceeding a few tenths of a milligram

may be heated without fear of disastrous consequences.
To test for stability, place not more than one milligram of the unknown
substance into a thin-walled test tube. Mount the tube behind a pane
of safety glass, and shove a lighted Bunsen burner under it so that the
substance is rapidly heated by a just non-luminous flame of about 8-cm
height. The unknown may be considered safe if neither explosion nor
deflagration occurs.
Observation of the Behavior on Heating. If reviewing of the already
established knowledge concerning the unknown does not suggest otherwise,
the testing is best continued by observing the behavior of solid material
under investigation while the temperature is raised until melting occurs
or decomposition sets in. If a cooling stage is available, liquids may be
frozen, and the behavior of the resulting solid studied in a like manner.
Solids too, are best studied with the use of a heating device mounted on
the stage of a microscope. The apparatus must be constructed so that a
thermometer or thermocouple indicates the temperature of the specimen (98,
159). A heating block and observation with a magnifying glass may be
used as a substitute, but they do not permit getting all of the information
which is obtained with the microscope hot stage. Finally, a melting point
determination may be carried out by the classical method with capillary
attached to a thermometer and inserted into a well-stirred bath liquid.
Heating to 350 C will not give any information if the material is
0

very hard, Mohs No.7 or above. Useful information may be expected

when the material is relatively soft and distinctly crystalline; in such
instances, the experiment should never be omitted if an organic substance
or an inorganic hydrate may be present.
Microscope hot stages and their use have been described by LUDWIG
and ADELHEID KOFLER (159, 160) and by MCCRONE (163). Apparatus is
commercially available, and instructions added by the manufacturer may
contain important information. Special attention must be paid to the
rate of heating since the data given in the literature assume adherence
to a standard procedure. It is general practice to reduce the rate of heating
when approaching the melting point; a rise of 4 degrees per minute is
permitted by the KOFLERS. In addition, it must be kept in mind that
the rate of heating will greatly affect the temperatures at which loss of
water of hydration, distillation, sublimation, and other transitions are
observed. The amount of loss by vaporization and the amount of decomposi-
tion, and consequently the melting point, are also determined by the
rate of heating. Thus, if nothing is known concerning the behavior of
the material upon heating, it may be best to sacrifice a very small amount
of sample for an exploratory experiment in which the heating is done
quite rapidly.
P.21 Observation of Transition Points Below 350 0 C 271

For heating on the microscope hot stage, use less than 0.1 mg of material
and crush larger particles or crystals by pressing between two glass slides.
Place the material so upon the slide that the individual particles do not
touch one another. Use a slide of the thickness suggested by the manufacturer
of the heating device and cover the material with a clean glass slip. Slightly
press upon the cover slip and impart a small rotary motion to distribute
the solid between the glass surfaces and to break up larger particles so
that the cover slip lies close to the surface of the slide. Place the slide
upon the microscope hot stage and move it until a suitable portion of the
preparation is in the field of vision. If several types of particles are
present-different color or transparency, crystals of different shape-make
certain that all types are included so in the field that they are not in contact.
Proper selection of the field is important at this time since most of the
apparatus do not make provisions for moving the slide when the heating
is once started. On the use of dark-field illumination see FELTON (488).
Finally place a glass bridge over the preparation if the directions
call for it, cover the hot stage, and adjust the heating rate in accordance
with instructions. Continuously observe the preparation until either all
particles have melted, or decomposition sets in, or the temperature limit
of the apparatus has been reached. Whenever a change occurs in the
preparation, record it together with the temperature. Use of polarizing
equipment may facilitate to observe transitions from the appearance,
change, and disappearance of interference colors; a compensator (A.j4-mica
or first-order red plate) used together with crossed polars gives sufficient
brightness to permit simultaneous observation of isotropic matter. If
projection is used for observation, be certain to insert infrared-absorbing
cells between the source of light and the preparation; otherwise transition
temperatures will be found too low since strongly absorbing particles
may acquire a temperature noticeably higher than that indicated by the
thermometer.
The following phenomena may be observed: (a) sublimation or distilla-
tion with the growing of new crystals or the appearance of droplets;
(b) crystallization of the distillate to give a phase stable at high temperature;
(c) color changes; (d) clear particles becoming opaque because of dehydration,
decomposition, or transition to another modification; (e) melting; (f) crys-
tallization of a new compound (anhydrous substance) or of a high -temperature
modification from a melt; (g) several sharp melting points if either different
substances are present, or the melting points of the hydrate, the anhydrous
substance, or of several modifications of one substance; (h) decomposition,
sudden or gradual, possibly accompanied by melting, boiling, discoloration,
charring, evolution of a gas, or separation of a solid.
Record the temperature interval in which the melting occurs since it
is a criterion of the purity of the substance. Usually, the smallest fragments
272 Olassification Tests P.21

liquify first; then the larger particles show a rounding-off and become
surrounded by melt. The temperature, at which the last trace of solid
dissolves in the melt, is closest to the melting point of the pure substance.
To observe the equilibrium between liquid and solid phase, stop the heating
when the particles begin to melt. After a short time, the remaining crystal
fragments will begin to grow in the melt, and by turning the heat on and
off, the temperature corresponding to equilibrium between the last trace
of solid and the liquid may be determined quite accurately.
In the evaluation of data, it is useful to know whether the unknown
substance is organic or inorganic in nature. This information, however,
may not be available at this time, and the evaluation may have to be
postponed (P.34) until the behavior on heating above 300 0 C has been
tested (P.22 to 28).
Inorganic Substances. The more common inorganic substances which
melt below 900 0 C are listed in Table 4. For convenience, they are
arranged in the order of their melting points; in some instances, transition
temperatures other than melting points have been used. The data are
taken from the literature, mostly from Lange's Handbook of Chemistry.
The procedure of arriving at the table implies that there are about 2000
of the more common inorganic compounds which do not melt below 900 0 C.
When using Table 4, proper allowance (98, 159, 160, 163) has to be
made for the fact that, as a rule, the listed transition temperatures have
not been determined under the microscope but either in the capillary or
with large amounts. Whenever doubt arises, an apparent identification
may be confirmed by observing the behavior of the known substance
having comparable purity and by performing a mixed melting· point.
This procedure will also resolve doubts concerning the accuracy of the
listed data. Obviously, the significance of the zeros is doubtful for most
transition temperatures given as 50, 80, 100, etc.; the mere fact that
relatively large numbers of substances are reported to melt at 30, 40, 50,
lOO, 110, 120, etc. calls for caution. The desirability of a revision and
expansion of the table after an experimental study of the behavior of the
substances on the microscope hot stage is obvious. It need hardly be
added that the author would be grateful for any information, corrections,
and references to published data.
As further aid in identification, Table 5 lists the more common inorganic
substances that sublime, and Table 6 those which burst into flame when
heated in air. It is generally known that sublimation temperatures vary
widely with the conditions of the experiment; tables usually give the
approximate temperature at which rapid sublimation starts when the
substance is heated without any special precautions.
Explosives. Table 7 gives a list of the explosive solids, extracted from
tabulations of the more common 2600 inorganic and about 6800 organic
P.21 Observation of Transition Points Below 350° C 273

compounds (10). Potential explosives are furthermore all mixtures of

oxidizable matter (S, P, P 4 S 3 , P 2 S3 , P 3 SS ' P 2 SS ' AS 2 S 2 , Sb 2 S3 , FeS 2 , Mg,
AI, C, organic substances) with oxidants such as liquid air, peroxides,
chlorites, chlorates, perchlorates, bromates, iodates, periodates, per-
manganates, nitrites, nitrates, and others. Intimate mixing is frequently
an important factor; on the other hand, the components may be solid,
liquid, and gaseous to give explosive emulsions, suspensions, sprays, and
dusts. Explosions of air-born combustible dust occur in coal mines and
flour mills.
Black powder is an intimate mixture of C, S, and KN0 3 which is usually
formed into definite shapes varying from small grain to large cubes. It is
very sensitive to heat, friction, and mechanical shock. The chemical
analysis of explosive mixtures usually presents little danger after the
oxidant has been separated from the fuel, which may be frequently
accomplished by extraction with water.
Small amounts of cellulose nitrate (nitrocellulose, gun cotton, collodion)
and smokeless powder deflagrate, i. e., decompose suddenly with a flash
of flame, when heated to 100° to 200 C, but do not explode. Smokeless
0

powder is obtained by treating nitrocellulose with nitroglycerin and giving

the resulting jelly the shape of flakes or grains.
Dynamite is obtained by absorbing nitroglycerin in suitable solids
(wood flour, diatomaceous earth, etc.) to obtain a plastic mass. It is
safer to handle than nitroglycerin itself, but will explode when heated.
Quite complex mixtures have been prepared in adapting explosives
to special uses. They may contain explosives (gun cotton, nitroglycerin,
nitroglycerin incorporated in a glue jelly, TNT, nitronaphthalene, dinitro-
benzenes, NH4 N0 3 ), oxidant [KN0 3 , NaN0 3 , NH 4N0 3 , Ba(N0 3 }2, Sr(N0 3 }2,
KMn04' K 2Cr 20 7 , K 3Fe(CN}s], fuel (S, C, AI, naphthalene, chioro-
naphthalene, kerosene, fatty oil, sugar, starch, wood flour, cellulose, resin),
moderator [NH4 CI, (NH.)2S04, aniline hydrochloride, NaCI, Na 2C0 3 •
. 10 H 20, magnesium sulfate], binder (kerosene, soap, glue, dextrine,
resin, starch), and absorbent (diatomaceous earth, flour, wood flour,
cellulose, etc.).
Organic Substances. If the substance under investigation is an organic
compound, consult suitable melting point tables and compile a list of
substances that have to be considered.
The KOFLERS (159, 160) and MCCRONE (163) list the substances in the
order of their melting points as observed upon the microscope hot stage.
These convenient tables may also be used for a preliminary orientation if
the melting point has been determined in the capillary, but in this case
it will be wise to include into consideration all substances melting in the
range from t observed to t +
10. For final identification, it may be necessary
to compare with listings of melting points determined in the capillary (2, 8,
Benedetti-Pichler, Identification 18
274 Classification Tests P. 22, 23, 24

10, 54). Tables listing melting points in the order of ascending values
may also be found in handbooks (10).
Residues from melting point determinations may be reserved for use
in heating to higher temperatures and for chemical analysis.
The heating on the microscope hot stage may be repeated with fragments
of the material under investigation immersed in paraffin oil (159, 160).
This will show whether or not a gas is given off on heating. In general,
however, it will be more efficient to get this information by heating in
a capillary, P. 26, which will permit identifying the nature of the gas.

P.22 Ignition Above 300 0 C

The material under investigation may be heated by itself in an essentially
inert atmosphere or with the addition of various reagent, solid or gaseous,
and by using various techniques depending upon the amount of material
available.

P.23
Heating in the Closed Tube. Customarily, a few milligrams of the
substance are placed into a test tube of hard glass, which is about 4 mm
in bore and 10 cm long. The procedure is described in Expt. 15.
If little substance is available, a narrower tube may be used, and a
particle of only a few micrograms weight might be heated in the sealed
end of a capillary tube of vitreous silica. Heating in a stream of inert gas
(see below) will be preferable, however, when dealing with small amounts
of substance. For interpretation of the observed phenomena see P.27.

P.24
Heating in an Open Tube. Description of the procedure is given in
Expt. 16. Condensates collect in the cooler part of the tube above the
substance; liquids may be removed with a capillary pipet, and solids may
be tested after the tube has cooled and has been cut into suitable sections.
Pointed strips of reagent papers and loops or capillaries with reagent drops
are applied at the upper opening of the tube to test for escaping gases.
Liquid reagents are best applied with an elliptical loop, about 10 mm long
and 4 mm wide, which may be formed at the end of a glass rod of 0.5- to
I-mm diameter, P.36. Colorations are seen by holding the loop in front
of a brightly illuminated, white surface. To test for turbidity, inspect
the drop in the loop before and after exposure to the gas in front of a black
background with strong light coming from the side.
To test for acidity, use wide-range pH paper. Record the odor of the
escaping gas, and test it with limewater (saturated solution of calcium
hydroxide), fuchsin paper, 1 % solution of silver nitrate, and KI-starch
paper. Additional reagents may be used, and for interpretation see P. 28.
P.26 Ignition Above 300 C0 275

To perform the test with small quantities of substance, use the technique
of P.26 for heating in oxygen or air.

P.25
Heating Upon Charcoal. Use the technique described in Expt. 17 if
sufficient material is available. Small amounts of substance are better
heated in hydrogen, P.29.
First heat the substance by itself. Record colors appearing in the flame.
Deflagration indicates the presence of an explosive material or of an oxidant
such as chlorate, perchlorate, iodate, nitrite, nitrate, etc. Compounds of
noble metals are converted to the metal. Oxides and compounds that are
converted to oxides on heating behave as stated below for heating with
sodium carbonate. Substances that do not react with either the carbon
or the hydrogen behave as when heated in an inert atmosphere with some
influx of oxygen from the periphery.
Germanium compounds, in the absence of an alkaline flux, are reduced
to the gray metal, and a deposit of white Ge0 2 may be obtained around
the heated zone. Elemental selenium melts and vaporizes; the brown
fumes condense around the heated area to give a deposit of gray metallic
selenium which may be surrounded by a ring of red selenium. The odor
of rotten radishes is characteristic for selenium. For interpretation of
melting and sublimation see also Tables 4 and 5.
Phosphates, borates, and silicates may fuse to a glassy bead, the color
of which may be indicative of metals present, P. 37.
If the substance is not reduced, but the presence of heavy metals is
suspected, mix some of the substance with twice the amount of anhydrous
sodium carbonate and again heat upon the charcoal or graphite. The
results are determined by the circumstance that oxygen has access to the
periphery of the reaction zone, whereas the flame and the support are
reducing. Thus the vapor of a volatile metal may be oxidized outside
the flame, and a deposit of oxide (incrustation) may be obtained in a ring
zone surrounding the flame.
The following list may aid in the interpretation of the observed
phenomena.
A white, strongly incandescent residue forms, which refuses to melt: oxides
of Ba, Sr, Ca, Mg, AI, and rare earths.
Reduction to metal occurs, but no deposit forms around the heated material:
a) a malleable bead is formed, which may be flattened in the mortar:
Cu, Ag, Au, Sn;
b) gray, metallic particles which may be
malleable: Pt, Ir, Rh, Pd;
not malleable, not magnetic: Mo, W, Re;
not malleable, ferromagnetic: Fe, Co, Ni.
18*
276 Classification Tests P.26

Reduction to metal occurs, and a deposit forms around the heated material:
a) malleable button and yellow deposit: Pb, In;
b) brittle metallic button that may be ground to a powder:
white deposit, Sb;
yellow deposit, Bi;
c) gray metallic powder and white deposit consisting of colorless
crystalline scales: Mo. The formation of a condensate of MoOs may not
occur since it requires use of an oxidizing flame. The scales of MoO, are
yellowish when hot; when they are touched with a reducing flame, a blue
oxide may form.
Reduction to metal does not occur, but a deposit forms around the heated
material: This deposit is
a) white: As, the deposit is volatile and an odor of garlic is perceived;
Zn, deposit is not volatile, but it turns yellow when heated;
TI, slight deposit at some distance from the hot reaction
zone; the flame becomes intensely green;
Te, the outer seam of the deposit has a brownish hue; the
deposit is readily volatilized by touching it with the
flame;
Mo, colorless crystalline scales, see above.
b) brown: Cd, brown rings may surround a blue central area to
imitate the appearance of the "eyes" in the plumage
of the peacock.
c) violet: Ru0 2•
It is understood that the presence of several heavy metals may lead
to complications in the interpretation of the observed phenomena, and
this possibility should be kept in mind. Colorations of the flame should
be noted since they may provide additional clues; for their interpretation
see P. 38. Metallic buttons are seen without difficulty, but for the detection
of powdery metals it is necessary to transfer the residue of the ignition
and some of the carbon supporting it to a mortar where it is ground, leached
with water, and freed from the carbon by floating off the latter with a
stream of water.
The Hepar test, P.59, should be tried with a portion of the residue;
blackening of the silver indicates the presence of compounds of S, Se,
or (and) Te. More information and additional tests may be found in the
special literature (37,47,55, 56, 57).

P.26
Heating in a Current of Gas (935). The material may be placed upon
a narrow slide which is then heated inside a combustion tube after the
air has been displaced by the desired gas. After cooling in the chosen
P.26 Ignition Above 300 C0
277

atmosphere, the slide is withdrawn from the tube for inspection of the
residue under the microscope. Vapors and gases produced upon heating
the substance will frequently escape detection, however, and the following
technique is recommended for the investigation of small samples.
The substance is heated in a capillary of hard glass or vitreous silica,
which has a reasonably heavy wall, a bore of 0.5 to 2 mm, and a length
of 10 to 12 cm. One end is drawn out to a bore of possibly less than 0.1 mm
and bent at a right angle, Fig. 74. The other end is mounted, by means
of a rubber stopper or heavy-walled rubber tubing with capillary bore,
in the opening of the tube which supplies the desired gas. A wad of cotton,

Fig. 74. Heating in a Stream of Gas

placed into space a, will retain dust and spray. The tube is held in a clamp
and may be provided with a manifold stopcock arrangement admitting
nitrogen, oxygen (air), or hydrogen. Gases like chlorine, hydrogen chloride,
or hydrogen sulfide should be supplied through special outlets; they must
not pass through the manifold so that there is no possibility for the contamina-
tion of either nitrogen, oxygen, or air. The purity of these latter gases
must be assured if the gas escaping from the fine capillary shall be tested
for sulfur and halogens. The presence of sulfur or halogen might also
modify the behavior of the solids observed in the capillary.
The substance may be heated without and (or) with the addition of
two parts of anhydrous sodium carbonate, and it may be heated first in
nitrogen and then in oxygen (or air), hydrogen, chlorine (or hydrogen
chloride), and finally in hydrogen sulfide. The choice of the experiments
should depend upon the information available at the time. Before, during,
and after heating, the material should be observed with suitable magnifica-
tion. A magnifier may be used or a simple microscope with stage and
278 Classification Tests P.26

substage removed and the tube in horizontal position. A simple mechanical

manipulator should be available for positioning the heating device which
may be a flame, an electric coil or wire, or a mirror or lens concentrating
the radiat:on of an arc upon the substance. The following directions are
based upon the simple procedure used by HERBERT E. SCHNEIDER.
Holding it at the wide end, clean the outside of the capillary by wiping
with a slightly moist and then with a dry cloth to remove all fingerprints.
Then introduce the material to be tested into the capillary which, to this
end, is best placed upon a sheet of glazed paper of contrasting color.
Transfer to the capillary a grain or small crystal upon a micro spatula or
upon a small piece of paper or metal foil and push it into the opening
and to location b, Fig. 74, by means of a thin glass rod. If the material
is a fine powder, proceed as follows. With a capillary pipet place a droplet
of water of 0.5- to I-mm diameter, too small to fill the cross section of
the capillary, upon a small piece of plastic sheet which repels water strongly
so that the drop becomes hemispherical. Using the flattened end of a
platinum wire, add some of the powder and stir to obtain a thick paste.
Cover the drop with a small watchglass, cup, or crucible and place it and
the capillary upon a metal surface having a temperature considerably
below the ice point (and best located within a dry box). When the drop
has become solid, press on it from the side with a metal spatula to get
it separated from the surface of the plastic. Then, without giving it time
for melting, transfer the congealed mixture like a grain to location b in
the capillary. Pass dry nitrogen through the capillary until the water
has been evaporated completely and the whole interior of the capillary
is again dry.
It may be possible to obtain a coherent grain by simply allowing the
water to evaporate from the paste, and this procedure may be tried first.
Another alternative is the use of the technique of Expt. 47 for collecting
the solid of a slurry in the cut-off end of a capillary. The end of the capillary
with the collected solid in it is placed into an oven for drying, and it is
then transferred to location b of Fig. 74; the disadvantage of this procedure
is that the observation of the heated solid is rendered difficult.
To obtain a particle representing a mixture of the substance with
sodium carbonate, grind the material to be tested with an equal volume
of anhydrous sodium carbonate to a very fine powder. Obtain a tiny
droplet of water on the strongly repelling plastic surface, and collect
next to it a pile of the powder, about twice the size of the droplet and
still too small to fill the cross section of the capillary. With a plastic needle
or a metal wire combine the solid and the droplet to form a tiny bead.
Allow 5 minutes for the crystallization of the hydrate which will cement
together the material. Then separate the bead from the plastic by pushing
from the side with a spatula and transfer it to location b in the capillary,
P.26 Ignition Above 300 0 C 279

Fig. 74, as suggested above. Place the capillary into a metal block, and
heat it slowly to remove the water, while a stream of dry gas is being
passed through. Raise the temperature to about 150 0 C.
If the available information does not suggest a different approach,
it is suggested to first heat the material in nitrogen. If it does not sublime
without leaving a residue and if it does not form a glassy bead when heated,
it is advisable to try heating in oxygen and then hydrogen. A destillate
or sublimate obtained in nitrogen should be tested by heating in oxygen.
If the sublimate appears to be an inorganic halogen compound or if the
substance fuses to a glassy bead, hydrogen reduction of its mixture with
sodium carbonate is recommended.
The required gases are best supplied from steel cylinders by means
of reducing valves, washed if necessary, and dried just before they enter
the manifold. Nitrogen and oxygen may be dried with Anhydrone which,
however, must not be used for drying flammable gases. Hydrogen as well
as chlorine and hydrogen chloride are dried by passing them through
concentrated sulfuric acid. By all means make certain that each gas has
displaced the air up to the stopcock leading into the manifold. Before
attaching the capillary to a, Fig. 74, test the gas escaping through a for
purity. When tested with a glowing splint, nitrogen should stop the
glowing, air will support it, and oxygen will cause the splint to burst into
a flame. Test hydrogen by collecting a sample of the gas in a test tube
(this may be done by displacing the air downward); when lighted, the
contents of the test tube should burn with a quiet flame.
Assemble the apparatus in a location where it will not be exposed to draft.
If tubing a, Fig. 74, is held in a clamp, the capillary will need no special
support. Adjust its position and focus the horizontal microscope upon
the substance which is illuminated by a strong beam of light coming from
the side of the observer. First turn on the stream of gas, and then insert
the orifice of the fine capillary into 5 to 10 pI O.I-F Ba(OH)g solution
contained in the tip of a microcone. Regulate to obtain a slow stream of
gas so that individual bubbles of gas may be seen rising through the liquid
which must remain clear.
For the collection of condensates, cool the capillary at a point about
25 mm downstream from the substance to be heated. A cooling block d,
Fig. 74, made of aluminum and filled with a mixture of dry ice and acetone,
is recommended, or a strip of wet filter paper may be hung over the capillary.
Start heating with a microflame of about I-mm diameter. Have the
flame burning at the orifice of a capillary drawn from one end of a tube
of 15- to 25-cm length, which is-near the other end-clamped to a simple
manipulator situated to the left of the microscope. By using the mechanical
motions slowly and stepwise, approach the substance with the flame while
watching through the microscope. Record the behavior of the substance
280 Classification Tests P.26

and, each time before moving the flame closer to the material, check the
appearance of the barium hydroxide solution and see whether or not a
condensate has formed in the capillary.
Sublimates usually appear close to the sample. If a liquid condensate
is obtained, lower the flame somewhat, stop the gas stream, and cut the
capillary at f. Collect the liquid in a capillary pipet, and then again draw
out a fine capillary at f and bend it as in Fig. 74. Use a small fraction of
the collected condensate for determining its pH. If it is close to pH 7
and the barium hydroxide solution is still clear, continue to use the latter.
Otherwise, get another microcone with a fresh portion of barium hydroxide
solution for the continuation of the heating experiment, and investigate
the contents of the first microcone.
Continue heating in the slow gas stream until either the whole material
has been volatilized or the softening point of the capillary is reached.
Remove condensing liquids and exchange the barium hydroxide solution
for a fresh batch whenever this seems indicated. Finally allow the capillary
to cool without stopping the gas flow in order to maintain the selected
atmosphere until the sample has room temperature.
Test liquid condensates as suggested in P.4. If the pH is close to 7,
no heavy fumes were seen in the capillary, and the barium hydroxide in
the microcone appears unchanged, it is probably water-and this may be
confirmed without expenditure of sample by determination of the boiling
and melting points, Expt.21, and p. 215. After performance of the
boiling point determination, the melting point is simply obtained according
to EMICH (152) by sealing the capillary at both ends and attaching it
with a rubber band to a thermometer which is then mounted by means
of a cork in a test tube half-filled with ethanol. Test tube and contents
are chilled to about - 10 C by immersion in an ice-salt mixture (if the
0

water droplet does not solidify, the thermometer with the capillary is
briefly withdrawn from the alcohol and chilled more strongly; EMICH
squirted ethyl chloride upon them). When the water has frozen in the
capillary, the test tube is taken from the cooling bath, wiped dry on the
outside, and, by means of a cork, mounted in a wide test tube which is
held in hand and continuously turned end over end to keep the contents
of the inner tube thoroughly mixed. The thermometer is read when the
solid in the capillary liquifies.
Test the contents of the micro cone as follows. Centrifuge if a precipitate
has separated. Transfer the solution to another microcone. Wash the
precipitate with 5,ttl O.I-F Ba(N0 3 )2, and discard the wash liquid.
Treat the precipitate with 10,ttl cold 3-F HN0 3 :
a) the precipitate dissolves completely and some gas may be liberated:
CO 2, S02; with a loop add 0.3,tt1 0.02-F KMn04 and mix: S02 is probably
P.26 Ignition Above 300 0 0 281

absent if the solution becomes pink; if the permanganate is reduced, add

more of it until the solution remains red upon mixing; centrifuge: a white
or pink precipitate of BaS0 4 in a still acid solution confirms S02'
b) The precipitate does not dissolve or does not completely dissolve:
centrifuge, transfer the solution to another cone and test it according to (c);
the insoluble precipitate may be BaS0 4 , BaF2' both finely crystalline
powders, or BaSiFs ' 2 H 20, monoclinic spindle-shaped crystals, singly, in
crosses or radiating masses, or (and) gelatinous silicic acid from the reaction
of SiF4 with water. The SiF 4 as well as HF may be obtained from fluorides
and fluosilicates especially when heated in presence of acid salts like KHS0 4 •
The residue may be transferred as a slurry to a platinum crucible and
tested for fluoride, P. 60. It may be digested with Na 200 a solution, and
the latter tested for sulfate with HOI and CaCI 2, P. 59.
c) Test the solution for S02 as under (a): if no reduction occurs and
some of the precipitate did dissolve, CO 2 should have been present. It is
suggested to use a fresh sample of the material under investigation for a
confirmatory test.

Test the solution removed from the original precipitate as follows.

a) Use a small fraction of the solution to test for ammonia, P.58.
b) Transfer another small fraction of the solution to a slide. Stir into
it some potato starch. Inspecting repeatedly under the microscope, first
add little O.l-F KI solution that does not color starch, and then expose
to fumes of concentrated HCI until the mixture has become acid. If the
starch grains assume a blue to black color, the barium hydroxide must
have absorbed some oxidant such as C1 2, Br2 , 12 or N0 2.
c) Treat the major portion of the solution with 2,ul 3-F HNOa and
then add up to 20,ul O.l-F AgNO s in small portions with inspecting, mixing,
and centrifuging after each addition. Layers of differently colored
precipitates may be obtained. Black Ag 2 S would separate first and may
be followed by yellow AgI, pale yellow AgBr, and white AgCl or AgCN.
The last two may be readily identified by recrystallization from ammonia
(P. 60) and nitric acid (P. 36), respectively. Black Ag 2S may be treated
with CaC1 2 and oxidized by bromine, P.36.
At the conclusion of the experiment, the capillary may be cut at various
points to separate the residue from sublimates, and these may be kept
for further investigation or immediately subjected to analysis.
Only a rather crude outline may be given for the interpretation of
the observations, and much must be left to the experience and imagination
of the observer and the information that may be obtained by a study of
the literature.
282 Classification Tests P.27

P.27
Heating in an Inert Atmosphere, Nitrogen. Heating in a stream of
nitrogen is approximately equivalent to heating in a "closed tube". Aside
from the fact that the alkaline earth metals combine with nitrogen to form
nitrides, there is no extraneous agent to react with the sample, to oxidize it,
or to reduce it. The observed phenomena may be interpreted with the use
of Tables 4 and 5. The transition temperatures may be crudely estimated
by touching a thermocouple to the heated capillary or by holding close
to it a thin rod of vitreous silica to which are attached fragments of tempera-
ture indicating crayon or of Tempil Pellets (13).
Transition and melting points at temperatures from 350° to 900° C
may be quite accurately determined under the microscope by using an
electric hot stage which, according to BRADLEY (466), may be made by
coating one side of a vitreous-silica microscope slide with a film of platinum
and fastening it to an aluminum adapter, 7.5 cm X 2.5 cm, which fits any
conventional mechanical stage. The temperature is simply determined
by empirically calibrating the settings of the variable transformer while
observing the melting points of known substances.
Still higher temperatures may be obtained by placing the particle
upon a strip of platinum foil which is heated by an electric current (730,
11l0, llll). The particle is observed with a horizontal reading microscope
or telescope, and the temperature is measured with an optical pyrometer.
Use of a s~itable dry box (458, 465) should permit working in nitrogen,
hydrogen, or oxygen. Concerning D. M. Olson's reflecting microscope for
temperatures up to 2500° C and low to 600 magnification see R. H.
MULLER (484).
Changes will be observed on the substance if it decomposes upon
heating; gases given off as well as sublimates and distillates may permit
conclusions concerning the nature of the original substance. Hydrides of
the alkalies and the alkaline earths will decompose with the formation
of nitrides. Water of hydration is lost upon heating; water is also given
off by some acids, by acid salts, as well as by hydroxides and basic salts.
Salts of oxygen acids (chlorates, perchlorates, nitrates etc.) and, obviously,
peroxides and salts of "per" acids may give off oxygen. HCI, HBr, etc.
may occur as a consequence of hydrolysis when water is given off at elevated
temperature. C1 2 , Br 2 , 12 may be derived from the decomposition of the
corresponding salts of the noble metals. SiF 4 may come from the de-
composition of fluosilicate or the action of acid fluoride upon Si0 2, silicate,
or the glass of the tube. Sulfur (Se) is given off by some simple ,sulfides
(FeS 2) when they are heated. But for those of the alkalies and some of
those of the alkaline earths, most sulfates, sulfites, nitrates, nitrites, and
carbonates decompose with the liberation of S02, S03, NO, N0 2, and
P.28 Ignition Above 300° 0 283

00 2 , Phosphates, borates, and some silicates may fuse to give a glassy

bead.
Most organic compounds will undergo pyrolysis with the formation of
a wide variety of substances; characteristic are evolution of flammable
gas and the formation of tarry products and a carbonaceous residue;
liberation of NH 3, HOI, S02' S03' N0 2, etc., are indicative of the particular
elemental composition. The heating of acetates may give a condensate
of acetone.
Salts of ammonium, hydrazine, hydroxylamine, and of ammonium
or aminobases may decompose with the liberation of NH3, HOI, S02'
S03' N0 2, etc.
Oomplete decomposition which gives only gaseous products outside of
water is observed with NH 4N03, oxalic acid, and various organic explosives
(glyceryl nitrate, cellulose nitrate, picric acid, etc.).
Some substances show characteristic reversible color changes: ZnO,
white to yellow; Sn0 2, yellow to brown; HgS, red to black; HgI2' red to
yellow; Ag 2HgI 4 , yellow to red at 50°; Ou 2HgI 4 , red to brown at 70°.
The multiplicity of phenomena which are possible when mixtures are
being heated in an inert atmosphere defies listing, and mentioning black
gunpowder and Thermit may suffice to indicate extremes of unpleasantness.
In general, interpretation will be simplified by separating the components
and testing them separately.
Absence of chemical change upon heating to high temperature in
nitrogen excludes organic and organized materials as well as compounds
of noble metals and is characteristic for: (a) most salts of the alkalies and
the sulfates of Ba, Sr, and Ca; (b) simple oxides, sulfides, selenides,
tellurides, nitrides, and carbides; (c) phosphates, borates, silicates; and
(d) anhydrous simple halides, many of which have a relatively high vapor
tension and may sublime at temperatures around 500° C.

P. 28
Heating in a Current of Oxygen. Heating in oxygen will produce useful
evidence when the presence of free elements, of organic compounds, or
of sulfides, selenides, carbides, or hydrides is suspected. In general, the treat-
ment will give oxides which then may be reduced by heating in hydrogen.
For information concerning the color of substances see P. 9; Tables 4
and 5 provide the data on melting points and volatility.
No change in appearance on heating in oxygen is characteristic for:
(a) noble metals, stainless steel, and aluminum; (b) oxides; (c) most salts
of the alkalies; (d) anhydrous halides; (e) phosphates, borates, and silicates.
Metals are oxidized to an oxide; some obtain tempering colors due to
the formation of an oxide film; the oxides may be classified according
to color, volatility, and melting point.
284 Classification Tests P.29

S, Se, Te, P, As, C, Si, and B give the oxides of which S02 and CO 2
are recognized by passing the combustion gases into barium hydroxide
solution. Se0 2, Te0 2, P 205' As 2 0 3 sublime; Te0 2 melts to a yellow liquid.
Diamond burns rather slowly.
Formation of water and oxide results when hydrides are heated in
oxygen; carbides burn to oxide and CO 2.
Carbon dioxide and H 20 are obtained upon heating organic and organized
materials in oxygen; in addition, other gases may be obtained (HCI, HBr,
S02' S03' NH 3 , NO, N0 2, etc.) which may be identified to reveal the
specific elemental composition. If the substance burns, record the color
of the flame and presence or absence of smoke. An ash may be left behind,.
which represents the mineral constituents.
Sulfides give S02 and oxides; selenides and tellurides give white
sublimates of Se0 2 and Te0 2; sulfides of phosphorus, S02 and a sublimate
of P 20 5 •

P.29
Heating in a Current of Hydrogen. Heating in hydrogen is not re-
commended when the unknown substance sublimes or when it fuses to
a glass. Sulfides should first be roasted to the oxides; halides, phosphates,
borates, and silicates should be treated with Na 2C0 3 to obtain the correspond-
ing carbonates or oxides. Sulfites, sulfates, nitrites, nitrates, carbonates, and
salts of organic acids frequently give carbonates or oxides on heating in
nitrogen or oxygen, and are then reduced by the hydrogen to the metal.
Heating in hydrogen is often useful for the investigation of the so-called
insoluble residue: silver halides give metallic silver and hydrogen halide;
sulfates of the alkaline earths are reduced to soluble sulfides; PbS0 4 and
TI 2S0 4 , to the metals; Sn0 2 and Sb 2 0 5 are reduced to the metals, and so
are Fe 20 3 , Cr 20 3 , and insoluble salts of chromium.
The condensate of neutral water on heating in hydrogen may also be
used to prove the presence of an oxide.
The heating in hydrogen may follow immediately after heating in
nitrogen if the presence of a reducible oxide is probable. It may follow
upon the burning of sulfides, selenides, tellurides, phosphides, or carbides
in oxygen. If halides, phosphates, borates, or silicates are present, it is
best to first fuse a sample of the unknown with Na 2C0 3 , to extract the
melt with water, and to use the residue of carbonates and (or) oxides for
testing in the current of hydrogen.
Heating in hydrogen produces the following effects.
The alkali and alkaline earth metals are converted to white hydrides
at temperatures ranging from 150 0 to 700 0 C; the hydrides decompose
when heated to higher temperatures, and they all react with water giving
hydrogen and hydroxide.
P.30 Solubility 285

Not reduced or apparently not reduced are the oxides of the alkalies;
Mg, Ca, Sr, Ba; Sc, Y; Ti, Zr, Th; Nb, Ta; U; Mn; AI, Ga; Si, and Ge.
Not reduced are borates and silicates in general, and the halides, oxides,
sulfides, and carbonates of the alkalies and the alkaline earths.
Reduced to sulfides and phosphides, respectively, are the sulfates and
the phosphates of the alkalies and the alkaline earth; see confirmatory tests
in P. 58 and 59.
Reduced to hydroxides and oxides are the nitrates and nitrites of the
alkalies and the alkaline earths; reduced are also the peroxides of these
metals.
Reduced to the corresponding halides are the alkali and alkaline earth
salts of the oxygen acids of the halogens.
Reduced to the elemental state are the oxides and the salts of volatile
oxygen acids of: V; Cr, Mo, W; Fe, Co, Ni; platinum metals; Cu, Ag, Au;
Zn, Cd, Hg; In, Tl; Sn, Pb; As, Sb, Bi; Se, and Te. Of these, Zn, Cd,
Hg, As, Se, Te distil to a colder part of the tube to give silve>y droplets
of Hg, silvery white metallic deposits of Zn and Cd, a brown to black
mirror of As, a steel gray and red sublimate of Se, and a gray to black
deposit of Te. In, TI, Sn, Pb, Sb, and Bi distil only partially when they
represent a large portion of the reduced metal. Fe, Co, and Ni are attracted
by a magnet applied to the outside of the tube.
Water is formed by the reduction of oxides and the salts of oxygen acids.
HCl, etc. are derived from reducible halides or halides that decompose
on mere heating.

P.30
Heating in a Current of Chlorine and of Hydrogen Sulfide. Heating
in a current of chlorine or HCl is suggested when heating in lJ.ydrogen
leads to a metal. Of course, the metal may be dissolved in acid for further
identification; conversion to chloride by heating in a stream of gaseous
reagent may be an attractive alternative. The resulting chloride may
be converted to the sulfide by heating in a current of H 2 S. Concerning
the appearance and behavior of the resulting compounds, the experimenter
will refer to P.9 to 13 and Tables I, 4, and 5.

Solubility
The information gained by determining the solubility may be dis-
appointing, especially if the material is organic in nature. This is partly
due to the fact that there is no clear boundary between soluble and insoluble
so that arbitrary limits must be used. On the other hand, the testing for
solubility does not cause loss of material, and information concerning a
suitable solvent is n~cessary for further investigation. Provided that the
286 Classification Tests P.30

amount of sample, solvent, and undissolved residue are determined so

that the concentration of the resulting solution becomes known, the solution
is well suited for chemical analysis and the estimation of the amount of
substances discovered in it. In addition, evaporation of solutions may
give well developed crystals for investigation under the microscope as
outlined in P. 10 to P. 13.
The customary solvents are water, acids, bases, and diethyl ether,
but any other solvent may be tried, which appears promibing from the
facts collected in preceding tests.
For the performance of solubility tests, the sample should be ground
to a fine powder, ii necessary, and weighed. The weight of small particles
may be estimated from the measurement of their dimensions under the
microscope; a reasonable guess concerning their density will usually be
possible. The volume of the solvent is chosen depending upon the nature
of the material. When equilibrium has been established, the decision is
made concerning the solubility. A solvent which is not effective is removed
and replaced by the next solvent to be tried-possibly after drying the
solid. EvapOlation of the removed ineffective solvent may furnish crystals
of an impurity or of a slightly soluble major or minor constit~ent. If the
material under investigation is a mixture, each clearly discriminating solvent
should be repeatedly applied to a obtain a satisfactorily complete separation;
the resulting solutions and residues are then used for further investigation.
Thd technique is essentially that of batch extraction. No further
comment is needed if the work is carried out in a centrifuge cona, micro cone,
or capillary cone. In the last instance, the sample will be in the field of
the microscope when the solvent is added, and evolution of a gas during
dissolution will not escape detection. When working on a larger scale,
there should be no difficulty in testing a liberated gas for CO 2, S02' H 2S,
C12, and NH 3 , see P. 36, 37.
When performing a solubility test on the microscope slide, it is rec-
ommended to place the particle(s) of the sample next to the drop of solvent.
After focusing upon the edge of the drop, one particle is picked up with
a glass needle, Expt.61, and placed into the solvent inside the field of
microscopic vision. The rounding off of the edges of the particle, gas
bubbles, appearance of a coloration or of a precipitate itl the solution
around the particle indicate action of the solvent upon the particle. If
soluble, the particle may finally vanish, and crystals may appear along
the edge of the drop where the solution gets concentrated as a consequence
of evaporation. The use of slides with a concave depression or (and)
treatment with Desicote (13) is (are) recommended for work with organic
solvents that spread on the surface of the slide.
For the determination of the solubility of small particles, MONKMAN (457)
and JAECKER and SCHNEIDER (959) expose it to the vapors of the solvent
P.31 Solubility 287

while observing under the microscope or with a 5 X to 10 X magnifying lens.

The latter transfer the particle onto a narrow slide which is inserted into
an absorption tube for semimicro combustion (as an alternative, a slide
or cover slip with the particle attached to the underside could be made
the removable cover for the circular or elliptical side opening of a glass
tubing, 8- to 10-mm bore, which is mounted on the stage). By use of
3-way stopcocks, nitrogen or air is passed either directly to the tube
containing the particle or via a gas washing bottle containing the liquid
solvent. A manifold stopcock arrangement may be used with several gas
washing bottles containing different solvents so that the sample may be
subjected to the action of a series of solvents. Of course, every solvent
should be completely removed by passing clean nitrogen or air through
the chamber before starting a test with another solvent.
The rate of gas flow is important. If too fast, the gas does not become
saturated with the solvent while passing through one simple gas washing
bottle. If the rate is too slow, the solvent vapor is brought to the sample
too slowly. A rate of 400 to 500 ml gas per minute was found satisfactory
in experiments at room temperature with ethanol, diethyl ether, carbon
tetrachloride, and benzene. Obviously, the gas washing bottles may
be supplied with water, strong hydrochloric acid, or ammonia solution;
in addition, the gas washing bottles may be placed into a shallow bath
of warm water in order to raise the vapor tension of the solvent. The
degree of solubility may be correlated to the time required for the conversion
of the solid particle to a liquid droplet. Anomalous behavior has been
discovered in several instances (965), where soluble substances refused
to liquify, which may be explained by the formation of compounds with
the solvent. Of course, withdrawal of the solvent by passing pure nitrogen
or air through the cell may furnish crystals for identification according
to P. 10 to P. 13.
Performance of Solubility Tests
Metallic materials are tested with acids only. Otherwise the solubility
in water is determined first, regardless whether the material is inorganic or
organic in nature.

P.31
Inorganic Substances are considered soluble if a 1 % or stronger solution
may be obtained (1 mgjO.1 ml or higher concentration); moderately soluble,
if 0.1 to 1 mg dissolve in 0.1 ml; and insoluble, if less dissolves, i. e., a
residue is left when treating 1 mg material with 1 ml solvent. Most inorganic
substances are either "soluble" or "insoluble". The relatively few substances
which are moderately soluble in water of room temperature are listed in
Table 8.
288 Classification Tests P.31

The material should be used in the form of small particles (powder)

so that equilibrium is rapidly attained. Add 0.1 ml water per mg sample
if only one substance seems to be present or 0.03 ml per mg sample if the
latter appears to be a mixture. Agitate the mixture for several minutes.
Dissolution may be hastened by heating on the steam bath, but finally
the mixture must be cooled to room temperature with the proper pre-
cautions to prevent the establishment of a supersaturated solution.
The substance is classified as soluble if a clear solution is obtained
and no solid residue is left behind. If the solid remains unchanged and
evaporation of some of the supernatant liquid does not give a significant
residue!, the material is considered insoluble. If some residue remains,
the microscopical inspection of it and of the residue of the evaporation
of some of the solution may show whether it is partial solubility of one
substance or the complete dissolution of one or several components of a
mixture. If it is limited solubility, addition of 0.9 ml more solvent per
1 mg sample will decide whether the substance is moderately soluble or
"insoluble". If a mixture of several substances is given, extraction with
water should be continued to get complete separation of the soluble material
from the insoluble. The amount of the insoluble part should be estimated
from the volume of the residue.
Evaporation of the aqueous solution may give well developed crystals
for investigation by P. 10 to P. 13. In the instance of mixtures, it is
advisable to analyze the aqueous extract separately.
Materials (residues) insoluble in water are tested with 0.01 mljmg solid
of the following acids. In each instance apply the acid first at room
temperature, and then heat on the steam bath. Try them in the order:
4-F HCl, 12-F HCI, 6-F HNO a, 16-F HNO a, and aqua regia (1 volume
16-F HNO a +3 volumes 12-F Hel) until a satisfactory solvent has been
found.
Tables on the solubility of common inorganic salts may be found in
books on qualitative analysis (40, 55, 56). The following rules and Table 8
will serve the same purpose.
Soluble in Water Are:
most compounds of the alkalies and ammonium;
most chlorides, hypochlorites, chlorates, perchlorates, bromides,
bromates, iodides, periodates, manganates, permanganates, cyanates,
acetates, thiosulfates, sulfates, selenates, nitrites, nitrates, hypophosphites,
and vanadates;

1 The solvent itself may give a slight residue either because of impurities
dissolved or became of attack of the apparatus; use of vitreous silica and comparison
with a solvent blank will aid in arriving at a decision.
P.31 Solubility 289

the alkaline earths cyanides, thiocyanates, ferrocyanides, ferricyanides,

hydrosuHides, polysulfides, bisuliites, primary phosphates and arsenates,
and bicarbonates;
AgF, SnF2, SnF4 , SbFs, BiFs;
TIOH, TICN, Hg(CN)2;
many complex compounds of heavy metals with ammonia and amines,
water, hydroxyl, cyanide, thiocyanate, thiosulfate, oxalate, tartrate,
citrate, and various enolic compounds.
Insoluble in Water, but Dissolved by 6-F HNO a or HOl:
most metallic oxides, hydroxides, and basic salts;
most water insoluble salts of weak acids such as fluorides, iodates
(those of Ag, Mg, Ca, Sr, Ba, Pb do not dissolve readily in HNO s), cyanides,
thiocyanates, ferro cyanides, ferricyanides, sulfides, sulfites, selenites
(tellurites are decomposed with separation of Te0 2), chromates, tungstates,
phosphates, arsenates, arsenites, borates, carbonates, oxalates, tartrates,
and other salts of organic acids;
the readily hydrolysable salts of Bi, Sb, Sn(4), Fe(3), AI, etc.;
the hydrides, nitrides, phosphides, and carbides of the alkaline earths;
BeC, Al 4Ca, LaC2, NdC 2, FeaC, NisC, MnaC, UC 2, CU 2C2, Ag 2C2;
AgCNO, Pb(CNO)2;
vanadates of Hg(l), Pb, Fe, Cr, AI;
some silicates are decomposed by acid.
Insoluble in 6-F Acid, but (Oxidized and) Dissolved by Hot 16-F HNO s or
Aqua Regia:
12 , S (slowly), P, As, B;
Cu 2Cl2, Cu 2I 2, Cu 2(CNS)2' AgCN, AgCNS, AuCI, PtCI 2, TICI, TlBr,
TlI, TI 2S0 4 , TICNS, Hg 2C1 2;
HgI2' Bil a, HgS, As 2Ss, As 2Ss, MoSs;
slowly attacked are CaF 2' Sn0 2, Sb2 0 s, some tungsten bronzes, some
silicates, Prussian blue.
Insoluble in All Solvents Tried:
S, C, Si; certain metals and alloys, P. 42;
AgCl, AgBr, AgI, AgCN, CaF 2 ;
Si0 2, Ti0 2 , Zr0 2 , Th0 2 , Sn0 2, Nb 20 s, Ta 20 S ' WOa, H 2W0 4 ;
Fe 20 a, Cr20 a, Al 20 a after exposure to high temperature;
BaS0 4 , SrS0 4 , (CaS0 4), PbS0 4, fused PbCr0 4;
pink anhydrous salts of chromium: CrCla, CrFa, Cr 2 (S04)3;
carbides: B 4 C, SiC, TiC, ZrC, HfO, NbC, TaC, (MoC, WC, VC);
nitrides: borazon=cubic BN, TiN, ZrN, TaN;
borides: TiB, ZrB;
silicates, chromites, aluminates formed at high temperature;
Benedettl·Pichler, Identification 19
290 Classification Tests P.32

tungsten bronzes characterized by metallic luster, intense color, high

density, and electric conductivity.

P.32
Organic Substances are classified as "soluble" if 30 mg or more dissolve
in 1 ml solvent of room temperature; as "insoluble" if less dissolves.
Agitation for several minutes will give equilibrium if the particles of the
sample are small, i. e., the sample has been ground to a powder.
If only one substance is present, use 0.033 ml solvent per mg sample;
if a mixture may be present, preferably try first 0.01 ml solvent per mg
sample. In the instance of a single compound, agitate until equilibrium
is obtained and then decide whether or not the substance has been dissolved
without leaving a residue; if there is a small residue, microscopical inspection
may help to decide whether or not it is an impurity. When dealing with
a mixture, inspect under magnification the residue of the solubility tests
as well as the residue from the evaporation of a small portion of the solution;
the appearance of the solids may indicate whether a separation has been
obtained and whether it has been complete. The amount of residue should
be estimated, and if a separation is obtained, it should be made complete
with measured amounts of solvent to permit estimation of the degree of
solubility. Any well developed crystals may be investigated according
to P. 10 to P. 13.
OHERONIS and ENTRIKIN (146) determine the solubility in water,
diethyl ether, 1.2-F HOI, 2.5-F NaOH, NaHOO a, and in IS-F H 2 S0 4 to
recognize the following divisions:
(S1) Soluble in Water and Ethe1':
generally monofunctional compounds with five carbons or less: alcohols,
aldehydes, ketones, carboxylic acids, acetals, anhydrides, esters, ethers,
lactones, some glycols, polyhydroxy phenols; amines, amides, amino
heterocyclics, nitriles, nitro paraffins, oximes; halogen substituted
compounds of the above list; hydroxy heterocyclic sulfur compounds,
mercapto acids, thio acids; halogenated amines, amides, and nitriles;
amino heterocyclic sulfur compounds.
(S2) Soluble in Water, Insoluble in Ether:
compounds with moderate molecular weight, having two or more polar
groups or being a sulfonic or sulfinic acid: dibasic and polybasic acids,
hydroxy acids, polyhydroxy alcohols and phenols, simple carbohydrates;
salts of acids and phenols, various metallic compounds; ammonium and
amine salts of organic acids, amines, amino acids, amides, amino alcohols,
semicarbazides, semicarbazones, ureas; halogenated acids, alcohols, and
aldehydes and acyl halides; sulfonic acids, alkylsulfuric acids, sulfinic acids;
P.33 Review: Inorganic Substances 291

amine salts of halogenated acids; aminodisulfinic acids, bisulfates of weak

bases, cyano and nitro sulfonic acids.
(B) Insoluble in Water, Soluble in 1.2-F HOl:
amines, amino acids, amphoteric compounds such as aminophenols,
aminothiophenols, aminosulfonamides, arylsubstituted hydrazines, N -di-
alkylamides.
(AI) Insoluble in Water and in 1.2-F HOl, but Soluble in 2.5-F NaOH
and 1.5-F (?) NaH00 3 :
acids (usually with 10 carbons or less) and anhydrides; amino, nitro,
and cyano acids, heterocyclic nitrogen carboxylic acids, polynitrophenols;
halogenated acids, polyhalogenated phenols; sulfonic and sulfinic acids;
aminosulfonic acids, nitrothiophenols, sulfates of weak bases; sulfonhalides.
(A 2 ) Insoluble in Water, 1.2-F HOl, and 1.5-F NaHOO a, but Soluble in
2.5-F NaOH:
high molecular weight acids, anhydrides, phenols including esters of
phenolic acids, enols; amino acids, nitrophenols, amides including N-mono-
alkyl amides, aminophenols, amphoteric compounds, cyanophenols, imides,
N-monoalkyl aromatic amines, N-substituted hydroxylamines, oximes,
p- and s-nitroparaffins, trinitro aromatic hydrocarbons, ureides; halogenated
phenols; mercaptans, thiophenols; polynitro halogenated aromatic hydro-
carbons; aminosulfonamides, aminosulfonic acids, aminothiophenols, sulfon-
amides, thioamides.
Insol~tble in Water, in HOl, and in NaOH:
(M) Oontaining Nitrogen or Sulfur: anilides and toluidides, amides,
nitro arylamines, nitro hydrocarbons, aminophenols, azo, hydrazo, and
azoxy compounds, di- and triarylamines, dinitrophenylhydrazines, nitrates,
nitriles; mercaptans, N -dialkylsulfonamides, sulfates, sulfonates, sulfides,
disulfides, sulfones, thioesters, thiourea derivatives; sulfonamides;
halogenated amines, amides, and nitriles;
(N) Not containing Nitrogen or Sulfur and Soluble in 18-F H 2S0 4 :
alcohols, aldehydes, ketones, esters, ethers, noncyclic unsaturated hydro-
carbons, unsaturated cyclics that are easily sulfonated (di- or polyalkyl-
substituted benzenes), acetals, anhydrides, lactones, polysaccharides
(charring noticeable);
(I) Not Containing Nitrogen or Sulfur and Insoluble in H 2S0 4 : cyclic
hydrocarbons, paraffins, halogenated hydrocarbons, and diaryl ethers.

P.33 Review: Inorganic Substances

The ignition test will have shown whether the material under investiga-
tion is an inorganic or an organic substance, or a mixture of both. The
solubility test may have shown a way for their separation. For suggestions
on the analysis of organic substances see P.34.
19·
292 Classification Tests P.33

If the material is inorganic, its history and the preceding tests (hardness)
may more or less definitely indicate its nature so that only the confirmation
of chemical identity remains as a final task. Analytical confirmatory tests
are compiled in P. 44 to 60. Means for distinguishing .similar substances
(different oxides of a metal, different hydrates of a salt, primary or secondary,
etc. salt of a polyprotic acid, choice between various complex compounds
of like qualitative composition, distinction between synthetic and natural
product, etc.) will usually have to be found from the description of the
substances involved (4, 5, S, 10) or from an inspection and study of samples
of the substances to be considered. On the recognition of various types
of carbon see CHAMOT (llS), p. ISS, and FEIGL (121), p.30S. On the
identification of corundum among natural and artificial associates see
CROSSMAN (442). Descriptions of the pigments used in paintings have
been compiled by GETTENS (1094, 1095). Books on mineralogy may be
consulted on the description and identification of minerals (169). Advice
concerning the recognition of soil minerals (particles) is given by FRY (93);
see also KmK (IS6).
If the preceding testing has furnished no definite clues for identification,
it will have produced sufficient information for arriving at a decision for
the continuation of the work. At least five different approaches, outlined
below, promise success. Aside from availability of apparatus, the decision
will be influenced by the amount of material available and its nature:
single substance or mixture, metal or non-metallic, simple or complex
substance, crystalline or glassy (amorphous), soluble or insoluble, "rare"
elements improbable or probable, naturally occurring or man-made, etc.
Crystallographical Optical Analysis. By the use of goniometer, elaborate
polarizing microscope, universal stage, and determination of refractive
indices (P. 10 to P. 13) solid particles (crystals and crystal fragments)
may be identified as "molecular species" without being destroyed. The
procedure is especially recommended in the instance of minerals and rock
constituents (S3, 94, 96, 97, 99, 101, 104, 107, 113, 114, 773).
X-Ray (or electron) Diffraction. This procedure likewise identifies
"molecular species" without causing destruction. In mixtures, it will
discover majors and minors. The amount of sample may be as small as 10 fhg.
Usually the material is used as a fine powder, but a!so single small crystals
and thin films may be investigated (SO, 467).
X-Ray Emission Spectrography (4S2, 770). This is a powerful and
efficient non -destructive method for revealing the elemental composition
if the elements below atomic number II (sodium) are of no interest. The
sample may be very small or a thin film; majors, minors, and traces down
to one part in ten thousand are discovered. Elements that are difficult
to discover in the wet way (rare earth, Ta, Nb, W, etc.) present, no
difficulties.
P.33 Review: Inorganic Substances 293

A beam of X-rays or electrons is focused upon the sample to excite

emission of K, L, M, N, 0, P series of lines in correspondence with the
atomic number of the elements present in the sample. The emitted radiation
(X-rays) is passed to an X-ray spectrograph or spectrometer. The far
greater simplicity of the spectra as compared with their optical counterparts,
the identical appearances of the series for all elements but the lightest ones,
and the systematic variation of wave length with atomic number are
distinct advantages for the interpretation.
For qualitative purposes, an instrument like the Microanalyzer Camera
of L. v. HAMOS (79) appears quite satisfactory. It gives a spectrogram
of all characteristic rays excited by an X-ray beam incident upon a very
small sample which may have a volume of 10-8 ml = 10 pI. Various
instruments are commercially available (77).
Solids may be exposed as such or diluted (mixed) with starch, lithium
carbonate, aluminum powder, alumina, or embedded in Lucite. Minerals
may be dissolved in a borax flux. The sample may also be dissolved in
water or suitable solvents (containing only C, H, N, and 0). Spots on
filter paper have been succesfully analyzed, and it seems obvious that
also the circular residues of the ring-oven technique will do.
The microsonde or electron probe (1292) uses an optical microscope
for selecting the field of study, whereupon a beam of electrons is focused
upon a detail in the field, that may be as small as 1 /hm or less in diameter:
less than 0.000001 mm2 in area and probably less than 10-12 ml or 0.001 pI
in volume. The elements present in the bombarded detail are approximately
indicated by the amount of backscattering of electrons and definitely
identified by the frequency of the emitted X-rays. Accurate quantitative
analyses can be performed by comparing the intensity of the X -rays emitted
by the specimen with that from the pure chemical elements and applying
appropriate interelement corrections (495, 1273).
At this time, the most highly developed instruments permit rapid
scanning of the whole field of the optical microscope and selection of the
effect which shall be observed: backscattering of electrons or emission
of dial selected X-rays lines characteristic for the element sought. Integra-
tion on the screen of oscillographs then reveals those details of the field,
in which the element is located, and by repeatedly resetting the selector
dial the distribution of the various elements is obtained (qualitatively and
at least semiquantitatively) within minutes. Photographic recording gives
remarkably sharp outlines of the parts of the microstructure containing
(e. g.) Ti, V, Cr, Mn, Fe, Co, Ni, Cu, Zn, Ga, etc., which match the structures
visible in the optical photomicrograph (1072).
The electronprobe requires that the specimen is mounted in a high
vacuum, withstands the heat produced by the electron beam, and is a
conductor to prevent building up a negative charge that would deflect
294: Classification Tests P.34

the stream of impinging electrons. It follows that volatile matter must

be absent.
Polished metallographic specimens satisfy all requirements, but polished
sections of rocks, ceramics, etc. must be coated with a suitable conducting
film, e. g. vacuum coating with carbon, 10 to 20 nm (mfi) thick. Loose
particles must be imbedded in a solid matrix provided with a conductive
film since otherwise they would become charged and propelled into space
when hit by the electron beam. For the study of the distribution of elements
in biological specimens, it is possible to use the mineral skeleton left
after cautious ashing of thin sections (194:).
Optical Emission Spectrochemical Analysis (78, 480, 494:, 1056). In spite
of the fact that excitation by use of flame, arc, or spark causes complete
loss of the sample, this may be justified if the metallic constituents are of
principal interest and also traces (down to 1 part per million) shall be found.
Only the elemental composition is revealed, but complex alloys, minerals,
ceramics, and glasses containing lanthanides, Ta, Cb, W, etc. do not offer
any particular difficulties.
The material under investigation may be supplied as a solid, 11 solution,
or a vapor. The size of the sample may be reduced to a fraction of a milligram,
but if no suitable precautions are being taken, one may be skeptical about
the use of very small amounts of readily volatile matter.
Chemical Analysis. Chemical analysis often permits saving parts of the
material of the sample in the form of precipitates or solutions; there is
not necessarily a total loss of the material. In addition to the elemental
composition represented by majors and minors, also some information
is obtained concerning valence state and complexes present.
The continuation of the investigation is outlined in P.36 and the
following paragraphs. Samples of 1 fig and less will suffice if the technique
of performance is properly selected. If very little material is available,
each step of the investigation may be preceded by a pilot experiment,
a control, on a known material containing small amounts of the substances
to be tested for. Success of the control will show that apparatus, reagents,
and technique ar~ suitable. A blank (experiments with solvents and
reagents only) should be performed simultaneously with the analysis of
the sample. Negative tests in the blank will then prove that substances
found have not been introduced by reagents, solvents, apparatus, or the
conditions of the surroundings (C0 2 , HCI, NH 3 , H 2 S in the air of the
laboratory, copper in the distilled water, copper oxide particles ejected
by the surface of a heated apparatus, etc.).

P.34 Organic Substances

If this was not done already, carbon compounds should be heated under
carefully controlled conditions for the determination of the melting point,
P.34 Organic Substances 295

P.21. This may lead to quick identification if only one substance is present
and a sharp melting point is obtained. If a mixture is present, it will
be preferable to separate the components for the determination of the
melting point; to this end, particles of various kinds may be selected
mechanically, Expt.61, and heated separately. If this is not possible,
other means for separation must be used.
If the melting is not "sharp", i. e., if it occurs during an interval of
several degrees, a mixture may be present. Observation of the crystallization
on cooling may show the separation of several types of crystals and of a
fine-grained eutectic mixture which either remains liquid or solidifies last.
In such instances, a mixture is obviously present, and the heating should
be repeated to obtain a record of the melting point of the eutectic mixture
and of the temperatures at which the last traces of the various types of
crystals melt (melting points of the pure components). An impure substance
will give one kind of crystals and only a small amount of eutectic mixture;
of course, the crystals may represent a solid solution (mixed crystals).
An impure solid may be purified for the determination of the melting point
as directed on pp. 151, 174.
As mentioned in P. 21, the determination of the melting point and use
of the tables of KOFLER (159) or MCCRONE (163) will still leave a choice
of three to ten compounds. As a rule, the final identification may be
brought about by determining the melting points of the eutectic mixtures
with the compounds listed in the table (159). A sample of the unknown
substance is mixed with an approximately equal volume of one of the
substances suggested in the table. The mixture is covered with a fragment
of a cover slip, heated until it is completely liquid, and then allowed to
solidify again. On reheating, the melting point of the eutectic mixture
is found as the temperature at which melting begins (well defined temper-
ature at which the melt is at equilibrium with the crystals of the eutectic
mixture). If necessary, the experiment may be repeated with the second
substance suggested by the table. Of course, the identification by melting
point and eutectic melting point should be in agreement with other facts
observed so far, and it should be confirmed by a "mixed melting point"
(m. pt. of a mixture of the unknown and a pure sample of the substance
it is supposed to be; crystallization of the melt should give only one kind of
crystals separating within a temperature interval of less than one degree).
Confirmatory tests may be suggested in the literature; or tests may be
designed after reading the descriptions of the substance that seems to be
pres:mt and closely related compounds.
If only one substance seems to be present, which decomposes on heating
so that no melting point can be obtained, clues to its identification may be
found by consideration of history, appearance, solubility, and elemental
composition which may be found according to P.35. The Tables for
296 Classification Tests P.34

Identification of Organic Compounds may be useful (54). If, however,

these lines of attack fail, it is suggested to consider the following selection
of instrumental methods in addition to the chemical procedure. If a mixture
seems to be present, low temperature sublimation may be considered in
addition to chromatography and liquid-liquid extraction as a means of
separation previous to instrumental analysis.
Crystallographic and Crystal Optical Methods (1l2) as well as X-Ray
Diffraction (82) offer non-destructive procedures for the identification of
molecular species, P. 10 to 13. LINDENBERG (1065) discusses the effect
of solvent upon the crystal habit of organic substances.
The collection of crystal optical data by WINCHELL (1l2) is being
continuously augmented by publication of data in Analytical Ohemistry
and other journals. Attention may be called to the contributions of SHELL,
POE, and WITT concerning chemotherapeutic drugs (933), anti-
histaminics (960), and steroid hormones (976, 977).
X-Ray Spectrography, P.33, can furnish, without destruction, the
elemental composition but for H, Li, Be, B, C, N, 0, and F.
Absorption Spectrophotometry provides additional means for non-
destructive identification of molecular species. Because of the complexity
of absorption spectra, recording instruments are desirable for qualitative
analysis. The visible and ultraviolet ranges (78, 81) are of obvious interest
for colored substances, dyes, and pigments.
Ultraviolet Spectrophotometry (746) is much used in the inve£tigation
of fats and oils and of hydrocarbons. A systematic procedure for the
identification of nearly 200 dangerous drugs, narcotics, and poisons (based
upon separation by solvent extraction) has been described by BRADFORD
and BRACKETT (946). Microspectrophotometry of objects as small as 1 pm
in diameter is possible (479,746).
Infrared Spectrophotometry (81 a) has been extensively used for qualitative
studies on organic substances, and CONLEY (713) has written on the use
of infrared spectroscopy in organic qualitative analysis; see also
CHERONIS (146).
The sample may be used in the solid state (KBr pellet), as a film, or
in a solvent which does not absorb in the regions of interest. Using the
bromide disk procedure, even industrial spectrophotometers do not require
more than 0.3 to 0.5 mg of sample. With a micromull technique, beam
condenser, and microscope attachment, the amount of sample may be
reduced to 50, 5, and even 0.5pg (475, 486, 487, 626,1092). Beam condensers
permit also the use of micro cells for work with down to 0.02 pI of 1 % and
weaker solutions (460, 462, 481, 486). An "Organic Microspectrophotometer"
for 10 to 50 pI solution is commercially available (1291).
Chemical Analysis (771). For a continuation of the work, it will be
advisable to first consult the books on qualitative organic analysis (35, 38,
P.34 Organic Substances 297

39, 41, 42, 46, 51, 52, 59). Obviously it will be helpful to refer to books
which have been written for work on a small scale. Of these, that of
CHERONIS (146) supplies much valuable advice on reasoning and the use
of the literature. SCHNEIDER (167) gives thorough consideration to the
procedure to be be followed with substances which have not been described
in the literature.
At this time, the task is to review all evidence collected this far and
to select additional orientation tests so that the preliminary identification
is accomplished with the expenditure of as little material as possible. Tests
should be selected, that promise a maximum of information and permit,
if possible, additional tests with the same material. Tests which may add
little or no new information should be postponed until there is a good
reason to try them.
With mixtures, separation must come first, of course, and preferably
by purely mechanical means: lifting solids out of a mixture, Expt. 61,
filtration, decantation, and separation of liquid layers. Exploratory (pilot)
experiments with very small amounts of substance are recommended for
finding the most efficient means of separation. The methods are discussed
in the books cited above; they also have been reviewed recently by MET-
CALFE (483). DAVIS, DUBBS, and ADAMS (485) have simplified Decker's (1060)
method for the elution of paperchromatogram spots for further investigation.
They cut the desired spot so from the chromatogram that a pointed tip
is obtained and then wash the spot into the point of the tip. The material
is collected within an area of just a few square millimeters; a recovery of
99.6 % was demonstrated with 2 to 10 fig adenine by cutting off and
extracting the tip of the paper. With sufficient substance, even a solid
deposit may be built up at the point of the paper, which may be lifted
off mechanically without need for any solvent. At times, well developed
crystals may be obtained for identification according to P. 10-13.
REIMERS (968) concentrates the material into a short line. On rapid
separation by thin-layer chromatography see PEIFER (972) and
WASICKY (493). For the concentrating of trace impurities and purification
by progressive freezing (zone melting), see MATTHEWS and COGGESHALL (474)
and the literature cited by them.
After separation and for the identification of the isolated components,
SCHNEIDER (166, 167) follows the system of MULLIKEN (46) who divides
the compounds into orders (according to elemental composition), genera
by "generic" tests, etc. CHERONIS and ENTRIKIN (146, 771) use a more
flexible approach in adding to the already collected evidence by testing
for solubility and degree of ionization (indicator method), for functional
groups, and for the "specific class". The tests with dilute and concentrated
sulfuric acid recommended by MCGOOKIN (42) may be tried when sufficient
material is available to observe odors and liberation of heat. Whenever
298 Classification Tests P.34

possible, the final identification should be based upon the preparation of.
several derivatives and their identification by the determination of
characteristic properties (771).
The following references to the recent literature are offered as additional
suggestions. Spot tests for functional groups and specific compounds may
be found in the books of FEIGL (121, 122), and lists of microchemical tests
for groups and compounds may be found in the reviews of CHERONIS (775,
780).
Alcohols. Melting points and eutectic melting points of the 2,4,6-tri-
nitrobenzoates of 29 alcohols are listed by LASKOWSKI and ADAMS (469).
A test for secondary alcohols and 1,2-diketones is described by FEIGL,
GENTIL, and STARK-MAYER (929).
Aldehydes. Spot tests have been described by ANGER and FISCHER (961)
and by MANNS and PFEIFER (for cyclic aldehydes) (949).
Fatty Acid8. On chromatographic separation and recognition 8ee
CHURACEK (967).
Amino Acid8. On the rapid identification of single crystals by determina-
tion of the eutectic melting point and the refractive indices 8ee LACOURT
et al. (914, 915).
E8ter8 of Nitric and Nitrous Acid8. Spot test by FEIGL et al. (939).
Nitro and Nitroso Compounds, Oximes, Hydroxylamines. Spot test by
FEIGL et al. (945).
Organic Bases. Microchemical tests by SANDRI (943).
Pyrimidine Derivatives. Microchemical tests by WEISS (966).
Synthetic Drugs. Microchemical tests are listed in the Methods of
Official Agricultural Chemists, pp. 511-515 (9).
Local Anesthetics. On identification via the melting points of salts see
BRANDSTATTER-KuHNERT and GRIMM (932).
Alkaloids. Electrochromatophoretic separation is treated by BROWN
and KIRK (938). Microchemical tests are described in the Methods of
Official Agricultural Chemists (9) and by WORMLEY (138), STEPHENSON
et al. (135, 1120), and WHITMORE and WOOD (879, 880). On ergot alkaloids
see KOLi:!EK (927), and on the test of MALQUIN-DENIGES for strychnine:
LUIS and CORAZZA (940).
Polymers. On characterization see ALLEN (33). SWANN and ADAMS (492)
describe a test for epoxy coatings.
Dyes. Testing by chromatography should not be omitted. On the
identification in food, see the Methods of Official Agricultural Chemists (9).
Inks. On electrochromatophoretic analysis see BROWN and KIRK (926).
Pigments. On classification and identification of organic pigments see
VESCE (206).
P.So Elemental Analysis of Organic Substances 299

P.36 Elemental Analysis of Organic Substances

All of the classical tests have been transposed onto the microscale, and
many of them have been modified repeatedly (146, 150, 152, 154, 161, 164,
166, 167). In this connection only the procedure of KORBL (925) shall be
described since it permits testing for most of the common elements with
the expenditure of only one sample.
Reagent. Treat a boiling solution of 19.4 g KMn0 4 in 400 ml water
with 20.4 g solid AgN0 3 • When the latter has dissolved, allow to cool.
Collect the crystals of AgMn0 4 on the fritted glass plate of a Buchner funnel
and wash with 150 ml cold water. Dissolve the washed crystals in 400 ml
boiling water, and filter the hot solution through a fritted glass filter.
Cool the filtrate and collect the crystals as before. Wash them with cold
water and dry them at 60° to 70° C. The yield is about 18 g.
To obtain the oxidant, place about 2 g of the AgMn0 4 into a clean test
tube and heat to 150° C. Sudden decomposition gives a black mass of
"silver manganite" of the approximate composition Ag 20· Mn0 2 • x 0,
which is used without further ignition and kept in a stoppered test tube
until needed.
Procedure of Korbl. Draw out one end of a glass tubing of 4-mm bore
(l-mm bore if only 5 to 20,Ltg of substance may be spared) to a,fine tip.
Then, at a distance of 6 cm from the tip, heat the tubing and let it collapse
to obtain a constriction with 0.3-mm bore. Cut the tubing 9 cm above
the constriction and select a syringe or rubber bulb which will fit this
wide end of the tube.
Place a wad of freshly ignited asbestos upon the constriction and press
it lightly together by means of a clean glass rod or steel wire. If that
much substance can be spared, mix 0.1 to 1 mg of it with about 50 mg of
the silver manganite and compact this mixture in the tube above the
asbestos wad. Finally introduce a second asbestos wad to wipe down
the interior of the tube and to hold the mixture in place. It is assumed
that even 5,Ltg. of substance and 3 mg silver manganite will suffice if the
quantities of all other reagents are reduced accordingly.
Start the reaction by cautious heating of the mixture with a microflame.
It is complete within a few seconds, whereafter the search for the products
of the reaction may be started without delay.
Hydrogen. Its presence is shown by droplets of water inside the upper
part of the tube. These may be collected in a capillary pipet and-after
testing for carbon-be used for the determination of pH, boiling point, and
melting point.
Oarbon. Do not delay this test. Connect the upper, wide opening of
the tube to the syringe or rubber bulb. Place 0.2 to 1 ml barium hydroxide
solution into an 8-mm test tube. Insert the combustion tube into the test
300 Classification Tests P.35

tube so that the tip is just above the liquid and slowly force about 5 ml air
through the combustion tube and onto the surface of the reagent solution.
A white precipitate of BaCO a at the surface of the liquid confirms the
presence of carbon.
Nitrogen. Introduce 0.1 to 0.2 ml water into the wide end of the tube
so that the drop fills the whole bore. Force the drop very slowly through
the tube and out the tip into a microcone. It will have dissolved any N0 2 ,
HN0 2 , or HNO a formed. Place a small amount of the extract upon the
surface of a drop of a 1 % solution of diphenylamine in IS-F H 2 S0 4, A blue
coloration confirms the presence of nitrogen.
Sulfur. Treat the rest of the solution obtained in the preceding paragraph
with 50,tt1 5 % (0.2-F) solution of barium nitrate. A precipitate of BaS0 4
may be collected with the centrifuge and used for further tests.
Ohlorine and Bromine. Introduce 0.1 to 0.2 ml 6-F NHa into the wide
opening of the combustion tube. By means of syringe or rubber bulb,
pass the drop slowly through the tube and into a microcone. Acidify the
extract. A white precipitate may be AgCI or AgBr or a mixture of the two.
Collect it with the centrifuge for further investigation, p. 323 (d). The solution
above the precipitate, test for:
Phosphate and Arsenate. Treat the centrifugate from the silver halides
with molybdate reagent and heat the mixture by inserting into water of
about 70° C. If a yellow precipitate forms, collect it in the point of the
cone, wash it with 2-F HNO a, and dissolve it in 3-F NH a. Treat the
ammoniacal solution with magnesia mixture to precipitate MgNH 4(P, AS)04 .
. 6 H 2 0. Wash the precipitate with I-F NH a. Transfer a few crystals to
a microscope slide, and treat them with AgNO a solution while observing
through the microscope, Expt.40. Dissolve the main portion of the
precipitate in 12-F HOI and saturate the cold solution with H 2 S. If a
yellow precipitate of arsenic sulfide is obtained, make certain that the
precipitation of arsenic is complete. Centrifuge, transfer the clear solution
to another microcone, make it ammoniacal, and treat it again with magnesia
mixture to test for the presence of phosphate.
Iodine. Wash the contents of the combustion tubing with 6-F NHa
until the last washing remains clear when acidified with nitric acid (absence
of chloride and bromide). Then extract the contents of the combustion
tube by passing through it 0.1 to 0.2 ml 30 % solution of AgNO a. In a
microcone, dilute the extract with water. If iodine is present, a silver
iodide-nitrate separates first and is converted to AgI on further diluting
with water.
Mercury. If the presence of mercury is suspected, one may forego testing
for carbon and hydrogen and pass a slow current of air through the tube
already during combustion. Droplets of metallic mercury will be found
at a short distance from the heated zone. KORBL (925) also tests with
P.36 Testing with Dilute Sulfuric Acid 301

dithizone. If material can be spared, a separate test by Carius combustion

(Expt.51) and isolation of the metallic mercury (Expt.49) seems more
attractive.
For the detection of refractory elements, organic material is usually
oxidized with or without the addition of suitable reagents. SPIALTER and
BALLESTER (491) describe a simple procedure for the ashing of organosilicon
and organometallic compounds. A hook of 0.8-mm radius is bent at the
end of a 0.8-mm platinum or nichrome wire. A Pyrex capillary, 1.6 mm o. d.
and 15 to 20 mm long, is placed over the wire so that it rests on the hook.
The solid or liquid sample is collected in the hook. For ashing, the wire
is held 45 degrees to the horizontal, and the hook is heated in the edge
of a just non-luminous Bunsen flame, halfway between barrel and tip of
flame. The material should ignite and continue to burn. When the burning
stops, the wire is held horizontally, and the whole capillary is heated in
the tip of the flame for about 5 seconds. Usually, a carbonaceous deposit
is now visible on the glass near the hook. The glass and the hook are
allowed to cool for one minute, and then they are again heated in the tip
of the flame (wire horizontal) until the carbon has been oxidized. Si, Ge,
Sn, and Pb give characteristic residues on the capillary near the hook.
The authors list also the burning characteristics of various types of organo-
silicon compounds.

P.36 Testing with Dilute Sulfuric Acid

If the solubility of the material in water and dilute acids has been
investigated, evolution of gases will have been already observed. A repetition
of the test with water or dilute sulfuric acid will have the principal purpose
of identifying the liberated gas, and the experiment will have to be arranged
to this end. If no efforvescence was observed in solubility tests, the
experiment with dilute sulfuric acid may serve to detect HCN, HN3 and
low concentrations of gases that are somewhat soluble (C0 2, S02' H 2S,
PH 3, AsH 3, C2H 2, C1 2) and vapors (Br2' 12)' To find small amounts of gases
contained in a large amount of some other gas (AsH3, PH3, H 2 S, or CH 4
in H 2, etc.), the treatment with dilute sulfuric acid must be carried out
so that detection of these gases becomes assured.
The tests with sulfuric acid serve the detection of anions or acidic
constituents. Obviously, one will not apply them to metals and alloys
unless one is searching for the presence of carbide, phosphide, sulfide,
arsenic, or antimony. If the material under investigation is a mineral
or ore, the test with dilute acid will serve mainly for the detection of
carbonate. Sulfides are better oxidized and recognized as sulfate, P.59.
If the material under investigation is not metallic in appearance and
the solubility has not yet been tested, it should first be tested with water.
302 Classification Tests P.36

Record all observations: solubility, rate of solution, color changes, change

of the appearance of the solid phase, heat evolved or consumed, pH of the
resulting solution, etc. The following gases may be liberated:
N2 which extinguishes a glowing splint may be derived from NHI!
salts, urea, etc. reacting with oxidants like nitrite, hypochlorite, and
peroxide;
H2 from hydrides of the alkalies and alkaline earths, and from metals
and alloys reacting with water: alkali metals, alkaline earth metals; Mg, Zn,
AI when finely dispersed;
O2 from hypochlorites; peroxides, percarbonates, perborates, persulfates
especially in presence of a catalyst;
H 2S from hydro sulfides and the hydrolysis of sulfides;
NH3 from the hydrolysis of ammonium salts (sulfide, carbonate,
acetate);
PH 3 from phosphides;
CHI! from carbides like A1 4 C3;
C2H 2 from carbides of the alkalies and the alkaline earths; and
anyone of the gases liberated by dilute sulfuric acid (see below) if the
material under investigation is a mixture containing a definitely acidic
constituent.
If the material reacts with water, the test with dilute sulfuric acid may
be carried out with the reaction mixture. The reaction with water is
allowed to come to completion, whereafter the mixture is treated with
an equal volume of 8-F H 2 S0 4 ,
As a rule, the material under investigation is treated with a small
volume of 4-F H 2S0 4 , The mixture is warmed if no reaction takes place
at room temperature. All phenomena (dissolution, color changes, separation
of new solid phases) should be recorded. The gases and vapors that may
be liberated are the following:
H 2, colorless and odorless, burning with a barely visible blue flame
to give water of pH 6 to 7, from hydrides, metals, alloys;
H 2S, colorless, odor of rotten eggs, burns with a blue flame to give
H 20 + S02 (sulfur separates from the reaction mixture in the instance
of polysulfides, or sulfide in presence of sulfite or oxidant; colored sulfides
may separate from thiocomplexes);
H 2Se, colorless, strong odor of H 2 S, burns with a reddish flame (odor
of rotten radishes) to give Se0 2 + H 20 - H 2 Se0 3; H 2Se is soluble in
water, and red flakes separate from the solution: 2 H 2 Se + O2 (air) - 2 Se +
+ 2 H 20; from selenides;
H 2Te, colorless, odor similar to H 2S, burns with a blue flame to give
Te0 2 (white) + H 20 - H 2Te0 3: H 2Te is soluble in water, and black
flakes separate from the solution: 2 H 2Te + O2 - 2 Te + 2 H 20; from
tellurides;
P.36 Testing with Dilute Sulfuric Acid 303

PH 3 , possibly with some P 2H4' colorless, garlic odor, burns with bright
yellow flame which has a characteristic emerald green core to give P 205 +
+ 3 H 20 ~ 2 HsPO 4' from phosphides or reaction of white phosphorus
with water;
AsHs, colorless, unpleasant odor, burns with a bluish white flame
(garlic odor) to give As 20 S + 3 H 20, from arsenides and all arsenic com-
pounds when acid acts upon metal with the liberation of hydrogen;
SbH a, colorless, odorless, burning with bluish green flame to give
Sb 20 a + 3 H 2 0; from antimony compounds when acid acts upon metal
with the liberation of hydrogen;
CH 4, colorless, burning with a non-luminous blue flame giving CO 2 +
+ 2 H 20; from Be2C, Al~Ca, FeaC, NiaC,' MnaC;
C2H 2, colorless, burning with a luminous yellow flame to give 2 CO 2 +
+ H 20; from the carbides of the alkalies and the alkaline earths, LaC2,
NdC 2, etc., UC 2, CU2C2' Ag 2C2;
O 2 , colorless, odorless, makes glowing splint burst into flame, from
percompounds especially in the presence of catalysts;
N0 2 , brown, unpleasant odor, from nitrites: 3 HN0 2 ~ HNO a + 2 NO +
+ H 20 and 2 NO + O 2 (air) ~ 2 N0 2 ;
C1 2 , greenish yellow;
Br 2, brown;
I 2 , violet, from halide and oxidant, reduction of halogen-oxygen acid,
slow decomposition of chloric acid;
S02' odor of burning sulfur, not supporting combustion, from sulfite,
thiosulfate, thionate (in the instance of the last two, sulfur may separate
from the solution; see also H 2S above);
CO 2 , does not support combustion, from carbonate, percarbonate,
cyanate;
HNa, garlic odor, explosive, azides, especially above the b. pt. of HNa,
37 0 C;
HCN, odor of bitter almonds, from simple and complex cyanides.
For the performance of the test, the solid is treated with a small volume
of· reagent in order to reduce the solubility loss: for 1 mg substance use
0.02 ml water or 0.03 ml 4-F H 2 S0 4 , If no reaction takes place, warm the
mixture to 90 0 C before testing for the presence of small amounts of gas.
Combustibility and color of flame are useful characteristics only when
working with 10 mg or more of solid sample per test. When working
with very small amounts, even the color and odor of the gas may no longer
be perceived, and one has to rely completely upon chemical tests.
Gram and Decigram Scale. Transfer about 10 mg of the solid to the
bottom of a test tube (best 10 mm in diameter) and add 0.3 ml 4-F H 2 S0 4
by means of a pipet so that the wall above the liquid remains dry. If there
is a rapid evolution of gas, begin the testing without delay; as soon as
304 Classification Tests P.36

the evolution stops. keep the tube lightly stoppered between tests. If little
or no gas is liberated, lightly place a stopper into the opening of the tube
and heat the contents by standing the test tube in a beaker containing
some water of approximately 90° C. The absence of efforvescence indicates
that significant amounts of difficultly soluble gases (H2' O2, N 2) are not
obtained; testing for these is omitted. Observe and record the following:

1
--------~~---------<=3

'1 I

I, /1
i

b c
Fig. 75. Testing for Gases. a with reagent paper on glass hook c; b with liquid reagent;
1-6 forming of glass loop; about 2/3 nat. size

1. Phenomena in the reaction mixture, color and odor of liberated gas.

2. If gas is copiously liberated, touch a small gas flame (burning at
the opening of a capillary) to the mouth of the test tube. A popping sound
indicates a mixture of combustible gas with air. A flame may form at the
mouth of the tube or creep down the inside of the tube. Note the color
of the flame and the odor of the combustion products (80 2 ),
3. If a colorless gas is copiously liberated, insert a glowing wooden
splint into the test tube. Oxygen is present if it bursts into flame. The
other colorless gases listed will put out the glowing.
4. This and the following tests should be delayed until the rapid
evolution of gas has stopped. Before testing for CO 2 , it is also necessary
to wait until the combustion products of the preceding test have been
displaced by the liberated gas.
P.36 Testing with Dilute Sulfuric Acid 305

A loop of glass is readily obtained as shown in Fig. 75, 1-6. A glass

thread of about O.5-mm diameter is bent by allowing it to follow the pull
of gravity while it is cautiously heated by applying a pinhead flame at
the points indicated by arrows. In step 5, the rod at the left and the loop
(by means of flat tipped forceps) are held while fusing the end of the thread
to the handle. Finally in 6, a small amount of bending by gravity pull
establishes a slight angle between handle and plane of the loop so that
the contents of the loop may be readily deposited upon a slide.
Insert the loop into a saturated solution of barium hydroxide. On
withdrawing, a clear film of solution should fill the loop. Insert it quickly
into the opening of the test tube. If much gas has formed, a white precipitate
will separate immediately (and may even dissolve again on formation of
bicarbonate or bisulfite). If there was no copious liberation of gas, insert
the loop deeply into the test tube, Fig. 75b, and warm the reaction mixture
in the test tube by standing the latter in water of 90° C. A white precipitate
may be BaCOs or BaSOs. Withdraw the loop from the test tube and
take up its contents into a capillary which may be kept in a horizontal
position until there is time for the following test: add to the contents of
the capillary a like volume of a mixture of equal parts of 4-F HNOs and
O.02-F KMn0 4 ; seal both ends of the tube and mix by centrifuging: a clear,
red solution proves the presence of CO 2 in the gas; a white or pink precipitate
of BaS0 4 (and fading of the Mn0 4 color) indicate SOa or HaS-and the
test for carbonate must be repeated with a mixture of equal volumes of
4-F HN0 3 and 0.02-F KMn0 4 in the place of the 4-F H 2 S0 4•
If lime water (saturated solution of calcium hydroxide) is used in place
of barium hydroxide, deposit the contents of the loop with the white
precipitate upon a slide and expose the drop to bromine fumes by inverting
the slide and placing it upon the opening of a bottle containing 12-F HCI
and a few drops of bromine. After 5 minutes, remove the slide and inspect
the drop under the microscope. Crystals of CaS0 4 • 2 H 20 indicate SOa or
HaS, see P. 59.
5. If violet vapors have been noticed, suspend from a glass hook,
Fig. 75 a, c, a small square or triangle of filter paper that has been impregnated
with starch. Moisten it with a droplet of water and introduce it into the
test tube. Blue color indicates 12 •
6. If the test (5) was negative, test in a like manner with KI-starch
paper. Blue color indicates CIa, Bra, or (and) N0 2 •
7. If tests (5) and (6) were negative, take up with the glass loop some
2% (O.l-F) silver nitrate solution and introduce it into the test tube,
Fig. 75b; warm the contents of the test tube to 90° C. A precipitate may be
white: AgCN or AgNs. On recrystallization from hot 8-F HNOs' both
compounds (and apparently also Ag aC2 ) give fine needles and bundles of
such. Recrystallization from NHa is suggested.
Benedetti-Pichler, Identification 20
306 Classification Tests P.36

Dip the loop 'with the white precipitate or turbidity into the gas space
of a bottle with conc. NHa and wait until the liquid in the loop is clear.
Then transfer it to a slide and observe under the microscope the crystalliza-
tion taking place as the NH3 concentration gradually decreases (compare
Expt. 32). AgCN gives fine needles that arrange themselves to imitate
the sides of polygons and of tree branches. Finally very thin plates appear
with serrated outlines, which may be circular or in the shape of leaves, etc.
AgN 3 gives fine needles and bundles of such.
Allow the drop to evaporate and then place upon the residue a large
drop of 8-F HN0 3. AgN3 gives a clear solution, whereas with AgCN (and
Ag 2C2 ) the drop of acid remains turbid.
first white, but quickly changing to yellow and light brown: Ag 2C2 • Treat
with NH3 ~and finally HN0 3 as directed for AgCN and AgN 3. The light
brown flocks seem difficultly soluble in ammonia so that the NHa solution
never becomes clear. On dissipation of the NH a, extremely small yellowish
brown squares, rectangles, and tiny grains separate.
yellow: AgaAs· 3 AgN0 3. This may happen if the AgNO a solution has
become concentrated by evaporation. Adding water gives a black precipitate:
6 Ag + HaAsOa + 3 HN0 3.
brown and quickly turning black: Ag 2 S. Transfer the contents of the
loop to a slide. Remove the liquid with filter paper (Expts.43, 45) and
wash once with water. Place upon the black precipitate a droplet of
1 % CaCl 2 (O.I-F) solution and expose to bromine vapors for 5 minutes,
P.59. Then place the preparation under the microscope and observe the
separation of crystals of CaS0 4 • 2 H 20 along the edge of the drop: sheaves
of fine needles, occasionally "swallow tail" twins.
black: possibly Ag 2 Se or Ag 2Te
PH3 + 3 AgN0 3 -->- Ag 3P (black) + 3 HN0 3,
AsH 3 + 6 AgN0 3 + 3 H 2 0 -->- 6 Ag (black) + HaAs03 + 6 HNO a,
SbHa + 6 AgNO a + 3 H 02 -->- 6 Ag + Sb(OH)a + 6 HNO a.
Transfer the contents of the loop to a microcone and add 3,tt1 16-F
HN0 3 and 10,tt1 12-F HC!. Use the acids to rinse the loop into the cone.
With frequent stirring heat the mixture for 15 minutes on the steam bath.
Centrifuge and transfer the clear centrifugate to another conel • Add an
equal volume of 6-F HCI and saturate with H 2S. A yellow precipitate
indicates As, and an orange precipitate, Sb. Alternately heat and saturate
with H 2 S to render the precipitation complete; then centrifuge and remove
the clear solution. By heating on the steam bath evaporate the latter

1 If Se or Te are thought to be present, the solution may be diluted with

water, saturated with S02' and heated for the precipitation of Se and Te. The
filtrate should be heated to expel S02 before treating it with H 2S.
P.36 Testing with Dilute Sulfuric Acid 307

inside a small area upon a microscope slide (compare Expt. 44). Dissolve
the residue in a droplet of 3-F HNO a) expose it to NHa to make it ammoniacal,
and add a grain of magnesium acetate to precipitate MgNH 4P04 • 6 H 2 0
(Expt.40).
Previous to the performance of confirmatory tests (Expts. 35 and 50, 51),
the sulfides of arsenic and antimony may be separated by digesting for
10 minutes at room temperature with a freshly prepared solution of
ammonium bicarbonate in water.

Table VIII. Sensitivity of Tests According to G. C. T. CHANG (1109)

Limit of Identification
in
Ion Reagent - Circ~; I Capillary
Cell I Chamber
pg pg

Sensitivity of Tests if the Gas is Liberated with 2-F H 2 S04

C03- Ba(OH)a 0.1
S03-' S20S- Ca(OH)g, later Bra 0.5 0.3
C12 KI-starch paper 0.01
S- AgNOs 0.05 0.002
CN- AgNOs 0.05 0.05
Sensitivity of Tests if the Gas is Liberated with 1S-F H 2SO4
C03 = Ba(OH)2 and negative SOa test 0.01
C2 0 4- Ba(OH)2 and negative S02 test 0.5
S03- bleaohing of aoid 0.1% KMn04 0.5 0.1
same Ca(OH)a and then Bra 0.5 0.2
SaOs- bleaohing of acid 0.1 % KMn04 1 0.5
same Ca(OH)a and then Br2 0.3 0.3
CNS- bleaohing of acid 0.1 % KMn04 0.5 0.2
same Ca(OH)2 and then Bra 0.3 0.3
S= AgNO a -+ AgaS 0.05 0.002
same AgNOa -> Ag2 S -+ CaS04 • 2 H 2O 0.3 0.3
same lead aoetate paper 0.02 0.002
Cl- AgNOa 0.5 0.01
Br- AgNOs 1 0.02
I- AgNOs giving blaok AgaS 1 0.1
CI03 - KI-staroh paper 0.5 0.05
C}- +KMn0 4 same 0.05 0.0005
Br + KMn0 4 same 0.3 0.0005
I- +KMn04 same 0.5 0.1
NO a- same 0.5 0.1

8. Insert into the test tube a very small triangle of moistened KI-staroh
paper that has been lightly blued by cautious exposure to bromine vapor.
308 Classification Tests P.36
•
The white color of the paper'will be restored if one or several of the following
gases is (are)' present:
H 2S + 12 = 2 HI + S,
S02 + 12 + 2 H 20 = H S0 2 4 ~- 2 HI,
HCN + 12 = ICN + HI,
2 HN3 + 12 = 3 N2 + 2 HI,
PH a, AsH a, and SbH a•
The last two may react with the separation of metallic arsenic or
antimony.
9. A colorless and odorless gas that burns with a blue flame may be
hydrogen, methane, or a mixture of the two. The presence of CH 4 may
be detected by testing the combustion products for CO 2, As an alternative
or confirmation, one may test the material under investigation for the
presence of carbide: FEIGL, p. 281 (121).
Milligram and Submilligram Scale. The gas may be liberated in a gas
reaction cell (Expt. 46 and p. 96), and the above suggested tests may be
used for the detection of most of the gases. G. C. T. CHANG (778) used
a circular gas reaction cell or chamber of 15-mm diameter and 5-mm height
as well as a capillary chamber of only about 1.2-,ul capacity and found
the limits of identification listed in Table VIII.
The test for carbonate requires comparison with a blank if the laboratory
air has access to the reagent solutionf'!. 'The listed limit of identification
for carbonate can be reached only by either working in a dry box (458, 465,
761) supplied with air free from carbon dioxide or by using the capillary
chamber and jacket of CHANG (778). Also variations of Emich's technique
for the sensitive detection of carbon seem feasible (150, 1141).
In all work with small amounts of gas, the volumes of aqueous solutions
should be kept to a minimum since all gases are quite soluble at room
temperature. Acetylene and carbon dioxide are soluble in an about equal

°
volume of water; hydrogen sulfide and sulfur dioxide are far more soluble.
One volume of NO, CH 4 , 2, CO, H 2, or N2 (760 mm pressure) requires
only from 21 to 67 volumes of water for complete dissolution. At 100 0 C,
the solubility is usually less than one half of that at room temperature.
Considering all circumstances, the identification of small amounts of
nitrogen, hydrogen, oxygen, carbon monoxide, or methane will require
liberation in a closed system above mercury and analysis with a micro gas
buret such as that of REEVE (721), BLACET (402, 404, 405, 408), or
BURKE (893). The sample or a short piece of capillary containing the same
may be treated with water or acid inside the conical top of a mercury-filled
gas reservoir, which may be heated by a jet of steam before drawing the
gas into the micro buret for analysis.
P.37 Test with Concentrated Sulfuric Acid 309

If the gas buret has a calibrated capillary of 0.25-mm diameter, 2 pI gas

are required to give a column of lO-mm length. On an average, this amount
of gas is furnished by 10,ug of Q. oompound (cube 01 OJ~-m.m. edge) W"h~ch
reacts with water or acid to give a gas. It will thus serve little purpose to
try working with solid samples much smaller than 10 f-lg. Using the reagents
and procedure of quantitative gas analysis, the limits of identification
may then be of the order of 0.2 to 1 f-lg.
Interesting in this connection is Chamot and Mason's test for the
evolution of a gas (118). The specimen to be tested is covered with a gel
or a very viscous liquid through which the reagent producing the gas diffuses.
The gas bubbles liberated are trapped in the viscous medium for inspection.
In a drop of freshly boiled water dissolve one or two square millimeters
of commercial sheet gelatin, just enough to get a solution that gels on
cooling. It is essential that the gel shall not set too fast. Nor shall it be
so thin as to require low temperature and considerable time for setting.
The specimen (particles of a powder on a slide or the evaporation
residue of a solution within a very small area of the slide) is warmed and
then coated with a thin film of the liquid gel, which should extend a small
distance beyond the specimen in all directions. (Collodion in ether might
be used if the specimen is soluble in water.) The preparation is t~en placed
upon a cool surface until the gel has set.
For testing, the specimen is focused under the microscope, whereupon
a drop of the reagent is applied to the jelly at a distance from the specimen.
The reagent slowly diffuses through the jelly. When it gets to the specimen,
the gas bubbles freed will persist at least long enough to be observed
before they dissolve in the gel. If the gas is not too soluble in the medium,
one may imagine it possible to take a sample into a micropipet, Fig. 59,
operated with a droplet of mercury in the tip, and then to blow it into a
reagent solution held ready in a capillary cone.

P.31 Test with Concentrated Sulfuric Acid

If the material under investigation did not react with dilute sulfuric
acid, the latter may be removed and the sample treated with the concentrated
acid. If the material reacted with the dilute acid but left an insoluble
residue, the latter may be treated with the concentrated acid. Since,
however, dilute acid may dissolve halides, nitrates, etc. without liberating
a gas or producing any other visible evidence, applying the test with
concentrated acid to a fresh sample may bring evidence which could not
be gotten with a residue from the treatment with dilute acid or water.
(It serves little purpose to test metals or alloys with concentrated sulfuric
acid.)
The following gases may be obtained with non-metallic materials.
310 Classification Tests P.37

HF, colorless, of penetrating odor, very corrosive, etching glass, is

obtained from simple and complex fluorides in the absence of silicon and
boron compounds (test performed with apparatus of platinum, lead, or
plastic) ;
SiF 4' colorless gas of penetrating odor, giving a white turbidity with
water (3 SiF4 + +
3 H 20 --. H 2Si0 3 2 H 2SiFs), obtained from simple and
complex fluorides, fluosilicates, in presence of silicon compounds (Si0 2,
silicates, glass) and in absence of more than very minor amounts of boron
compounds;
BF3, colorless gas, giving heavy white fumes with moist air, very
soluble in water, imparting a yellowish green color to flames, obtained
from fluorides and fluoride complexes in presence of boron and boron
compounds, borates;
HCI, colorless gas of penetrating odor, giving fumes with moist air
and very heavy white fumes when approached with a loop containing strong
NH3: from chlorides;
O 2, colorless, odorless gas causing a glowing splint to burst into flame:
from Mn0 2, persulfates;
CO, colorless, odorless gas burning with a bright blue flame to give
only CO 2: derived from simple and complex cyanides, thiocyanate, formiates,
oxalates, citrate, and other organic compounds;
CO 2, colorless gas with slightly acid odor, that quenches a glowing
splint: from thiosulfate, oxalate, and other organic compounds;
COS, colorless, burning with a blue flame to give CO 2 and S02: derived
from the decomposition of thiocyanate, which is accompanied by a very
disagreeable smell;
S02' colorless, odor of burning sulfur, quenching a glowing splint:
from sulfite, thiosulfate, thionate, thiocyanate, and from the reduction
of sulfuric acid by iodide, H 3P0 2, H 3P0 3, and organic compounds;
+
C1 2, greenish yellow gas of characteristic odor: from chlorides oxidants
or chlorates and perchlorates + reductant;
CI0 2, greenish yellow gas of chlorine-like odor, that explodes on heating
or contact with organic matter (glowing splint): from chlorate, 3 HCI0 3 -
-+ HC10 4 + +
2 C10 2 H 20;
Cr0 2C1 2, brownish red, heavy fumes which react with water to give
Cr0 3 and HCI: derived from chloride in presence of chromate;
Br2 (and some HBr), brown, heavy vapors of characteristic odor: from
bromide or bromate + reductant;
12, violet fumes, especially upon warming, of characteristic odor: from
iodide, or iodate (periodate) + reductant;
Mn 20 7 (1), violet smoke of very characteristic odor which is obtained
when some of the green or brown solution of permanganate in concentrated
P.37 Test with Concentrated Sulfuric Acid 311

H 2 S0 4 decomposes (detonates) because of heat (40 0 C may suffice) or

presence of reductant (organic matter);
N0 2 (from NO + 02)' brown gas of characteristic odor obtained from
nitrite or nitrate+ some reducing agent.
The mixture of the sample with the concentrated sulfuric acid changes
to dark yellow in the presence of chlorate, to green in presence of per-
manganate; such mixtures may explode and should be discarded by pouring
slowly into running water. The separation of sulfur may indicate the
presence of thiosulfate, thionate, or sulfide. Charring, which may be
accompanied by the odor of burning sugar, occurs with organic substances.
Gram and Decigram Scale. If there is any possibility of the presence
of explosive mixtures (strong oxidants, such as chlorate or permanganate,
and organic' substances), treat first a sample of a few tenths of a milligram.
If the material under investigation appears harmless, place 1 mg of it and
a like amount of glass powder into a test tube and, with a pipet, add about
0.05 ml 18-F H 2 S0 4 , If no reaction takes place, warm slowly with a micro-
flame. If a colored gas is obtained, test with the appropriate reagents
(4, below). If a colorless gas is obtained, that cannot be identified
satisfactorily, repeat the test with 10 mg sample, a like amount of glass
powder, and 0.5 ml of the acid.
In testing the liberated gas, already available information (history,
origin, nature of material) may indicate the choice of tests. If no such
clues are available, use the following outline and the technique with glass
loop and hook described in P.36.
1. Observe the reaction mixture, the color and (cautiously) odor of
any liberated gas as well as its effect upon a small triangle of moistened
pH test paper.
2. If a colorless gas is liberated at a good rate, touch a microflame
burning at the opening of a glass capillary to the mouth of the test tube:
popping sound: mixture of combustible gas and air;
gas burns with a blue flame: 00, COS;
test flame burns yellowish green: BFs;
test flame extinguished: 0° 2 , S02, HOI.
3. If the gas is colorless, try the following tests.
a) If the gas is strongly acidic, insert a loop with water. A white
turbidity indicates SiF4 , Transfer, with a capillary pipet, some of the clear
solution to a microscope slide and add a small grain of NaOl. Separation
of pink hexagons or six-pointed stars of Na 2SiF6 , P.57, confirms the
presence of fluoride.
b) If the gas is strongly acidic, insert a loop with 1 % AgNOs. If a white
turbidity or precipitate is obtained, transfer the loop into the gas space
of an ammonia bottle. When the contents of the loop have become clear
312 Classification Tests P.37

(or distinctly ammoniacal) transfer them to a microscope slide to observe

the crystallization of AgCl, Expt. 32.
c) If the gas is not strongly acidic or has a bleaching action, test with
Ba(OH)2 or Ca(OH)2 for CO 2 and S02 as told in P. 36.
d) Test for small amounts of oxidizing gas by inserting a small triangle
of moistened K1-starch paper. If it turns blue, it may be C1 2 , Br 2, 12,
CI0 2, Cr0 2C1 2, N0 2, Mn 20 7 •
e) If test (d) is negative, test for small amounts of reducing gas with
a small triangle of filter paper that has been moistened with a mixture of
equal volumes of 0.1-F freshly prepared solutions of FeCl3 and KaFe(CN)6'
If the paper turns blue, it may be S02'
f) If (d) and (e) were negative, one may test for CO by inserting a small
triangle of filter paper that has been moistened with an 0.1 % solution of
PdCI 2 · 2 H 20 in water. The test is also given by S02, H 2S, NHa, and
hydrocarbons, but significant amounts of these cannot be present.
4. Colored gases have also characteristic odors so that they may be
recognized by these properties. The following tests serve mainly for
confirmation:
a) 12 : insert a small triangle of moistened starch paper which will
turn blue. (Moisten the paper with a solution of starch in 15% NaCl brine.)
b) Br 2 : insert a small triangle of filter paper which has been dipped
into a saturated solution of fluorescein in 50 % ethanol. A change to red
proves Br 2. Test (a) must be negative.
c) C1 2 : if tests (a) and (b) have been negative, insert a small triangle
of filter paper which has been treated with a droplet of 1 % KBr solution,
dried, and then with a droplet of the fluorescein solution. A change to
red proves C1 2 •
d) Cr0 2C1 2 : insert a loop containing 1 % AgNO a. A white precipitate
forms in a yellow solution. By means of a capillary pipet transfer the
clear solution to a slide. Place upon a steam bath and evaporate to dryness.
Treat the residue with a 1 % solution of AgNO a in 3-F HN0 3 • Crystals
of Ag 2Cr 2 0 7 may appear immediately or on evaporation of the drop, Expt. 31.
e) N0 2 : insert a paper which has been moistened with a freshly prepared
mixture of equal volumes of sulfanilic acid (1 g warmed with 100 ml
30 % acetic acid) and eX-naphthylamine (30 mg boiled with 70 ml water;
the colorless solution decanted from the residue and treated with 30 ml
glacial acetic acid). Red coloration proves N0 2 (857).
The test with conc. H 2 S0 4 may be extended by adding ethanol and
warming the reaction mixture. An aromatic odor of ethyl acctate and
absence of carbonization indicate acetic acid. A green flame when the
alcohol is boiled off and the vapors are lighted proves boric acid.
VORTMANN and LIEBER (61) finally add to the reaction mixture some
metallic zinc. Liberation of N0 2 indicates presence of nitrate. A change
P.38 Flame Tests 313

of the color of the reaction mixture from orange or red to green suggests
chromate; from colorless or yellow to blue, Mo, V, or W; from green or
violet to colorless, Mn.
STEINBACH (1262) prefers the use of 15-F H sP04 which, as pointed
out by the above authors (61), lacks the oxidizing action of concentrated
H 2 S0 4 and liberates H 2 S from sulfides, HCN from cyanides, HBr from
bromides, and HI (and some 12) from iodides (instead of S02, CO, Br2,
and 12 , respectively, obtained with H 2 S0 4 ),
Regardless of the particular acid or concentration used (dilute or
concentrated), a gas that burns in air may be efficiently identified by
drying it with CaCls and then burning it with the use of a chimney (glass
tube) permitting the testing of condensates and combustion gases. For
additional criteria may serve the color of the flame, the mirror obtained
when holding a cold porcelain plate into the flame, the pH of the condensate,
and specific tests with condensate, combustion gases, and mirror.
Milligram and Submilligram Scale. The technique and limitations
have been discussed in P. 36. Some of the identification limits have been
determined by CHANG (778) and are included in Table VIII.

P.38 Flame Tests

Flame tests may be useful down to the microgram scale if some of the
material can be sacrificed. If a test on charcoal has been carried out
(P. 25, Expt. 17), the coloration imparted to the flame by the material
may have been already observed and a repetition may be unnecessary.
On the other hand, it may seem desirable to repeat the observation, possibly
with the use of a spectroscope.
It is understood that any strong flame coloration may obscure any
other color effect and also that the observed color may be a mixture
(green = yellow + blue, etc.). The perception of several colorations is
often made possible without recourse to filters or spectroscope by making
use of a difference in volatility, i. e., raising the temperature of the sample
in steps or addition of reagents.
The flame may be viewed through filters which absorb interfering
colorations. Most widely used is blue cobalt glass which should absorb
the yellow sodium light but allow passage of the radiation given by potassium.
The general technique has been demonstrated in Expt. 18. To comple-
ment it, any simple spectroscope will do, but an instrument with a wave-
length scale will reduce the need for control experiments which are required
for arbitrary scales. A lens may be used to give an image of the flame
upon the slit of the spectroscope.
For the sake of efficiency, two people should cooperate if a spectroscope
is used so that one person may concentrate upon the recording of the
314 Classification Tests P.38

spectroscopic image while the other, who is feeding the sample into the
flame, may give full attention to the proper procedure and also record the
behavior of the material: color change, foaming, melting, sintering,
volatilization, incandescence, etc. If the spectrum is somewhat complex,
it will be probably necessary to repeat the test several times, and material
would be saved by the use of a spectrograph.

Table IX. Colorations Observed in Flame Tests and During Heating on Charcoal
Approximate Location of Lines and Bands
Coloration II
Substance
Temperature nm

crimson red low Li lines: R 670.8, OY 610.3

purplish red high Sr bands: RO 686, 674, 662, 649,
646, 635, 606, B 460.7
yellowish red high Ca bands: OY 620.3, 618.2,
YG 554.4
orange yellow low Na line: Y 589.3
greenish yellow high Mo0 4
pale yel.-green H 3P0 4
yellowish green MnCl2 bands: two in 0, G 559.2, 539.2,
515.8
yellowish green low bands: laurinO, Y 548.1, 544.0,
G 519.3,491.2, two in B
yellowish green high Ba bands: G 513,524, 534, B 487;
line: G 553.5
green high CuO esp. distinct on coal
green, fleeting Mn MnCl 2 only, spectrum between
500-600
bright green low TI line: G 535.0
green low Hg(CN)2 with purple border
greenish white low Sb around bead on coal
greenish white Re
pale green high Te
pale bluish green ' Zn around the metal bead on coal
bluish white . As, Bi, Sn
sky blue low CuCl 2 lines: G 550.7,538.6;
bands: B 443.7, 441.3, 435.4,
433.2
deep blue low Se esp. distinct on coal
deep blue low Pb around bead on coal
bluish violet low In lines: B 451.1, V 410.1
violet low Ga lines: V 417.2, 403.3
violet low HgCl 2
violet very low Cs lines: B 455.5, 459.3
violet very low Rb lines: R 795.0, 780.0,
V 421.5, 420.2
violet red low K lines: R 769.9, 766.5, V 404.4
R = red, Y = yellow, 0 = orange, G = green, B = blue, V = violet, and
the same meaning in combinations as RO, OY.
P.39 Bead Tests 315

The phenomena are summarized in Table IX which also lists the

positions of the characteristic spectral lines and the approximate positions
of the characteristic bands. Notations in italics refer to bands or lines
that mayor may not be observed.

P.39 Bead Tests

The sensitivity of the bead tests for the common metals is usually far
inferior to that observed with the cobalt bead, and relatively large amounts
of material are needed to obtain a coloration. The intensity of the color
is naturally proportional to the amount of coloring matter supplied.
Furthermore the colors listed in Table X may be greatly modified, changed,
or completely obscured by the presence of other metals. Consequently
and in general, bead tests may be recommended mainly when the elements
of the middle of the periodic table are of special interest.
The material used for bead tests is not necessarily lost, however, since
the bead may be dissolved for further investigation, isolation, or identifica-
tion. In addition, inspection of beads in ultraviolet light provides a
sensitive method for the detection of uranium and various lanthanides.
Finally, use of the procedure of DONAU allows sensitive tests for some
noble metals.
The classical bead tests are performed with either borax or microcosmic
salt. Both fuse to a clear glass which is able to dissolve metal salts.
Na 2B 40 7 • 10 H 20 --'; Na 2 B 4 0 7 , spongy mass + 10 H 0,
2

Na 2B 4 0 7 + CoO --'; 2 NaB0 2 + CO(B0 2)2,

NaNH 4HP0 4 • 4,H 20 --'; NaP0 3 + NHa + 5 H 0, 2

NaPOa + CoO --'; NaCoP0 4•

Sulfidic ores and compounds should be roasted for conversion to oxides

before introducing them into the bead: 4 FeS 2 + II O2 -+ 2 Fe 20 a + 8 S02'
The molten borax glass adheres better to the wire than the phosphate.
Since both kinds of glass give about the same phenomena, the phosphate
bead is recommended solely for the confirmatory test for titanium and
wolfram (tungsten), for which it is better suited than the borax bead.
Even large borax beads may be fused on a straight wire with the same
technique as used for microbeads in Expt. 59. The procedure may be
refined for the use on a microgram scale by adopting the fusion technique
described on p. 204. Electrically heated loops for fusion under the micro-
scope have been used by EMICH (149) and by KOCH, MALISSA, and
DITGES (570).
For work on the milligram scale, which is in this instance the customary
classical scale, one end of a straight platinum wire of 0.3- to 0.5-mm diameter
and 3-cm length is sealed into the end of a glass tube which serves for handle.
316 Classification Tests P.39

Using a non-luminous Bunsen flame, the free end of the wire is heated
to redness and then quickly dipped into a small supply of borax on a watch
glass. Some salt attaches itself to the wire and is fused to a glass by heating
in the flame so as to hold the bead at the free end of the wire, Expt. 59.
More borax is added and fused in the same manner until the bead has
the desired size, 4- to 5-mm diameter at the "equator". Then some of
the material to be tested is placed upon a watch glass or slide, and a small
sample of it is picked up by touching it with the hot bead which is then
heated in the lower oxidizing zone (b in Fig. 10) until the sample has
dissolved, whereafter the color of the bead is observed while hot and after
cooling to room temperature.
More sample may be added if there is no color or the coloring is too
faint. More borax may be taken up and fused into the bead if the color
is too dark.
To obtain reducing action, the bead is heated at c or d (Fig. 10) and
then allowed to cool in the stream of gas emerging from the barrel of the
burner (base of the inner blue cone of the flame). If the particular gas
used should not give a suitable flame, reduction in a candle flame, near
the inner dark cone, may be tried.
The colorations that may be obtained are listed in Table X. If uranium
or lanthanides may be present, the bead should also be inspected under
Table X. Interpretation of the Oolors Shown by a Borax Bead
Heated in the Oxidizing Zone Heated in the Reducing Zone
--_.-------- ------------ Metal
Cold Bead Hot Bead Cold Bead Hot Bead

violet red violet red violet red violet red Nd

reddish brown brownish violet violet gray gray Ni or
Fe+Mn
pale yellow orange yellow pale green pale green U
pale yellow orange yellow green green Fe
yellow yellow brown colorless colorless Ce
yellow green yellow brown green pale green V
yellow green yellow green yellow green yellow green Pr
yellow green yellow green green green Cr
greenish blue green ruby or colorless Cu
opaque red !

blue blue blue blue Co

violet violet blue blue Co+Mn
brown violet violet colorless colorless Mn or
Ni + Co
colorless i pale yellow dark gray and i brown Mo
opaque
colorless pale yellow yellow brown yellow W
colorless pale yellow pale violet gray Ti
colorless pale yellow gray gray Sb, Bi, (Zn)
P.40 Review of Findings 317

ultraviolet light. Uranium in the amount of two millionths of the weight

of the bead still gives a green fluorescence (580). HAITINGER (1087) lists
the following fluorescence colors and limits of identification for the
lanthanides: red, 25f-tg Eu; orange, 5f-tg Sm; pale yellow, Ho; yellow,
5 f-tg Dy or 50 f-tg Gd; yellowish green, 2.5 f-tg Tb; blue, 0.5 f-tg Ce; and
violet blue, Tm.
For the confirmation of titanium or wolfram, fuse a bead of microcosmic
salt (NaNH 4HP0 4 • 4 H 20) in a loop obtained by bending the free end
of the platium wire into the shape of a U (8 -mm height, 3-mm inner width).
In the oxidizing flame, the bead is colorless, but on reducing it turns violet
with Ti and blue with W. Now add a small grain of FeS0 4 to the bead
and fuse again in the reducing flame; the bead turns blood red with both,
Ti and W.
The phosphate bead may also be used to test for silica. Most, but not
all, silicates leave an insoluble skeleton floating in the molten bead. On
prolonged heating the· skeleton slowly disappears: Si0 2 + 2 NaPO s -.
-. Na 2 SiOs + P 20 S ' Sn0 2 and other oxides are also "insoluble" and may
give the impression of a silica skeleton.
Sensitive tests for noble metals are obtained by t.he procedure described
by DONAU (1003). The test 801ution is added to the spongy mass obtained
by just dehydrating the borax on the wire. The spongy mass is then fused
to a clear glass (the proper zone of heating may have to be found by tests
with controls). The color is given by the colloidal dispersion of the reduced
metal and fades as the colloidal particles grow on prolonged heating.
The following colors and limits of identification are listed by DONAU:
red (on prolonged heating violet, blue, and colorless), 0.025 f-tg Au: yellow,
0.2f-tg Ag, brown and turbid, 0.05f-tg Pt; brown and clear, Os, Ir; brown
but slate gray in reflected light, Rh; black, Pd, Ru.
The beads are removed and the wire cleaned as described in Expt. 59.
Also other substances may be used for bead tests, and oxidizing fluxes
have been applied in connection with classical blowpipe analysis. According
to FEIGL (340) fusion with quinolinol provides sensitive tests for ferric
iron (greenish black) and vanadium (brownish black) and WEST and
GRANATELLI (453) use melts of the same compound for the detection of
water and the identification of chloride, nitrate, sulfate, uranyl, chromium,
copper, zinc, calcium, and magnesium by the separation of characteristic
crystals.

P.40 Review of Findings

At this time the nature of the material under investigation may have
been already recognized, and it is merely necessary to select confirmatory
tests which prove the correctness of the identification beyond any reasonable
318 Classification Tests P.40

doubt. These need not be chemical tests, and the recognition of a particular
variety (modification) would even require the observation or determination
of physical properties such as color, crystal form, melting point, density,
electric conductivity, diffraction pattern, etc. See also P.73.
If the identity is still in doubt, a decision concerning the general nature
of the material under investigation must be made to have a guide for the
continuation of the search. It should be possible to decide whether it is
a simple substance (one compound), a simple mixture (two or three com-
pounds), or a complex material (alloy, glass, silicate), or a mixture of
such materials (soil, rock). Even this decision may require additional tests.
It may be necessary to determine whether or not silica is present by warming
a sample with sulfuric acid and ammonium fluoride as directed in P. 57;
the material used for this test may be fully utilized since the residue left
after evaporation of the acids may be used for the search for cations.
If this is the intention, the amount of sample taken for the test should
be carefully estimated.
If only little material is available, one may consider what useful tests
might be performed with condensates, residues, or solutions left from
preceding tests.
The presence of ammonium compounds may have been discovered by
this time. If not, a fresh sample is usually needed. The test may be per-
formed at this time and furnish additional information concerning the
sample. The technique of Expt. 46 may be used or that with test tube
and glass loop described in P.36.
Use as little NaOH solution as possible, and observe the behavior of
the sample: dissolution, color changes, separation of new phases. If no
NHs is liberated and the test is negative, allow the reaction mixture to
cool if it has been heated, add some Na 2 S for the conversion of all metallic
compounds to sulfides and again test for NHs; it may now be liberated
if it was held in a compound not decomposed by alkali alone. The reaction
mixture may contain characteristic sulfide and hydroxide precipitates
and thiocomplexes of various metals (As, Sb, Sn, Hg, Mo, W, V) in solution.
It may be used for the detection of metals (61).
If the material under investigation appears to be a single compound,
systematic testing should be able to discover the identity. If a solvent
has been found, the solution obtained in P. 31 may be treated with a series
of reagents to find the analytical group to which the metal(s) belong.
When working in the test tube or centrifuge cone, first add a drop of
3-F HOI. If there is a precipitate (AgOl, Hg 201 2 , PbOI 2 , TIOl, SbOOI, BiOOI;
Si0 2 • X H 20, H 2W0 4 , and metal hydroxides may separate if the solution
was alkaline), it is a matter of choice whether or not to collect it for further
investigation. The next reagent is then added to the filtrate or decantate
from the HOI precipitate. As an alternative, one may allow the HOI
P.lO Re.view of Finding:; 319

precipitate to settle and add the next reagent so that one may observe
first what it does to the clear solution and later, on mixing, its effect. upon
the precipitate.
Next treat with a drop of 3-F H 2 SO t • A white precipitate lllay he
BaSO Jl SrSO t , CaS0 4 • 2 H 20, or PhSO.. ; it will hardly interfere if left
in the solution which may be next treated with a little methyl red and
6-F NaOH until distinctly alkaline. Just record the appearance of an~'
precipitates, test their solubility by adding a small excess of NaOH, and
then add sodium sulfide. Record the sulfide precipitates and the color
of the solution. Either treat the lllixture with 3-F H 2 S04 until acid, or
separate precipitate and solution and treat them separately with the acid.
When working with a droplet on a slide, expose first to fumes of HOI and
then to NHs vapors until alkaline. Add a grain of Na 2CO S , expose to an
atmosphere of H 2 S, and finally to fumes of HCl until again acidic. Variation!>
of the procednre and additional tests should be made as circumstances
suggest.
If the material is insoluble in acids, fusing a very small sample with
some NasO a (which has been tested by a control fusion with OrsOs) may
give valuable information, see also P. 42.
If everything ei<le fails, try a method of systematic elimination. Prepare
a list of all possible substances and then cross off the items that are definitely
excluded by the observed facts. Recall that many things are mutually
exclusive. The presence of an oxidant excludes strongly reducing substance;,;
in the same solution, and ,vice 'versa (compare P. 70). Solubility excludes
insoluble combinations, and insolubility excludes soluble substances. If
either a cation or anion is once known to be present, the solubility and
color or lack of it may exclude whole series of combinations. Tables of
properties of inorganic substances (8, 10) may be used to search for substances
having the properties of the material under investigation and to exclude
imagined possibilities; see also P. 73.
If the material seems to be a mixture of two or several simple solid
substances, it may be best to perform a mechanical separation (Expt. 61)
and to test the components separately starting with P. 5. If this procedure
is not feasible, fractionation by extraction (alcohol, water, a,mmonia, acid)
or by a combination of extraction and sublimation may be tried. If fractions
containing only one or two solutes may be obtained, the task is greatly
simplified. Solutions containing not more than two substances should
not require separations but permit finding the cations and anions involved
by systematic testing (with Hal, H 2 S0 4 , NaOH, NallS, NaaCO a, etc.)
possibly with the aid of a few filtrations or decantations.
An elaborate procedure of systematic separations may be required to
find the essential composition of complex substances such as alloys, ores,
silicates. etc. Of course, ana.lY8is b~' systematic separation may be chosen
320 Classification Tests P.41

even for simple substances when very little material is available for testing.
One will try to avoid it, however, because of the effort and time involved
if economy with the material under investigation is not a compelling reason.
It is assumed, of course, that the problem which prompts the investigation
is important enough to warrant its continuation. Depending upon the
circumstances, it might suffice to know that the material is a silicate
rock, a soil, some common glass, or an alloy steel.
More specific information may appear desirable when considering that
there is a great variety of alloy steels, that inorganic glasses may be silicates,
borates, phosphates, or even combinations of sulfur, selenium, arsenic,
and thallium which become liquid above 100° 0 but have the chemical
resistance of silicate glass. Industrial ceramics also include a wide and
growing assortment of ingredients such as MgO, Mg(Al0 2 )2, BeO, Oe0 2,
Zr0 2 , ZrSi0 4 , Th0 2 as well as oxide-metal and oxide-carbide combina-
tions (6, 12).
If it is decided to use an elaborate separation scheme, it will be well
worth to keep in mind the advice given by NIEUWENBURG and LIGTEN (47)
that much time and effort may be saved if the work is started only after
making certain, by preliminary tests, which substances form the major
constituents. During the performance of the separations, one may then
focus the attention upon the minor constituents and the clarification of
those details which the orientation tests failed to reveal or were unable
to decide.
If very little material is available for investigation, one may also
consider the advisability of a sodium carbonate fusion with the original
material or that part of it which is insoluble in water; it may permit
identifying the anions in the aqueous extract of the melt, while the residue
from the aqueous extract is used for the separation of the cations.

P.41 Dissolution of the Sample

T~e preparation of a solution of the material under investigation is
necessary for a continuation of chemical testing regardless whether just a
few simple confirmatory tests are needed or an involved separation has to
be carried out. A suitable solvent may be known already from preceding
trials, or it will haye to be found now by turning to the tests of P.3l.
If a mixture of substances is present, it may not be possible to find a
solvent that would dissolve all substances. The use of several solvents
becomes necessary, and the outcome may depend upon the order in which
they are applied. Thus a mixture of Ag aP0 4 and Pb a0 4 will dissolve if
first extracted with dilute nitric acid and then boiled with HOI, but it
will leave insoluble AgOI if the order is reversed. The proper solvents
and the order of their application may be suggested by the history of 'the
P.41 Dissolution of the Sample 321
material and the findings in the orientation tests. Obviously, but not
necessarily, one may try to avoid HCl if silver is present or H 2 S04 if alkaline
earths must be expected.
Metals and alloys are traditionally dissolved in HNO a since this will
prevent the loss of S, P, As, and Sb as hydrides. Silicon, tin, and antimony
are converted to insoluble oxides; the carbon of carbides is oxidized, but
graphite is not attacked. The residue of the treatment with nitric acid
may therefore contain graphite, Si0 2 , Sn0 2 , Sb 20 5 ; if Sn0 2 is present,
it carries down also P0 4 and As0 4 and small amounts of Bi, Cu, Pb, Fe, etc.
Of course, various metals (AI, Cr, W, Au, Pt, Ta, Nb) and alloys (of Au, Pt;
stainless steels) are not dissolved by nitric acid. Aluminum dissolves in
HCI and (readily) in strong NaOH. Gold, the platinum metals, and their
alloys may be successfully treated with aqua regia. Powerful oxidants
+
(HCI0 4 HNOa or fusion with Na 20 2) are most successful with alloy steels.
For dissolution in nitric acid, cover 1 g (mg, flg) of the metallic material
with 0.01 liter (ml, fll) water and, in small portions, add an equal volume
of 16-F HNOa so that the reaction never becomes too violent. If necessary,
make dissolution complete by finally heating upon the steam bath. To
obtain a more complete separation of tin and silica, the mixture may be
evaporated on the steam bath and the residue extracted with 3-F HNOa,
but this is not recommended if much iron (or other readily hydrolyzed ion)
is in solution. Finally, separate residue and solution, preferably with the
use of the centrifuge. For a simple scheme of separation see the literature (55,
56, 57).
Non-metallic materials (if there is no reason for a different procedure)
are traditionally treated, in the order of listing, with water, dilute HCI,
concentrated HOI, dilute HNO a, concentrated HNO a, and aqua regia
(I volume of concentrated HNOa +
3 volumes of concentrated HOI).
The preliminary determination of the solubility will have shown which
of these steps may be omitted; it may have shown that the material is
insoluble in acid. For insoluble materials and the treatment of a residue
insoluble in the acids listed see P. 42.
For obvious reasons, one tries to use as little solvent as possible. The
following amounts may be profitably taken per 1 g (mg, flg) of sample
or what is left of it from the preceding extraction(s):
0.01 to 0.05 liter (ml, fll) of water depending upon whether only part
or the whole of the sample is dissolved;
0.005 to 0.014liter (ml, fll) 3-F HOI;
0.003 to 0.0061iter (ml, fll) 12-F HOI;
0.002 to 0.01 liter (ml, fll) 4-F HNO a ; and
0.002 to 0.006 liter (ml, fll) 16-F HNOa or aqua regia.
The determination of the solubility will have shown whether the
dissulution may be carried out at room temperature or whether it is
Benedetti·Pichler, Identification 21
322 Classification Tests P.42

necessary to heat the mixture. Concentrated HCI and aqua regia should
not be heated above 50° C since this would cause a rapid loss of HCl.
Strong acids are most effectively applied in small portions with removal
of the solution obtained before adding the next portion of acid.
If the solvents are used sparingly, the resulting solutions will not
contain an objectionably high concentration of acid. Solutions in nitric
acid may be evaporated to dryness on the steam bath without any danger
of losing a volatile compound of metals. The residue may then be dissolved
in an acid of the desired concentration. Solutions in aqua' regia may be
evaporated after adding an equal volume of 16-F HN0 3 •
The procedure of NOYES and BRAY (49, 162), P.68, is definitely
recommended if elements giving volatile oxides or halides, platinum metals,
and elements of the tungsten and tantalum groups (W, Mo, V, Ti, Nb,
Ta, Zr) are to be detected. It starts with the solid sample, and the dissolution
is part of the procedure and is conducted in such manner that the bromides
of Se, As, and Ge as well as the tetroxides of Os and Ru are simultaneously
isolated in distillates. A suitable technique for the distillation is available
even for the microgram scale of work, Expt. 63.

P.42 Treatment of Substances Insoluble in Acids

The systematic treatment of insoluble substances is part of the scheme
of NOYES and BRAY, P. 68, and is a matter of little concern if this scheme
of separation is used. The following discussion is intended as aid in systematic
testing and in the use of the classical hydrogen sulfide scheme of analysis.
A selection of procedures is given. Suitable techniques are described
for the gram scale, p. 68; the centigram scale, p. 81; the milligram scale,
p. 104; and the microgram scale, p. 193.
a) Heating with H2F 2 and H 2S0 4 in platinum apparatus will bring
into solution: Si0 2, ZI0 2 , and silicates.
Per gram (mg, flg) of solid use 0.001 liter (ml, fll) 9-F H 2 S0 4 and
0.003 liter (ml, fll) 24-F (50%) HF from a dispensing plastic bottle (as an
alternative, 2 ml 18-F H 2 S0 4 and 2,5 g NR,F may be used). Use a fume
hood and be certain to avoid getting any hydrofluoric acid on the skin.
Allow the reaction mixture to stand for five minutes and then evaporate
at low temperature so that there is no boiling at any time. Finally raise
the temperature to drive off the sulfuric acid. Add to the residue a small
amount of 4-F H 2 S0 4 just sufficient to obtain a clear solution and again
evaporate until fumes of S03 appear. This should remove all fluoride.
The residue may be dissolved by rinsing with water into apparatus selected
for the performance of the analysis.
Silicon and boron are completely removed by the treatment, and
there may be significant vaporization of As, Ge, Se, Re, Sb, and Cr (as
Cr0 2F 2 1); also some manganese may be lost (446). If fluoride is left behind,
P.42 Treatment of Substances Insoluble in Acids 323

it may interfere with the precipitation of aluminum and possibly other

elements by NHs.
b) Digestion with HsF 2 in platinum apparatus will dissolve the metals
Nb and Ta as well as their oxides Nb 2 0 s and TasOs. Treating the solution
with KF may precipitate the somewhat difficultly soluble K 2TaF 7; the
more soluble niobium compound may hydrolyze in weakly acid solution
to the readily soluble oxyfluoniobate K 2NbOFs' Evaporation with sulfuric
acid is not advisable since hydrolysis will bring back the insoluble pent-
oxides.
c) Fusion with Na 2COS (K 2CO S for Ta 20 S and Nb 20 S ) in platinum
apparatus for silicates, chromites, aluminates, Fe 20 s, Al 2 0 S ' C('20S' CaF2 ,
Ba80 4 , 8r80 4 , Ca80 4, and insoluble phosphates and fluorides of Ti, Zr,
Th, and the lanthanides.
Treat 1 part (by weight) of the finely ground sample with not more
than 4 to 6 parts of anhydrous Na 2CO S ' Fuse 'until the melt becomes clear
or until efforvescence stops, 1000 to 1200° C. The treatment of the melt
is simple when the work is performed on a small scale. Useful advice
concerning removal of the meH from a crucible may be found on p. 846 of
Applied Inorganic Analysis (7). The melt may be decomposed with dilute
acid as practiced in quantitative analysis, or it may be extracted with
water and the aqueous extract used for the search for anions. It should
be understood that the aqueous extract may also contain antimonate,
stannate, plumbate, zincate, aluminate, chromite, chromate, vanadate,
wolframate (tungstate), etc. in addition to silicate and the usual anions.
A green color of the melt indicates manganate and a yellow color, chromate.
d) Fusion with KNaCO s (mixture of the carbonates in the mole ratio
1: 1) in porcelain for the halides of silver and insoluble salts of lead: PbS0 4
and PbCr0 4 , fused.
Proceed as under (c). Extract the melt with water and dissolve the
carefully washed residue in 3-F HNOa• The anions will be found in the
aqueous extract.
e) Fusion with K 2 8 20 7 in platinum, porcelain, vitreous silica, or glass
apparatus for the dissolution of rhodium metal (finely divided), Fe 2 0 s,
Cr 2 0 3 , Al 20 S, Ti0 2 , Zr0 2 , Th0 2 , and insoluble phosphates and fluorides
of Ti, Zr, Th, and the lanthanides.
If the anhydrous pyrosulfate is not available, heat some KHS0 4 until
no more steam is given off and the melt becomes quiet, 2 KHS0 4 -+
-+ K 2 S 2 0 7 + H 2 0. Pour the melt into clean porcelain dishes, break up
the thin sheets of salt while still warm, and store in a stoppered bottle.
Use 3 parts (per weight) of pyrosulfate for 1 part of solid sample which
should be finely powdered. Fuse at as low a temperature as possible to
prevent rapid decomposition of the salt, K 2 S2 0 7 -+ K 2 S0 4 + SOs, which
would not allow sufficient time for the action of the SOs upon the sample.
21·
324 Confirmatory Tests P.43

Just maintain a liquid flux until the sample has completely dissolved.
Mter cooling, dissolve the melt in 3-F H 2 S0 4 ,
f) Fusion with Na 20 2 in apparatus of nickel or iron for the dissolution
of carbon, wolfram metal, finely divided ruthenium and rhodium metal,
alloy steels, insoluble chromium compounds, chromite, stannic and anti-
monic acids, silicates, Ti0 2 , Zr0 2 , W0 3 , and acid insoluble carbides, borides,
and nitrides.
When working on the gram or centigram scale, mix 1 part (by weight)
of the finely powdered sample with 5 parts of dry peroxide and cover the
mixture with 1 more part of the reagent. Use eye protection. Make certain
that flame gases will not get to the mixture. Heat to first expel any water
from the mixture, then raise the temperature gradually to about 700 0 C
and hold it there for one minute. Mter the melt has cooled to room
temperature, decompose it by the gradual addition of small amounts of
water in such a manner that the violent action will not cause loss of material.
With like caution, finally acidify with 3-F HCl. Since the apparatus is
strongly attacked during the fusion, the solution will contain considerable
amounts of Ni or Fe.
Concerning the use of zirconium for fusions with KOH see DODSON (489).
g) Heating in a current of hydrogen is convenient for the reduction
of the halides of silver and the insoluble oxides of antimony and tin. Water
may be condensed for identification, and the hydrogen halides may be
absorbed to the same end.
P.43 Confirmatory Tests
The elements are taken up in the order of the groups, starting with
the alkali metals, group I A, of the long form of the periodic table.
Controls and Blanks must be carried out before applying a test
to a sample which is difficult to replace. Closely imitating the conditions
of actual use, try one control with a moderate amount of substance X
sought and another with a small amount. Some more practicing may be
desirable if the experimenter is not familiar with the test.
Controls are tests with a known amount of sought substance X.
Blanks are controls closely imitating the conditions under which
the test is applied, but with the sought substance X absent.
The Sensitivity Statements are made according to F. L. HAHN (962):
Limit of Identification, L. I., is given in gram of X;
Limiting Concentration, L. C., is given in g X/ml;
Limiting Proportions, L. P., are given as ratio of largest mass of inter-
fering substance per unit mass of X sought.
The statements are given in the following forms:
If a solid sample is used:
[p L. I.; symbol(s) of interfering substance(s): log L. P.]
P.44 Group I A: Alkali Metals 325

If a sample solution is used:

[p L. C. - pL. I.; symbol(s) of interfering substance(s): log L. P.)
P L. I. = - log L. I. P L. C. = - log L. C,
P L. C. - p L. I. = log v.
v = volume of test solution in mI.
It is suggested, however, to take the sensitivity data with reservations.
The decimals are given mainly to permit calculation of the test volume.
It must be kept in mind that limit of identification and limiting concentra-
tion are not exactly related via the test volume as the equations pretend.
To obtain the limit of identification, slide tests are usually allowed to
evaporate so that the test volume becomes an unknown quantity; in
addition, a refined method of observation may be used. The equations
silently assume that the techniques of working and the methods of observa-
tion retain their efficacy regardless of scale of work; this, of course, is not true.
The tests described represent merely the preferences of the author.
Many others may be found in the literature.

P.44 Group I A: Alkali Metals

No.3: Lithium, 6.939
Flame Test (9.0).
Crimson red flame, see P.38 and Table IX..
Hexamethylene and Potassium-Iron (2 or 3) Cyanide (5.7-7.2).
The test solution must not contain ammonium salts or metals other
than the alkalies (854). Place a small droplet of test solution upon a slide
and evaporate to dryness. If necessary, ignite for the removal of ammonium
salts. Place a small drop of a 15 % solution of hexamethylenetetramine
upon the residue. When the latter has dissolved, take up the solution with
a capillary pipet for transfer to another slide. There deposit it to give
two drops of like size and next, but not too close, to one another. Place
into one drop a small grain of KaFe(CN)6 and into the other, one of K 4Fe(CN)6'
The ferricyanide gives yellow octahedra with lithium. These should appear
upon addition of the reagent; on evaporation, also blank tests give octahedra.
The ferrocyanide produces in presence of lithium short rods which may
combine to give X.-forms and radial clusters. The crystals seem to be
isotropic.
Lithium-Zinc-Uranyl Acetate (3.4-6.4) (875).
The test should be used only after the lithium has been separated from
sodium and the other alkali metals.
Reagent. Dissolve 10 g uranyl acetate dihydrate in 6 g 30% acetic acid
and 49 ml water; dissolve 30 g zinc acetate trihydrate in 3 g 30 % acetic acid
and 32 ml water; mix the two solutions and filter after 24 hours (7).
326 Confirmatory Tests P.44

Performance of Test. Place a droplet of the neutral test solution upon

a microscope slide and next to it a droplet of like size of the reagent solution
so that the two drops are about 1 mm apart. Connect the two droplets
by drawing a narrow channel with a glass needle. Lithium-zinc-uranyl
acetate separates immediately and mostly in granular shapes. Regularly
developed octahedra are quite abundant and can hardly be confused with
the rather characteristic, often large, elongated forms of the sodium-zinc-
uranyl acetate. With the latter, sturdily developed prismatic crystals
and elongated hexagons are frequent, whereas regularly developed 'octahedra
are rarely found.
On evaporation of the test drop, the reagent itself gives crystals that
might be mistaken for those of the triple acetate. They appear along the
circumference of the drop. Between crossed polars (nicols), however, the
crystals of the reagent appear bright and colored, whereas even large
crystals of the triple acetate show only a light gray of the first order.
Insertion of a first-order red selenite plate reveals the composite structure
of many of the octahedral crystals of the triple acetate; some sectors of
these crystals appear yellow while others, at the same time, display the
blue of second order.

No. 11: Sodium, 22.9898

Flame Test (8.4).
Orange yellow flame absorbed by blue cobalt glass, see P.38 and
Table IX.
Sodium-Uranyl Acetate (1.9-5.2; Li: 2.7 (1.4-4.7; K: 2.7).
Obtain a neutral solution of the chloride, nitrate, or sulfate, which
contains about 1 to 10 mg Na/ml. Deposit a small droplet of this solution
upon a slide and add a small grain of ammonium-uranyl acetate. The
isotropic tetrahedra (triangles) of the sodium-uranyl acetate appear
immediately after adding the reagent. The corresponding lithium salt is
rather soluble, but may crystallize along the circumference of the drop
when the concentration rises because of evaporation. The lithium salt
gives prismatic crystals, irregular hexagons, and forms which more or less
imitate the appearance of the octahedron or the tetrahedron. The crystals
of the lithium salt, however, are strongly anisotropic and easily distinguished
from those of the sodium salt. The presence of 10 parts of lithium for
1 part of sodium does not prevent the instantaneous separation of the
tetrahedra of the sodium-uranyl acetate (875).

No. 19: Potassium, 39.102

Flame Test (7 ~; Na: 2.3).
The bluish violet flame may be made visible in the presence of sodium
and lithium by means of blue cobalt glass. See P. 38 and Table IX.
P.44 Group I A: Alkali Metals 327

Potassium Chloroplatinate (3-7; N a: 1.2).

The test cannot be carried out in presence of NHl! Rb, Cs, TI(I), organic
amines, and alkaloids, all of which are also precipitated. Iodide, sulfate,
and nitrate interfere.
If Rb, Cs, TI(l) may be assumed to be absent, the performance of the
test is simple: Place upon a slide a solid sample or a volume of test solution
corresponding to about 1 p,g K. Evaporate the solution to dryness. If it
seems advisable, ignite the evaporation residue or the solid'sample for the
removal of ammonium salts and organic substance. Treat the cold residue
with 1 p,l 5 % H 2PtCl e solution. If potassium is present, yellow octahedra
of K 2PtCI e separate immediately. The test is not reliable in a laboratory,
the air of which is laden with NH 3 •
The following procedure is recommended if the potassium has been
separated from rubidium and cesium: the resulting fraction containing
the potassium is evaporated and, if necessary, the residue is ignited to
remove ammonium salts.
Prepare an aqueous solution which contains approximately 10 mg K/ml.
Of this solution, deposit two separate drops, each of about I-p,l volume,
side by side upon a slide. Place next to them in a systematic manner four
more drops of like volume, two drops of pure KCI solution and two drops
of pure RbCI solution, all of them containing 1 mg metal/ml. Gently
heat the slide to evaporate all drops to dryness. When the slide has cooled
to room temperature, treat three of the residues--one of each kind-with
about 1 p,l each of 5 % H 2PtCl e. Without delay treat the other three drops
with about 1 p,l each of a solution of chlorostannic acid obtained by dissolving
2 g SnCI4 • 5 H 20 in 5 ml 13-F HCI and 15 ml water.
If pure potassium salt has been isolated, the residues of the test solution
will behave exactly as the residues of the pure KCI. solution: they will
give a copious precipitation with chloroplatinic acid, but no precipitate
with the stannic chloride solution. If the residue of the isolated fraction
contains heavy alkali (Rb, Cs), comparison of the chlorostannate precipitates
will show whether or not the heavy alkali may account for all of the
chloroplatinate obtained with the unknown residue. Make the comparison
of quantities with the unaided eye after placing the slide upon a dark
paper (875).
No. 37: Rubidium, 85.47
Flame Test (6.7).
All alkalies and the strontium interfere so that their presence calls for
the use of a spectroscope. Pure rubidium salts color the flame violet.
See also P.38 and Table IX.
Rubidium-Silver-Gold Chloride (3.7-7.0).
Ammonium must be absent; K, Li, Na, and the alkaline earths do not
328 Confirmatory Tests P.44

interfere. Rubidium may be recognized in presence of cesium which gives

a more insoluble compound of the same type (ll8, 1008, 1012, 1013).
Reagent. Saturate with solid silver chloride a 5 % solution of AuOls '
. 2 H 20 in 13-F HOI.
Procedure. If a solid is to be tested, place a droplet of the reagent upon
a slide and introduce a small partiCle of the solid into the edge of the drop.
To test a solution, evaporate some of it upon a slide so that a few micrograms
of solid residue remain behind in a very small area; then place upon the
residue a droplet of the reagent.
If any cesium is present, it is immediately precipitated in finely granular
form, but after some time black squares, hexagons, triangles, four- and
six-pointed stars may be recognized with a magnification of 200 or more
diameters. Well developed orthorhombic plates and prisms of the more
soluble rubidium salt, Rb 2AgAuOl 6 (1), which vary in color from yellowish
red through red to almost black depending upon their thickness, appear
after most of the cesium salt has separated. Twinning may take place,
and radial clusters of rods may form. In addition, the crystals of the
rubidium salt may show a play of spectral colors on their brilliantly
reflecting surfaces.
No. 55: Cesium, 132.905
Flame Test (6.0).
The violet coloration is hidden in presence of anyone of the other
alkalies or strontium, and a spectroscope must be used in such instances.
See also P.38 and Table IX.
Cesium Iodobismuthite (4.4-7.7; K, Rb: 2 or better).
Reagent. Dissolve 0.3 g Bi 20 s in 1 ml HI, sp. gr. 1.6, and dilute with
2 ml water (ll59).
It is assumed that the test solution does not contain ammonium, organic
amines, alkaloids, or heavy metals.
Place a droplet of the solution to be tested upon a slide and deposit
an equal volume of the reagent close to it. By means of a glass thread,
make a connecting. channel. The separation of hexagonal plates or six-
pointed stars which may be, depending upon thickness, yellow, orange,
red, or nearly black, proves the presence of cesium. If it happens that
the crystals do not separate immediately after adding the reagent, allow
the test to stand until complete evaporation has taken place. Then search
for the hexagonal stars or plates which are dark between crossed polars
and remain dark during the full rotation of the stage. The reagmt itself
gives yellow to dark red crystals which are strongly birefringent and
separate as needles, rods, prismatic plates, diamonds, dendrites, or in
ornamental patterns (875).
Cesium-Silver-Gold Chloride. See under Rubidium.
P.45 Group II A: Alkaline Earths 329

P.45 Group II A: Alkaline Earths

No.4: Beryllium, 9.0122
Test with Quinalizarin (5.5-6.8; alkalies, Ca, Sr, Ba, AI, Zn, Cd, Sn,
Bi, Se, Te, Nb, Ta: 2.5; Fe, Mn, Co, Mo, W, As, Sb, Ag: 2.0).
In alkaline solution, the violet color of the dye is changed to a pure
blue on adding beryllium (1144). Ammonium, magnesium, Y, Ti, Zr, Th,
lanthanides, Y, uranyl, and Cu must be absent. The sensitivity given
above refers to the use of a white spot plate; it may be improved with
the coloriscopic capillary, Expt. 55.
To a drop of test solution, which must not contain free acid, add a
drop of a saturated solution of quinalizarin in ethanol (0.005%) and then
a drop of 0.1-F NaOH. Perform a blank and compare the colors.
To test for beryllium in presence of magnesium which produces the
same color change proceed as follows. To a drop of the "neutral" test
solution add two drops of a 0.05 % solution of the dye in 2-F NH3 and
then 1 ml saturated bromine water. The blue solution may become less
intensely blue, but keeps the color if beryllium is present. Compare with
a blank that contains some magnesium.
Test with p-Nitrobenzene-azoorcinol (5.3-6.7; alkalies, alkaline earths,
AI, Y, lanthanides, Ti, Zr, Th, Nb, Ta, Y, Mo, W, Fe, Bi, Ca, As, Sb, Sn,
Se, Te: 2.5; Ag, Pb, Mn, Ni: 1.5; Mg, Cu, Zn, Co must be absent).
The yellow alkaline solution of the dye gives an orange red lake with
beryllium; magnesium gives a brownish yellow lake that would interfere
with the beryllium test.
Place a drop of 0.025 % solution of p-nitrobenzene-azoorcinol in 1-F
NaOH on filter paper. Take the "neutral" test solution into a capillary
pipet and deliver it to the center of the yellow spot of dye. Finally add
another drop of dye solution. Depending upon the amount of beryllium,
either the whole spot or just its center becomes orange red.

No. 12: Magnesium, 24.312

Magnesium-Ammonium Phosphate (4.7-8.0; Na, K: 3; Ca: 1.7).
All cations that are precipitated by phosphate should be absent. If
calcium is present, treat the test solution with some ammonium citrate.
Mn, Zn, and Co give metal-ammonium phosphates of the same appearance,
but the Zn and Co compounds are soluble in ammonia, and the Mn salt
may be recognized by the formation of brown Mn0 2 when treating the
crystals with H 20 2 , W. BOTTGER.
Transfer the test drop to a slide and evaporate to dryness. Dissolve
the residue in such a volume of 2-F HN0 3 to give a 0.1 to 1 % solution
of magnesium salt. Invert the slide and place the hanging drop over
the opening of a bottle with 12-F NH 3. After a few minutes, transfer
330 Confirmatory Tests P.46
the preparation to the stage of the microscope and push a small grain
(0.2,tt1 volume, 0.6-mm diameter) of Na 2HPO, into the edge of the drop.
Observe with a magnification of about 80 diameters. MgNH 4PO,' 6 H 20
forms dendrites, X shapes, and prismatic crystals that are insoluble in
dilute ammonia. See also Expt. 40.
Test with Magneson (4.6-5.6; Ca, Sr, Ba: 2).
Ni, Co, and Cj give the same test and must be absent. Ammonium salts
are best removed.
On the spot plate, in a micro cone , or in a capillary treat some of the
test solution (which should not contain free acid) with a like volume of
a 0.1 % solution of p-nitrobenzene-azo-resorcinol in 50% ethanol and a
volume of O.l-F KOH equal to the volume of test and dye solution. A blue
coloration or precipitate indicates magnesium; a blank should remain red.

No. 20: Calcium, 40.08

Flame Test (4, 1).
O.lly the volatile salts, especially the chloride, color the flame which
seems to become yellow in the interior and red on the outside. See also
P.38 and Table IX.
Calcium Sullate Dihydrate (4.5-7.5; Mg: 1.3; Sr, Ba: 0) and (3.8-6.8;
Mg: 2; Sr, Ba: 1).
Fe(3), Cr(3), Pb, Sc, lanthanides, and Th interfere with the test.
Place the test droplet upon a slide and evaporate to dryness. After
cooling, dissolve the residue in a volume of 6-F HCI that promises a 0.1
to 1 % solution of calcium. Place 1,tt1 1-F H 2 S0 4 close to the test drop,
and connect the two drops by drawing a narrow channel. CaS0 4 • 2 H 20
separates as fine needles, sheaves of needles, and occasionally in rhomboids
that combine to twins which imitate the outline of the tail of a swallow
(or arrow). If very little calcium is present, the crystals form only on
complete evaporation of the drop.
Other Tests for Calcium. The test with sulfuric acid is entirely satisfactory
for the identification of calcium after separation of the alkaline earth group.
In nature, barium and strontium occur only rarely in high concentration
and will not interfere with the detection of calcium.
Qllite recently, sensitive tests for calcium have been described, which
permit the detection of calcium in presence of most other elements (342,
472). The spot test described by WEISZ (950), based upon involved
reasoning, will certainly serve within the alkaline earth group.

No. 38: Strontium, 87 . 62

Flame Test (6). After lithium and potassium have been vaporized at
low temperature, the purplish red coloration due to strontium may be
perceived without difficulty. See also P. 38 and Table IX.
P.46 Group III B: Scandium Group 331

Potassium-Strontium-Cupric Nitrite (4.0-7.3; Mg: 2). V, Mo, W, Pd,

Au, Pb, Se, Te must be absent; Ba inhibits the test if its quantity is
ten times that of the Sr.
Reagent. Prepare the reagent before use by mixing equal volumes of
aqueous solutions of 50 g KN0 2 per 100 ml and of 45 g sodium acetate
and 10 ml glacial acetic acid per 100 ml. The reagent remains effective
for at least two days; the KN0 2 solution should be renewed every year.
Place a droplet of the test solution upon a slide and evaporate to dryness.
Treat the residue with a volume of 1 % cupric nitrate or acetate solution
that contains about five times as much copper as there is strontium in
the residue. Again evaporate to dryness. After cooling, moisten the
residue or part of it with a very small amount of reagent without touching
the surface of the slide with the tool. Use transmitted light of high intensity
to observe the green squares of the triple nitrite, which separate within
a few minutes (875).
No. 56: Barium, 137.34
Flame Test. See P.38 and Table IX.
Barium Chromate (5.0-8.0; Ca, Sr, Mg: 2). Ag, Hg, Pb, Bi, TI must be
absent.
Place the test droplet upon a slide and evaporate to dryness. Mter
cooling, dissolve the residue in I-F acetic acid to obtain an approximately
1 % solution of barium ion. Push a small kernel of K 2Cr 20 7 into the edge
of the drop. BaCr0 4 separates as very small yellow squares, rectangles,
and rhombs.
Barium Fluosilicate (3.8-6.8; Mg, Ca, Sr: 1) and (3.0-6.0; Mg, Ca, Sr: 2).
Zr, Mo, AI, Mn must be absent.
Place the test droplet upon a slide and evaporate it to dryness. Dissolve
the residue in 2-F HCI to obtain an approximately 1 % solution of barium
ion. Push into the edge of the drop a relatively large kernel of ammonium
fluosilicate. The crystals of BaSiF 6 appear after a few minutes and have
the shape of rather large lentils, spears, and bundles of spears.

P.46 Group III B: Scandium Group

No. 21: Scandium, 44.956
Cochineal Test (4.7-4.2; alkalies, alkaline earths, Be, Y, lanthanides,
Nb, Ta, Cr, Mn, Co, Ni, Zn, Cd, AI, TI, Pb: 1). V, Ti, Zr, uranyl, Cu, Ag,
Au, Hg, Sn, and fluoride must be absent (137).
In a test tube, treat 5 ml of the solution to be investigated with a few
drops of tincture of cochineal. Add drops of 2-F NaOH until a violet
coloration is obtained, warm the solution a little, and then add a drop
or two of glacial acetic acid. Scandium gives a blue coloration or precipi-
tate (878, 911).
332 Confirmatory Tests P.46

No. 39: Yttrium, and the Lanthanides, Nos. 57 to 71

The chemical behavior of yttrium and the lanthanides is sufficiently
different from that of the other elements so that the "rare earths" may
be isolated or recognized as a group. Inside this group, however, the
chemical behavior is sufficiently uniform to make unprofitable the use
of classical analytical procedures. The various forms of spectroscopy are
suited for identification and estimation. Whereas these elements do not
give colored flames, they give characteristic arc and spark spectra. In
addition, absorption spectroscopy may be applied to solutions, and also
the light reflected by minerals, oxide or oxalate mixtures, and colored
salts gives useful spectra. Concerning separation on anion exchange resin
see FARIS and W ARTO"N (490). On the use of the fluorescence of borax
beads, see P.39. Useful contributions of the classical method are the
following group tests.
Test for Lanthanides, yttrium, and 'fhorium. These give a (colorless)
crystalline precipitate when a drop of the test solution (which should be
slightly acid with HCI or HNO a) is treated with a saturated aqueous
solution of oxalic acid (136).
Test for Elements 57 to 62, La to 8m, Inclusive. Transfer a drop of
test solution to a slide and evaporate to dryness. Dissolve the residue
in as little water as possible and add an excess of ammonium succinate
in the form of a few crystals of it. Elements 57 to 62 give fine colorless
needles (136).
Test for Elements 63 to 71, Eu to Lu, Inclusive. Place a drop of the
test solution upon a slide and evaporate just to dryness. Dissolve the
residue in a just sufficient amount of water and add an equal volume of
85% lactic acid of reagent quality. Sc, Y, and elements 63 to 71 give
colorless prismatic crystals (136).
Salts and solutions of the following ions show color: reddish violet,
Nd(3); pink, EU(3), Er(3); reddish brown, Sm(2); orange yellow, Ce(4);
yellow, Sm(3), Ho(3); greenish yellow, Dy(3); green, Pr(3); pale green,
Tm(3) and Yb(2).

No. 58: Cerium, 140.12

Borax Bead. On heating in an oxidizing zone, the bead is orange yellow
while hot and yellow when cold. The color may be seen with 1 part of
cerium in 950 parts of borax when cold and in 4700 parts, when hot (1138).
Cerium Perhydroxide (5.2-6.4).
This test is specific inside the lanthanide group. Strongly colored ions
interfere; the precipitation of ferric hydroxide may be prevented by adding
Rochelle salt (tartrate).
P.47 Group IV B: Titanium Group 333

Mix a drop each of test solution, of 3 % hydrogen peroxide, and of

3-F NH3 in a porcelain crucible and warm gently. A yellow to orange
coloration or precipitate indicates Ce(OH)300H (121, 640, 722).
Test with Anthranilic Acid (4.0-7.3; rare earths and nearly all other
metals: 2) and [3.5-6.8; Fe(3), Cr(3), Cu, Bi, As( 5): 1].
Au, V, chromate, Sn, and chloride must be absent.
The test substance should be dissolved in about 3-F HNO s' If iron
is present, add strong H aP0 4 to suppress the ferric color. On a spot plate,
mix one drop of this solution with one drop of a 5 % solution of anthranilic
acid in ethanol and add some Pb0 2 • Ce causes a bluish black precipitate
to appear, which again quickly dissolves to give a brown solution (136).

No. 63: Europium, 151.96

Cacotheline Test (5.5-5.5). Oxidants like N0 2 -, NO a-, ClOa -, and the
metals Mo, W, U, Ti, V, Nb, Re and Sn(2) must be absent.
In a test tube, treat 1 ml of solution to be investigated with a few drops
of 4-F HCI and a granule of metallic zinc. Then add a few drops of 0.25 %
aqueous solution of cacotheline. Presence of europium is indicated by a
violet coloration (136).

No. 70: Ytterbium, 173.04

Naphthoresorcinol Test (5.3-5.0).
The test is specific inside the rare earth group. The interferences
are about the same as in the cacotheline test above.
Reagents. (a) Add 0.5 cm3 (0.5 g) metallic s,odium to 3 to 4 ml mercury
and stir until solution is complete. (b) Aqueous saturated solution of
oxalic acid. Boil it briefly before use to remove dissolved oxygen. (c) Solid
1,3-dihydroxynaphthalene.
To 1 or 2 ml of the test solution, which should be about I-F with H 2 S0 4 ,
add a small piece of ice and 0.5 ml sodium amalgam. Then add 3 to 4 drops
of boiling hot oxalic acid, about 3 mg naphthoresorcinol, and 3 to 4 drops
6-F HCI. Boil 2 minutes, cool, and then shake with 2 ml diethyl ether.
The latter turns pink if ytterbium is present (137).

P.47 Group IV B: Titanium Group

No. 22: Titanium, 47.90
Bead Test, see P.39 and Table X.
Test with Hydrogen Peroxide (5.0-5.0; Be, La, Ce, Zr, Th, Cr, U, Fe,
AI, Ga, silicate: 2). Nitrate, chloride, bromide, and strongly colored ions
decrease the sensitivity; acetate and formiate must be absent.
Use a solution of the material in 2-F sulfuric acid. Treat 1 ml of the
solution with a drop of 3 % H 20 2 • Titanium gives a yellow orange coloration,
334 Confirmatory Tests P.47'

H 2 • Ti02(S04)2' Molybdenum and vanadium give similar colorations, but

these persist when some solid NH,F is added, whereas the dioxydisullato-
titanic acid is destroyed.
In presence of Mo and (or) V, the proof for titanium is merely the
partial loss of color on adding fluoride. Since also the color of ferric iron
fades upon adding fluoride, it becomes necessary to remove the ferric
color before the fluoride treatment by adding some phosphoric acid.
Peroxide Test after Isolation of Titanium (6.0-6.3; in presence of large'
amounts of Be, La, Ce, Zr, Th, Cr, U, Fe, AI, etc.). Sullate must be absent.
Titanium is separated from other ions by coprecipit'ation with zirconium
arsenate and then tested with hydrogen peroxide.
Dissolve the test material in approximately I-F HCl. In a centrifuge
cone, treat 0.5 ml of this solution with 0.05 ml 20 % arsenic acid and 0.01 ml
1 % ZrOCI 2 in water. Heat in the steam bath, centrifuge, remove the
solution, and wash the precipitate once with I-F HCl containing some
H 3AsO, if V, Mo, or much Fe are present. Mix 2 m12-F H 2S0 4 with 1 drop.
3 % H 20 2 and add 0.2 mi of this mixture to the precipitate. If titanium
has been present, a yellow solution is' obtained, which turns colorless on
adding some solid ammonium fluoride.
Test with Chromotropic Acid (4.7-6.0; in presence of Be, La, Ce, Zr"
Th, Cr, U, Fe, AI, Ga, etc.).
Reagent. Solution of 20 mg 1,8-dioxy-3,6-disullonic acid in 20 ml 18-F
H 2 S0 4 ,
On a spot plate, treat one drop of a solution of the test material in
dilute sulfuric or hydrochloric acid with one drop of the reagent solution.
A blood red coloration indicates Ti .
.
No. 40: Zirconium, 91.22
Rubidium Heptafluorozirconate (3.8-6.8; alkalies, alkaline earths, Be,
V, Cr, W, U, Co, Ni, Pd, Pt, Rh, Au, Zn, Cd, AI, Tl, As, Sb, Bi: 1.3) and
(2.8-5.8; rare earths, Ti, Th, Nb, Ta, Mo, Mn, Cu, Ag, Hg, Sn, Pb, Se,
Te: 1.3).
The test is best performed upon a slide of plastic. Treat a droplet of
the acid solution with a small grain of NH4F and a like particle of RbCl.
RbsZrF 7 separates in colorless octahedra, and six-sided plates (U8).
Test with Alizarin S (4.5-5.8; Y, lanthanides, V, Nb, Ta, Cr, Mo, W,
Mn, Fe, Co, Ni, platinum metals, Ag, Au, Zn, Cd, Hg, AI, Sn, Pb, As,
Sb, Bi, Se, Te: 2.0) and (3.7-5.0; Be, Sc, Ti, Th, U, Cu: 1.3). Fluorides,
phosphate, and organic hydroxy acids interfere; the interference of sullate
may be eliminated by adding BaCIa. The test material is best dissolved in HCl.
On a spot plate treat one drop of the weakly acidic test solution with
one drop of 1 % aqueous solution of sodium alizarin-3-sulfonate and one
drop of 12-F HCl. Zirconium gives a red precipitate.
P.48 Group VB: Vanadium Group 335

No. 90: Thorium, 232.038

radio-active
Thallium Salt of the Carbonate Complex (4.0-7.0; Na, K, Cr, W, Pd,
Pt: 1.5) and (3.4-6.4; Li, alk. earths, Y, lanthanides, Zr, Nb, Ta, Mo,
Mn, Co, Ni, Cu, Sn, Pb, Te: 1.0). Uranyl must be absent.
Place a drop of the neutral or slightly acid test solution upon a slide
and mix with a small drop of 10% (NH4)2COa solution. Add a small crystal
of TINO a. The colorless crystals of TI6Th(COa)5 are very small and have
the shape of somewhat deformed rectangles, squares, and rhombs with
slightly curved outlines. Some appear black because of their high refractivity.
Some irregular shapes suggest twins.
Thorium Iodate (5.4-5.7; alkalies, alk. earths, Be, Y, lanthanides, V,
Nb, Ta, Cr, Mo, W, U, Mn, Fe, Co, Ni, Pt, Ir, Cu, Zn, Hg-ic, Al, TI, Pb,
Se, Te: 1.3). Ti, Zr, Hg, and Sn give similar precipitates and must be
absent.
In a centrifuge cone, treat 0.5 ml of the slightly acidic test solution with
25,u1 8-F HNO a and 10 to 20,u1 of a saturated aqueous solution of KIO a·
Thorium gives a white precipitate of KIO a . 4 Th(IO a)4 . 18 H 20.

P.48 Group V B: Vanadium Group

No. 23: Vanadium, 50.942
Bead Test. See P.39 and Table X.
Quinolinol Fusion. This will detect I ,ug V in presence of 38000 ,ug MoOs
and 7000,ug WO a (340). See also P. 39.
In a porcelain crucible treat some of the solid material (or residue
of the evaporation of a test drop) with a few milligrams of solid 8-hydroxy-
quinoline. Heat upon the steam bath. A brownish black melt indicates
vanadium.
Test with Benzidine Acetate (5.3-7.3; Mo: 2.0) and (5.0-7.0; W: 3.3).
Chromate and other oxidants must be absent.
Transfer a droplet of the test solution to filter paper and add a drop
of a saturated solution of benzidine in 10% (1.7-F) acetic acid. Pentavalent
vanadium oxidizes to "benzidine blue". With much vanadium, a green
coloration results (136).

No. 41: Niobium, 92.906

Niobium and tantalum have in common that, because of the pronounced
tendency to hydrolyze with the separation of the pentoxides, it is very
difficult to keep their compounds in solution.
Sodium Metaniobate (3.4-6.1). Ta gives the same test; Wand Fe do
not interfere.
336 Confirmatory Tests P.49

Transfer one droplet of the alkaline test solution to a slide; add I drop
of 2-F NaOH and I drop of saturated aqueous solution of sodium acetate.
Colorless monoclinic needles and prisms prove the presence of Nb or Ta
(NaNbO a or NaTaO a).
Zinc and Thiocyanate (4.3-4.3; Ta, Ti, and W do not interfere).
Place a few crystals of KCNS into a small test tube. Add I ml of
the alkaline test solution, a few granules of metallic zinc, and 5 drops
I2-F HCI. Depending upon the amount of niobium, the resulting coloration
deepens from yellow to reddish brown (136). Perform a blank test.

No. 73: Tantalum, 180.948

See also Niobium, above.
Dipotassium Heptafluotantalate (4.0-6.7). Nb and Ti interfere and must
be absent.
Transfer a drop of the alkaline test solution to a slide and add a drop
of saturated aqueous solution of KF. The crystals of K 2TaF 7 or K~bF 7
appear after about ten minutes: colorless orthorhombic needles which
are weakly birefringent.
Test with Tetraethylrhodamine (4.6-4.6; Nb, Ti: 2.5). Mo, W, Fe, Au,
Hg, and Sb interfere and must be absent.
In a test tube, treat I ml of the test solution which contains the tantalum
as fluoride (TaF5 or K 2TaF 7) with 50 mg solid rhodamine B (tetraethyl-
rhodamine). Tantalum gives a violet coloration or precipitate; a blank
test must be carried out.

P.49 Group VI B: Chromium Group

No. 24: Chromium, 51.996
Bead Test. L. I. = 2 p,g Cr. See P. 39 and Table X.
Silver Dichromate. L. I. = 0.06 p,g Cr0 4 • All anions interfere, that
give acid insoluble silver salts.
The test is an inversion of the silver test of Expt. 25. Add a small grain
of solid silver nitrate to the test solution which should be about 3-F in
HNO a• A suitable test solution of chromate is obtained by dissolving
an evaporation residue in HNO a of the given concentration.
Perchromic Acid (approx. 4.0-4.0; V: 0.7; As: 1.0; Ti, Fe-ous, Ir,
Tl-ic, Bi, Te: 1.2; and in presence of nearly all other metallic ions: 2.0).
Place into a small test tube I drop 3 % H 20 2 , I drop 2-F HCl, a few
drops of diethyl ether, and finally 1 ml of the slightly acidic solution to
be tested for chromate ion. Shake the mixture briefly and cool by running
tap water over the outside of the test tube. The blue coloration given
by Cr0 6 = is not stable and disappears soon. With much chromate, the
P.49 Group VI B: Ohromium Group 337

whole ether turns blue; with small amounts of chromate, the blue coloration
appears only at the interface of the liquids.
Test with Strychnine (6.0-7.3; most metal ions including Mo, W, Au,
Rh, Pd, Ir: 2.0), (5.5-6.8; Th, U, Fe, 00, Pb, Se, Te: 1.5); stannous tin,
ceric ion, and other strong oxidants must be absent.
Also this test, as all others, applies to chromate ion.
On a spot plate, treat one drop of the solution of the test substance
in water or dilute H 2 S0 4 with 1 drop of a 1 % solution of strychnine in
18-F H 2 S0 4 , Ohromate gives a pink coloration which may require 15 minutes
to appear if only little chromate is present.

No. 42: Molybdenum, 95.94

Flame Test. See P.38 and Table IX.
Borax Bead. See P. 39 and Table X. One part of MoOs may be recognized
III 60 parts of borax.

Thallous Molybdate and Wolfram ate (5.5-8.5; As, Sb, Sn, selenite,
tellurite, and tellurate do not interfere).
On a slide, treat one drop of definitely alkaline molybdate solution
(2-F in NaOH) (or tungstate solution) with a small crystal of TINO s'
The colorless or pale yellow crystals of Tl 2Mo0 4 or Tl 2W0 4 form regular
hexagons, six-pointed stars, skeletons, or rosettes. Dendrites may first
grow out of the grain of reagent.
Test with Thiocyanate and Stannous Chloride (5.7-7.2; in presence of
arsenite, phosphate, and antimony) and (4.3-5.8; wolframate: 2.0; also
stannate and germanate may be present). Selenite, tellurite, and tellurate
must be absent.
The test may be carried out in the centrifuge cone, on a spot plate,
on spot test paper, etc. If necessary, the test solution is evaporated with
HOI for the removal of nitrate. The residue is moistened with 12-F NHa
which is then allowed to evaporate. The residue from the ammonia treat-
ment is dissolved in 3-F HOI to give an about 0.1 % Mo solution.
To perform the test, first add to the acid test solution a small volume
of 10% (I-F) solution of KONS. A red coloration is obtained in presence
of ferric ion. Molybdate may give a yellow coloration. Now add a 5 %
solution of SnOl 2 in 3-F HOI. The red coloration given by ferric thiocyanate
disappears, and any red color persisting or appearing is now due to the
complex Mo(ONS)6=-'

No. 74: Wolfram (Tungsten), 183.85

Bead Test. On the borax bead see Table X; concerning the differentiation
between Wand Ti by means of the phosphate bead, P.39.
Thallous Wolframate, see Molybdenum.
Benedetti-Pichler, Identification 22
338 Confirmatory Tests P.50

Test with Stannous Chloride (4.0-5.7). Mo, Nb, Ta do not interfere.

A wolframate solution is used for the test.
Place a large drop of 12-F HCI upon spot test paper and, without delay,
add the test drop to the center of the wet spot. A yellow fleck of W0 3
may become visible. Now add 1 drop of a 25% solution of SnCl 2 in about
3-F HCI and 1 drop of 10% (I-F) KCNS. A blue spot in the center indicates
W, and a red ring around it may suggest Mo.

No. 92: Uranium, 238.03

Bead Tests. See P. 39 and Table X. Uranium in a bead of NaF gives
a violet fluorescence with ultraviolet light, which permits detection of
0.02 ",g U; by using a spectrograph, this limit of identification may be
improved to 1 ng U (119).
Sodium-Zinc-Uranyl Acetate. L.1. = 0.6 ",g U. The test is specific
for uranium.
Transfer a droplet of test solution to a slide and evaporate to dryness.
Dissolve the residue in a volume of 3-F acetic acid, that promises a 1 %
uranium solution. Place into the edge of the drop first a grain of sodium
acetate and then, at a point not too far away, a grain of zinc acetate.
Search under the microscope for tetrahedra (triangles) of the sodium-
uranyl acetate and for octahedra-like birefringent crystals of the triple salt.
Oompare under Lithium and Sodium in P.44. Observe the fluorescence
of the crystals in ultraviolet light.
Quinolinol Fusion (453). Probably quite specific.
Treat a residue of uranyl salt upon the slide with sufficient solid
8-hydroxyquinoline so that the space between the slide and a fragment
of a cover slip will just fill with the melt. Place the piece of cover glass
upon the reagent and heat upon the steam bath. When the quinolinol
melts, at about 80° C, the uranyl salt separates in rods and willow leaf-
shaped crystals having pleochroism (yellow-red). At about 120 C, these
0

crystals dissolve in the melt, and upon cooling separate rectangular crystals
having an oblique extinction of 8 degrees (both indices of refraction are
higher than that of the melt at 35° C).

P.oo Group VII B: Manganese Group

No. 25: Manganese, 54.9380
Flame Test. Only MnCl 2 gives a very fleeting green coloration to the
flame, see Table IX.
Borax Bead. L.1. = 20 ",g Mn. See P.39 and Table X.
Chlorate Fusion. L. 1. = 0.1 ",g in presence of 3 ",g Cu, Co, or Ni;
10 ",g Fe, Cr, U; or 100 ",g alk. earths, Be, Y, lanthanides, Ti, Mo, W,
Re, Cd, Pb, As, Se.
P.I)O Group VII B: Manganese Group 339

On a platinum wire or foil-or in a platinum spoon or crucible-and in

a non-reducing flame, fuse the solid particle to be tested (or the test drop)
with a small amount of a solid mixture of 1 part (volume or weight) NaCIO s
with 5 parts Na 2CO a• On cooling, the bead or melt will show the green
color of manganate. Dissolve in a small volume of 2-F acetic acid to obtain
the pink to purple permanganate coloration.
If the bead is dissolved upon the slide in I-F HCIO" adding solid KCI
or RbCI into the edge of the drop will give mixed crystals of permanganate
and perchlorate, see below.
Thermoluminescence Test. L. I. = 1 ng Mn; see Expt. 60.
PersuHate Oxydation (5.7-7.0; Cr: 2.5; Ce: 3.0). The test works in
presence of most other metals without impairment of the sensitivity.
Place the test solution upon a slide and evaporate to dryness. Three
times treat the residue with 3-F HNO s followed by evaporation to dryness
(for the removal of chloride). Finally dissolve the residue in I-F HN0 3
and transfer the solution to a centrifuge cone. Add a very small volume
of 1 % AgNOa and a small crystal of KzSzOs. Mix and heat to about 60° C.
The pink or purple color of the permanganate proves the presence of
manganese. It may be necessary to centrifuge the mixture and to inspect
the clear solution in the coloriscopic capillary, Expt.55. The solution
may finally be used for the rubidium perchlorate test. With the red
crystals as well as with the coloriscopic capillary, the identification may
be improved by the use of a spectroscopic attachment to the eyepiece
of the microscope.
Rubidium Perchlorate-Permanganate (5-7.7; for most metals including
Re: 3.0).
Transfer the solution of permanganate to a slide and push into the
edge of the drop-at two places somewhat distant from one another-, a
crystal of NaCI0 4 and a crystal of RbN0 3 or RbCl. Prismatic mixed
crystals of Rb(CIO" MnO,) are obtained, and the color varies from colorless
to pink, red, and black depending upon the ratio of permanganate to
perchlorate.
Manganese Cyanurate (4.5-7.8; alkaline earths, Cr, V, Fe, Co, Zn,
chloride: 2.0; Be, Y, lanthanides, Ti, Mo, Re, Ni, AI: 2.5). Magnesium should
be absent.
Place a droplet of the solution of manganous salt upon a slide and
evaporate to dryness. After cooling, treat the residue with a droplet of
a saturated solution of cyanuric acid, (HNCO)s' 2 H 20, in I-F NH 3 •
Colorless needles and pointed rods indicate manganese. If only little
manganese is present, hasten crystallization by covering the test drop
with a small watch glass to prevent evaporation and warming slightly.
22·
340 Confirmatory Tests P.51

No. 75: Rhenium, 186.2

Flame Test. Greenish white flame.
Acridine Perrhenate (4.7-8.0; molybdate, wolframate: 2.0; alkaline
earths, Be, Y, lanthanides, Cr-ic, Mn-ous, Fe, Co, Ni, Cu, Zn, AI, Sn: 2.7).
Chromate, permanganate, and uranyl must be absent.
Transfer the neutral or slightly acidic perrhenate solution to a slide.
Place upon the steam bath and evaporate the solution to dryness or nearly
to dryness. Remove from the steam bath and add to the warm residue
a small drop of a saturated aqueous solution of acridine hydrochloride.
Perrhenate gives yellow pointed needles and loose sheaves of such (1202).
Permanganate, chromate, and uranyl give similar crystallizations.
Reduction of Tellurate (6.3-7.6; alkaline earths, Be, Y, lanthanides,
V, Nb, Ta, Ti, Zr, Th, Cr, U, Mn, Fe, Co, Ni, Ag, Zn, Cd, AI, TI, Sn, Pb,
As, Sb: 2.0), (4.0-5.3; tungstate, Se: 1.5; Pt, Ir, Cu: 2.2). Titanium,
molybdate, Rh, Pd, Ag, Au, and Hg must be absent since they also give
dark precipitates.
Perrhenate is reduced by SnCl 2 to Rh(3) which, in turn reduces tellurate
to elemental tellurium.
On a spot plate, treat a drop of the slightly acidic test solution with
1 drop of 1 % aqueous solution of Na 2Te0 4 and 1 drop of 50 % SnC1 2 in
10-F HCI. Perrhenate causes precipitation of black Te.

P.ol Group VIII: Fe-Ni Triad

No. 26: Iron, 55.847
Bead Test. See P. 39 and Table x.
Quinolinol Fusion. L. 1. = 40 ng Fe.
Perform the test as directed under Vanadium, P. 48. Iron gives a
greenish black coloration.
Prussian Blue (4.5-5.8; alkaline earths, Be, Y, lanthanides, Nb, Ta,
Pd, Au, AI, As, Sb, Te: 2.0), (4.0-5.3; Cr, Th, Mn, Ni, Pt, Ir, Rh, Ag,
Zn, Cd, Bi, Pb, Se: 1.5; Ti, Zr, V, W, Co, Hg, TI, Sb-5: 1.3), and (2.8-4.1;
molybdate: 0.7; uranyl: 0.3). Fluoride, phosphate, and aliphatic hydroxy
acids must be absent since they hold the iron in stable complexes.
If not certain that the iron is present in the ferric state, expose the
test drop to bromine vapors and then evaporate just to dryness. Dissolve
the residue in 0.2-F HCI and treat a drop of the solution upon the spot
plate with 1 drop of 10% (2-F) potassium ferrocyanide solution. Ferric
iron gives a dark blue precipitate or a light blue coloration with little iron.
A blank is essential if only a coloration is obtained.
The sensitivity may be improved by using small drops, performing
the test on paper or on a textile fiber impregnated with K4Fe(CN)6 or
K2ZnFe(CN)6.
P.1)1 Group VIII: Fe-Ni Triad 341

To discover about 2 pg of iron, place a small droplet of the test solution

upon a slide and evaporate just to dryness. Allow to cool and then moisten
the residue with a small volume of a mixture of equal parts of 0.5% (O.l-F)
potassium ferro cyanide solution and 1-F HCl, which has been centrifuged
just before use to remove any blue sediment. The "amorphous" blue
precipitate should be observed with strong transmitted light and with
reflected light before a white and before an orange background.
Insoluble iron compounds may be dissolved in a tiny bead of K 2S 20 7
or of borax, which is then placed upon the slide and treated with the
above acid reagent solution. Separation of flocks of Prussian blue may
then be observed during the slow dissolution of the bead.
Ferric Thiocyanate (4.0-7.0). Fluoride, phosphate, arsenate, iodide,
iodate, acetate, and complex forming organic oxy acids must be absent.
Place the test droplet upon a slide, treat with a small amount of a
mixture of 1 volume 16-F HNO s with 8 volumes 12-FHCI, and evaporate
just to dryness. Dissolve the residue in a small volume of a mixture of
equal parts of 1-F KCNS and 6-F HCI. If necessary use the coloriscopic
capillary for the observation of the red color, Expt. 55. The red color is
discharged upon adding stannous chloride, see Titanium, P. 47.
o-Phenanthroline Test for Ferrous Iron (6.5-8.8; alkaline earths, Be,
Sc, Y, lanthanides, Ti, Zr, Th, V, Nb, Cr, Mo, W, U, Mn, Ni, Ru, Rh,
Pd, Os, Pt, Ag, Au, Zn, Cd, Hg, AI, Ga, In, TI, Sn, Pb, As, Sb, Bi, Se,
Te, fluoride, phosphate, aliphatic oxy acids: 3.7), (5.2-6.5; Co, Sb: 2.7),
and (4.2-5.5; Cu: 1. 7).
Ferrous iron gives a red cation, Fe (C12HsN 2)S ++. Of course, the treatment
with hydroxylamine must be omitted when testing for ferrous iron
specifically. Additional precautions would be necessary.
Evaporate the test solution just to dryness, and dissolve the residue
in a small volume of O.l-F HCI. On a spot plate, treat with a kernel of
NH 20H . HCI. Hasten the dissolution by stirring, and then treat with
a like volume of 0.025-F aqueous solution of o-phenanthroline. If necessary,
observe the red coloration in a coloriscopic capillary, Expt. 55.

No. 27: Cobalt, 58.9332

Borax Bead. The blue coloration may be recognized with O.~ p,g Co,
see Expt. 59, P.39, and Table X.
Cobalt-Mercuric Thiocyanate (6.0-10.0; Be, Y, lanthanides, Ti, Mn, Re,
Zn: 2.3; alkaline earths, Zr, V, Cr, Mo, W, U, Fe, Ni, Rh, Os, Ir, Pt, Cu,
Ag, Au, Zn, Cd, Hg, Sn, Pb, As, Sb, Bi, Se, Te: 2.0).
For the performance compare Expt. 37. To obtain highest sensitivity
for cobalt, proceed as follows. Place a droplet of test solution upon a slide
and evaporate to dryness. (At this time, zinc may be added to increase
the sensitivity. If this seems desirable, add a droplet of 1 % zinc solution
342 Confirmatory Tests P.ol
to the residue and again evaporate.) Allow to cool to room temperature,
then add as little reagent as possible with the use of a loop or capillary
pipet. Be certain not to touch the surface of the slide with the tool since
this may cause seeding and the separation of many tiny crystals tracing
the scratches. Observe the blue needles and clusters of spear-shaped
crystals with strong transmitted light. The crystals of CoHg(CNS)4 are
relatively thick and dark blue, but if zinc has been added, they will be
lighter in color. Ferric ion gives a dark red solution which may be decolorized
by adding NH4F.
Cobalt Rubeanate (6.0-7.5; Cr, Re, Zn, AI: 2.0), (5.0-6.5; Mn, Fe-ic: 2.0).
Ag, Ni, and Cu interfere.
Concerning the performance see Expt. 27.

No. 28: Nickel, 58.71

Bead Test. See P. 39 and Table X.
Molybdate Test (3.7-7.0; Co and perrhenate do not interfere). AI, TI,
Ce, Fe-ous, and Mn should be absent. The interference by ferric ion may
be prevented by adding NaF.
Place the droplet of test solution upon a slide and evaporate to dryness.
After cooling, dissolve the residue in a small volume of O.I-F HCI and add
a grain of NaF if iron is present. Next to the test drop, deposit a drop of
a saturated aqueous solution of (NH4)6M07024' 4 H 20 and draw a channel
to connect the drops. Separation of colorless squares, rhombs, and wedges
indicates nickel. Aluminum and ferrous iron give crystals of the same shape.
Nickel Dimetbylglyoxime (5.0-8.5; Co: 1.7) and 0.3 ng Ni may be
detected with darkfield illumination when the test is performed upon a
slide. The test is quite specific.
Place the droplet of test solution upon a slide and evaporate to dryness.
After cooling, dissolve the residue in 0.3 to 3,tt1 6-F NH a. Without delay,
add a small crystal of dimethylglyoxime and cover with a small (I-cm
diameter) watch glass to prevent evaporation of the ammonia. After five
minutes inspect under the microscope. The nickel salt crystallizes as long,
fine, red needles which show pleochroism. They appear red when they
are parallel to the vibration direction of the polar (nicol). The far more
soluble copper salt of similar color appears red when the vibration direction
of the polar is vertical to the long edges of the prismatic crystals.
If the test is carried out on the spot plate or on spot paper, the sensitiv-
ities are specified as follows (5.5-6.8; Mn, perrhenate, chromate, Zn,
Al: 3.0), (4.5-5.8; Fe: 3.3; Co, Pd-ous: 2.0), and (3.5-4.8; Cu: 1.0).
For reagents serve a saturated solution of dimethylglyoxime in 95 %
ethanol and 6-F NH a.
P.52 Group VIn: Ru-Pd Triad 343

Nickel Rubeanate (6.0-9.0; alkaline earths, V, Cr, Mo, W, Pt, Au, Zn,
Cd, AI, TI, Sn, Pb, As, Sb, Bi, Se, Te, perrhenate: 2.0), (5.0-8.0; ferric
ion: 2.0). Cu, Ag, and mercurous ion must be absent.
For the performance see Expt. 27.

P.52 Group VIII: Ru-Pd Triad

Catalytic Glow Test for Platinum Metals (697, 1003). The test detects
all platinum metals with the exception of ruthenium and osmium. When
using test drops of 1 ,Ltl volume, the limits of identification are: 20 ng Rh,
10 ng Pd, 180 ng Ir, and 40 ng Pt. The test is quite indifferent to the
presence of other elements, and Mo, U, Fe, Co, and Cu may be present
in thousandfold excess. For As 2 0 a, which poisons the catalysts, the ratio
may be 50 to 1 Pt. The procedure for the performance of the test has been
recommended by HAHN (856). As a possible alternative, one might take
up the test solution in the tip of a small asbestos triangle cut to give a
fine point (881).
Cut asbestos paper of 0.5-mm thickness to a strip of 2 cm X 6 cm.
Moisten one end of the paper and and make there a depression with the
rounded end of a glass rod and the paper resting upon the flat surface of
a cork having a pit that serves for matrix. Finally form at the lowest
point of the cup a sharp conical depression by pressing with the other end
of the glass rod, which is drawn out to a sharp point. The paper may be
slightly pierced by this operation. Dry and then strongly ignite the shaped
end of the strip by heating in the reducing zone of a Bunsen flame.
With a capillary pipet transfer the test solution to the point of the
tiny cone. Ignite again and then, before the paper gets quite cold, expose
it to a stream of pure and dry hydrogen gas which escapes from a capillary
tube. The gas should be supplied with low pressure (20 cm water column)
so that it gives at the opening of the capillary a flame of 5- to 10-mm height.
Extinguish the flame, and hold the paper so that the gas blows into the
cup shaped depression. The point of the conical tip will start glowing.
Find the position which gives the strongest incandescence by varying
the distance of the paper from the orifice of the capillary.

No. 44: Ruthenium, 101.07

Borax Bead. Black. See P. 39.
Test with Thiourea (136) (5.0-6.3). Osmium gives a red coloration
and consequently interferes with the test. A solution of the test material
in hydrochloric acid is preferred.
Place 1 drop lO-F HCI into a small test tube. Add a drop of the solution
to be tested and a few small grains of thiourea. Warm gently. A blue
coloration shows the presence of ruthenium.
344 Confirmatory Tests P.52

Test with Rubeanic Acid (5.0-6.0). Co, Ni, and Cu supposedly interfere,
but Os does not. Ruthenium shall be present in the trivalent state.
In a porcelain crucible, treat a drop of the test solution with 1 drop
2-F HCI and 1 drop of 0.2 % rubeanic acid in glacial acetic acid. Heat
with a microflame. A blue coloration is caused by ruthenium. Platinum
and palladium give red precipitates; to render the blue coloration visible,
transfer the mixture to a microcone, whirl in the centrifuge, and take
the solution into a coloriscopic capillary, Expt. 55, if necessary.

No. 45: Rhodium, 102.905

Borax Bead: brown, slate gray in reflected light. See P. 39.
Potassium Pyrosulfate (162). A trace of metallic rhodium heated with
K 2 S 20 7 on platinum foil or wire gives a melt which is red when hot and
yellow when cold.
Tests with Aqueous Solutions of Rhodium. For the following tests,
KONIG and CROWELL (883) treat the residue to be tested with oxidant
and potassium salt to obtain the complex RhCI 6 =-. Potassium rhodonitrite
is repeatedly evaporated at 125 0 C with aqua regia, and this treatment
is followed by evaporation with 1 drop HCI (12-F) and a crystal KCIO s'
The residue is dissolved in water. An ethylxanthate precipitate is given
just the chlorate treatment described above.
Test with Hexamethylenetetramine (3-5.3; Pd, Ir, Pt, Au: 1.0).
On a slide, treat a droplet of the test solution with a crystal of hexa-
methylenetetramine hydrochloride. After 5 to 10 minutes, bright red
crystals will appear: rods crossed at right angles in presence of sodium,
or thick hexagonal plates in presence of potassium (885). Controls are
recommended.
If iridium is coprecipitated, the color of the crystals changes to brown,
and this may already happen with quite small amounts of iridium
(1 iridium: 50 rhodium).
Stannous Chloride Test (5.0-6.0). All platinum metals as well as Ag,
Hg, and As are reduced to the metallic state, but only Au gives a red
coloration, the purple of CASSroS (883).
In a centrifuge cone, heat 2 drops of the test solution with 1 drop of
a 40 % solution of SnCl 2 in 30 % (9.5-F) HCI for 5 to 10 minutes. A purplish
red coloration which develops on cooling indicates rhodium.

No. 46: Palladium, 105.4

Borax Bead: black, see P. 39.
Test with Dimethylglyoxime (4.0-7.0). Only Au and Ni are also
precipitated. The composition of the precipitate is analogous to that of
the nickel dimethylglyoxime.
P.53 Group VIII: Os-Pt Triad 345

Place a drop of the weakly acidic test solution upon a slide and introduce
a crystal of dimethylglyoxime. Yellow pointed needles, sheaves and stars
of such, separate in a short time.
Test with Thioglycolic Acid (5.0-6.3; Ir, Pt, Au: 2.4; Rh: 1.4). Ru, Os,
Cu, Ag, nitrite, and cyanide interfere. Simple as well as complex ions of
bivalent and tetravalent palladium give the test.
Transfer a drop of the test solution to a white spot plate and add a
drop of 10% aqueous solution of thioglycolic acid or of a correspondingly
smaller volume of the commercially available solution of 80 %. A yellow
color develops immediately if palladium is present. High concentrations
of palladium give a yellow precipitate that dissolves upon diluting to
give the yellow coloration. Similarly, also Au, Pt, Ir (present in high
concentration) give white amorphous precipitates; these also dissolve upon
diluting with water but give colorless solutions (884).

P.53 Group VIII: Os-Pt Triad

No. 76: Osmium, 190.2
Borax Bead: brown and clear, see P. 39.
Potassium Nitrate Bead (119).
The bead is first completely black, but becomes brown on further
heating and finally colorless since the metal is volatilized as OsO" the
chlorine-like odor of which may be perceived with 20 ng of the compound
per milliliter air. OS04 is known in two modifications: one yellow, m. pt.
41 0 C; the other white, m. pt. 39.5 0 C. The boiling point is 134 0 C. The
solution in water, "osmic acid" or "perosmic acid", and the vapor are
very poisonous and injurious to the eyes and mucuous membranes.
Potassium Osmate.
By using the volatility of OS04' the test like any other osmium test
may be made specific.
Treat the solid test substance with some 8-F HN0 3 or treat a test
solution with an equal volume of 16-F HNO a. Heat the mixture gently
in any suitable apparatus that permits absorbing the liberated OS04 in
2-F KOH. Apparatus shown in Figs. 75b, 15a, 18c, and 69 may be used.
The KOH will become yellow to light red as it absorbs increasing amounts
of OS04' and the resulting solution is used for confirmatory tests; RuO,
is not liberated under the conditions.
Transfer a small volume of the distillate to a slide and introduce a
small grain of KN0 2 to aid in the reduction to OS02' Garnet red, small
but well developed, strongly birefringent (orthorhombic) "octahedra" of
K 20sO, . 2 H 20 separate (865).
346 Confirmatory Tests P.53

Osmium Thioureacomplex (5.0-6.0) (136).

In a small test tube, treat 0.1 ml of the test solution, which may contain
osmium of the valence + 8 or + 4 and which should be 1.5-F with HCI,
with a few grains of thiourea. Warm the mixture. A red coloration
indicates osmium.
Test with Thiocyanate (6.0-7.0). Au, Ir, Pt, Fe, Co, and various other
elements interfere. The Os shall be in the valence state + 8 (162).
In a small test tube, treat 0.1 ml of the solution, which shall be lightly
acidified with HNO a, with a few grains of KCNS or NH 4CNS. Osmium
gives a blue coloration which may be extracted with ether or amyl
alcohol.
No. 77: Iridium, 192.2
Borax Bead: brown and clear, see P. 39.
Potassium Chioriridate, see Platinum. The hexamethylenetetramine
tests used by KONIG and CROWELL (883, 885) deserve consideration.
Test with Sulfuric Acid and Nitrate (3.9-5.2). Appreciable amounts
of Ru, Rh, or Pd interfere with the test (162).
In a porcelain crucible, treat 1 drop of the test solution with 2 to 3 drops
18-F H 2S0 4 and heat to drive out HCI. Then add a few crystal grains of
AgNO a and heat again. It is assumed that the blue coloration is due to
some hydrate of Ir0 2.

No. 78: Platinum, 195.09

Borax Bead: brown and turbid, L.1. = 0.2,Ltg. See P.39.
Alkali Chloroplatinate (For K 2PtC16 : 4.0-7.5). Os and Ir interfere.
A suitable test solution is obtained by evaporating with aqua regia and
dissolving the residue in 2-F HCI. This procedure will also eliminate
osmium.
Transfer a small droplet of the test solution to a slide and add a tiny
crystal of KCI. Yellow isotropic octahedra of K 2PtCl 6 prove the presence of
platinum. The corresponding iridium compound, K 2IrCI 6 , is red. Solutions
containing both, Pt and Ir, give mixed crystals, the color of which depends
upon the ratio Pt: Ir.
The sensitivity of the test may be improved by substituting Rb, Cs,
or TI+ for the potassium. The solubility of the salts decreases in the given
order.
Test with Rubeanic Acid (4.0-5.3). Ru, Pd, Au, and Bi interfere.
On a spot plate, treat 1 drop of the solution of the test substance in
hydrochloric acid (see K 2PtCl 6 test above) with 1 drop of 0.02% rubeanic
acid in glacial acetic acid. The precipitate of Pt(C2HaN2S2)2 is reddish
violet.
P.54 Group I B: Copper Group 347

P.54 Group I B: Copper Group

No. 29: Copper, 63.54
Flame Test. L. I. = 0.3/lg Cu. See P. 38 and Table IX and P. 60.
Bead Test. See P. 39 and Table X.
Copper-Mercuric Thiocyanate (5.0-8.0; V, Mo, W, Pt, Ag, Au, Cd, TI,
Pb, As, Sb, Bi, Se, Te: 2.0). Fe-3, Co, and Ni interfere and should not be
present in appreciable amounts.
Proceed as directed in Expt. 37.
Copper-Ammonia Complex (4.0-6.0). Ni, Co, Cr, and other substances
interfere, that give colored ammoniacal solutions.
Proceed as directed in Expt.55.
Cupric Rubeanate (6.0-8.0; V, Mo, W, Pt, Au, Cd, TI, Sn, Pb, As, Sb,
Bi, Se, Te: 2.0). Ag, Hg, and in some measure Co and Ni interfere.
Proceed as directed in Expt. 27.

No. 47: Silver, 107.870

Borax Bead: yellow; L.1. = 0.2/lg Ag. See P. 39.
Silver Dichromate (3.2-6.7). Apparently no interferences that have
practical consequences.
Proceed as directed in Expt.31.
Silver Chloride (7.0-9.0). Mercurous ion interferes with the test but
may be converted to mercuric ion. Aside from this, the test is specific
when performed so that a small amount of hydrochloric acid is added to
an already acid solution. Neither PbCl 2 (orthorhombic needles) nor TICI
(cubic) should separate under these conditions.
Test tube, centrifuge cone, microcone, capillary, capillary cone may be
used, or the droplet may be spread between slide and fragment of cover
slip or suspended in oil. For the detection of small amounts, it is essential
to inspect the solution before and after adding the HCI with strong lateral
illumination before a black background (Tyndall illumination or darkfield
condenser) .
Depending upon circumstances, dilute the test solution with water or
acidify it ,with HNO a so that it will be about 2-F with free acid. Inspect
the solution to make certain that it is clear. If necessary, centrifuge and
use the clear solution after careful inspection for absence of turbidity.
Treat the test solution with an amount of I-F HCI, that will bring the
chloride concentration of the test solution to about 0.001 to 0.01 formal.
A white precipitate proves the presence of silver. Small amounts of silver
will give only a turbidity which is made visible by suitable illumination:
strong light from the side and black background or use of a darkfield
condenser.
348 Confirmatory Tests P.54

For recrystallization from NH 3, see Expt. 32, the precipitation of the

AgCI is made complete by adding more HCI, if necessary, and heating on
the steam bath for 5 to 15 minutes to hasten flocculation. The precipitate
is collected and washed once with O.I-F HN0 3. Without the aid of micro-
manipulation, about 0.1 flg AgCI may be recrystallized with success. The
recrystallization from NH3 establishes the identity of the AgCl.

No. 79: Gold, 196.967

Borax Bead: L. I.: 25 ng; ruby red by colloidal gold, see P. 39.
Metallic Gold (4.7-6; Cu: 1.3 if the bead of gold is taken for critel'ion)
or (6.0-7.3; Cu: 2.0 if the pink line of colloidal gold serves this purpose) ..
The test is based upon the experience that all compounds of gold are
reduced to the metal when heated. Aside from Rh, Pd, Ir, and Pt, no
other metal can interfere, and the anion to which the gold is bound is
without consequence.
DUVAL and FAUCONNIER (912) give the following directions and suggest
that the experimenter should not wear any objects of gold. From soft
glass or Pyrex glass tubing, which has been properly cleaned, prepare
a medicine dropper with a rather thick-walled capillary tip. Attach the
rubber bulb and take about 0.05 ml of the sample solution into the tip.
Try to do this so that the solution remains with the outer meniscus at the
orifice of the tip. Holding the dropper horizontal and squeezing the bulb
to hold the drop continuously at the opening, cause it to evaporate very
slowly without spattering by warming cautiously over a microflame or
under an infrared lamp. Finally, remove the bulb and heat the tip of the
dropper with suitable rotating in the Bunsen flame to obtain a solid sphere
of glass of a few millimeters in diameter. The glass sphere serves for magnifyer
and will show a tiny globule of metallic gold if 1 flg of it or more was present.
With less than this and down to 0.05 flg Au, a pink line of colloidal gold
is visible.
Copper, if present in larger amounts than 18 flg, may give a metallic
bead that could be mistaken for gold. If less copper is present, it will
give green specks in lively contrast with the bead of gold or the red streak
of colloidal gold.
Colloidal Gold Upon a Fiber (5.4-8.7). Platinum metals do not interfere.
Alkalies should be absent, and significant amounts of Ag and Mo interfere.
Reagent. Use clean viscose-rayon fibers or fibers of raw true silk.
Soak them in 10% (2-F) KOH for several hours and then wash them
thoroughly. Dissolve 10 g SnCI 2 • 2 H 20 in 5 ml 12-F HCI; add 95 ml water
and filter. In small portions and with continuous stirring, add to the
filtrate 109 white crystals of pyrogallol. Place the washed fibers into.
this solution and heat 10 minutes upon the steam bath. Wash the fibers
with distilled water and then dry them by pressing between filter paper.
P.55 Group II B: Zinc Group 349

Keep them wrapped in clean filter paper until they are assuredly dry,
and then store in a stoppered amber bottle (better in a dark closet).
Perform the test as described by DONAU (1002). Place the test drop
upon a watch glass and evaporate to dryness on the steam bath. Cover
the residue with 12-F HCI, and again evaporate. After cooling, extract
the residue with 0.5,u1 water, and transfer the extract to a slide by means
of a capillary pipet. Using the technique of Expt.57, insert the end of
one of the prepared fibers into the droplet, and allow the latter to evaporate.
If gold is present, the end of the fiber turns red, violet, or blue-depending
npon the amount of gold.
Molybdates give a blue color without any reddish tinge. Silver imparts
a brown color if present in large amount, but it does not conceal the color
produced by gold unless very little of the latter is present.
Reduction with Stannous Chloride (5.4-5.7) (137).
Reagent. Stir 22.6 g SnCI 2 • 2 H 20 with 2.5 ml 12-F HCI and heat for
two to three minutes. Cool to about 50° C, and then pour the solution
into 400 ml water.
In a small test tube, treat 0.5 ml of the weakly acid test solution with
0.1 ml (3 drops) of the reagent and mix. The reduction to colloidal gold
gives a yellow to yellowish brown coloration.
Pyridine Bromaurate (4.5-7.5; Mo, V, Pt, Cu, Cd, Hg, Sn, Bi, Se,
Te: 2.0; W, As, Sb: 1.0). Mercurous mercury, silver, lead, and thallium
should be absent (136).
Transfer to a narrow slide a drop of test solution that promises to
contain between 0.1 and l,ug Au. Evaporate upon the steam bath so
that the residue is collected in a very small area. After cooling, add to
the residue 0.3 to l,ul of a solution of 1 volume of pyridine in 9 volumes
40% HBr. The thin, almost rectangular plates of the bromaurate appear
yellow, brown, or dull red. They show oblique extinction (angle of 10 degrees)
and strong pleochroism from colorless or yellow to brown when the long
edges are at 80° to the vibration direction of the polar.
Use filter paper or a capillary pipet to remove the mother liquor.
When the preparation has become dry, ignite it over a small Bunsen flame.
After cooling observe the pseudomorphs of gold having the shape of the
bromaurate crystals.

P.oo Group II B: Zinc Group

No. 30: Zinc, 65.37
Metallic Zinc. For the detection of zinc or cadmium in an alloy (150),
place some of the latter into the sealed end of an 8-cm length of narrow
glass tubing (4-mm i. d.) or of a capillary. Heat in the flame of a Bunsen
burner. Cadmium (b. pt. 767° C) and zinc (b. pt. 907° C) will distil and
350 Confirmatory Tests P.55

give a condensate consisting partly of the metal and partly of the oxide
(ZnO, white; and CdO, brown). The tube may be cut apart, and condensate
and residue may be tested separately, which may save tedious separations.
Interferences may be due to Hg (b. pt. 356.6° C) and As (sublimes copiously
at 615° C), but will rarely happen~
Zinc and cadmium compounds may be heated with sodium oxalate
to get reduction to the metals; if the heating is performed in a current
of hydrogen or of inert gas, P. 27, 29, the condensate will consist of metal
only.
Zinc-Mercuric Thiocyanate (5.5-8.8) and (4.0-7.3; alkalies, alkaline
earths, Be, Y, lanthanides, Zr, Th, V, Nb, Ta, Cr, Mo, W, U, Fe, Ni, Rh,
Pd, Ir, Pt, Cd, Hg, AI, Sn, Bi, Se, Te: 2.0; Mn, Co, Ag, Au, TI, Pb, As,
Sb: 1.0).
Proceed as directed in Expt. 37.
The test may be carried out upon a spot plate: Treat the test solution
with some H 2 S0 4 and evaporate to dryness. Dissolve the residue in such
a volume of 0.2-F H 2 S0 4 to obtain an approximately 0.1 % solution.
Transfer 1 drop of this solution to a spot plate and add 1 drop each of
0.1 % cupric sulfate solution and a solution of 2.7 g HgCl 2 and 3 gNH,CNS
in 100 ml water. A chocolate brown or violet black precipitate of mixed
crystals is obtained if zinc is present.
Rinnmann's Green (3.2-6.2). The absence of appreciable amounts of
other metals is required. One part of zinc may be detected in the presence
of 5 parts Cd, 2 parts AI, 1 part Ni or Ti, 0.5 part Co, or 0.1 part Mn.
Aluminum gives a light blue ash; titanium, a dark blue one. Tin and
antimony should be absent since they may give a green ash. The zinc
may be quite simply isolated for the test by electrolytic precipitation
from NaOH solution (1143).
Reagent Paper. Soak "ash-free" filter paper in a solution of 4 g KaCo(CN)6
and 1 g KCIO a in 100 ml water. Dry at room temperature or at 100° C
in the oven.
Procedure. Place a droplet of the test solution upon a watch glass
and evaporate to dryness upon the steam bath. Moisten the residue with
6-F HN0 3 and again evaporate to dryness. Dissolve the residue in such
a volume of 2-F HNO a as promises a 0.1 to 1 % solution of zinc. Take up
some of this solution into a capillary pipet and transfer it to the test paper
by using the technique of Expt. 24. Dry the moist spot by holding the
paper high above a small Bunsen flame. First, a yellow circle will appear
and indicate the circumference of the drop. Continue drying until the
center turns brown. Then grasp the far end of the paper with forceps,
light the paper, and allow the ash to drop into a porcelain crucible or onto
a watch glass held ready for the purpose. If zinc is present, some green
ash (a solid solution of some CoO in ZnO: Rinnmann's green) will be visible
P.56 Group III A: Boron-Thallium Group 351

at the center of the spot where the tip of the pipet touched the paper.
This green spot (with little zinc, just a few green fibers) is usually surrounded
by a circular zone containing very little ash, often only a delicate network
of black fibers connecting to the surrounding sheet of black ash (403).
The test may be carried out in a more sensitive and convincing manner
by first precipitating and inspecting the KZnCo(CN)6 on a microscope
slide (1l43).
No. 48: Cadmium, 112.40
Cadmium Metal. See Zinc, above.
Test with Brucine Acetate and Sodium Bromide (4.0-7.0; Cu, Ag, Zn,
TI, Sn, Pb, As, Sb, Se, Te: 2.0; Mo, Pt, Au: 0.0). Mercury, Bi, V, and W
interfere and should be absent.
Follow the directions given in Expt. 38.
Cadmium-Mercuric Thiocyanate (3.0-6.0; Co, Cu, Zn: 0.0).
Proceed as directed in Expt. 37.
Test with Cadion 3 B (6.0-9.0; Na, K, NH 4 , Zn, Pb: 4.0; Mg, Be, Cr,
Mn, Co, Ni, Cu, AI, As, Sb, Bi: 3.7; and Ca: 3.0).
Proceed- as directed in Expt. 28.

No. 80: Mercury, 200.59

Flame Test: see Table IX.
Metallic Mercury and Conversion to Iodide (3.7-6.7; Ag: 2.0). The
test is specific for mercury.
Proceed as directed in Expt.49.
Cupric-Zinc-Mercuric Thiocyanate (4.3-7.3).
Proceed as directed in Expt. 39.
Mercuric Iodide (3.3-6.3); spot test specific in presence of all common
metals.
Proceed as directed in Expt. 24.

P.1l6 Group III A: Boron-Thallium Group

No.5: Boron, 10.811
The following tests are given by boric acid and borates.
Flame Test: see P.38 and Table IX.
FEIGL and SUTER (620) place a drop of the test solution into a porcelain
crucible, add 1 drop 18-F H 2 S0 4 and 5 drops methanol, warm the mixture,
and light the escaping vapor. The volatile methyl borate gives a green
flame. According to LENHER and WELLS (695), Ba, Cu, TI, MoOs, Te0 2 ,
and H sP0 4 may interfere by giving a similar coloration to the flame.
The interference by substances that do not form volatile ester may
be eliminated as follows. Prepare a miniature gas washing bottle from
352 Confirmatory Tests P.56

a soft-glass test tube, about 16-mm i. d. and cut short to S-cm length.
Provide it with a 2-hole rubber stopper, an inlet tube that goes to the
bottom of the test tube, and a short straight outlet tube ending in a capillary
tip. Introduce into the test tube some of the solid sample, 0.5 ml lS-F
H 2 S0 4 , and 3 ml methanol. Close the apparatus, mix by swirling, and
blow a stream of air through the mixture and into a non-luminous Bunsen
flame. Any methyl borate, B(CH 3)3' carried along with the air .will color
the flame green. It should be possible to detect 20 ",g boric acid or less.
A plug of glass wool may be inserted into the outlet tube.
Isolation of Boric Acid via Methyl Borate. The volatility of methyl
borate may be used to isolate boric acid for increasing the specificity of
the tests performed with the distillate. Suitable apparatus are suggested
by Figs. 75b, 15a, lSc, and 69.
Collect the methyl borate in a drop of water where it will undergo
saponification. Treat the solid sample (solutions may be made alkaline,
if necessary, and evaporated in the "pot" of the still) with like volumes
of first lS-F H 2 S0 4 and then methanol. Then heat the mixture by inserting
the pot into water of SO° C.
Turmeric Test. L. 1. = 0.5 ng B; 1 part of boron may be detected in
presence of 100 parts of Mg, Ca, silicate, and phosphate; or 10 parts of
ferric ion. Ti, Zr, Hf, Nb, Ta, and Mo give similar colorations in acid
solution, but do not give the blue color with NH 3 •
The limit of identification applies to the use of a textile fiber as outlined
in Expt. 57. In most instances, use of turmeric paper will suffice; the
test solution may be taken up with a hook or a loop, exposed to fumes
of HCI to acidify it, and then transferred to the edge or a sharp point
of the paper. For drying, the paper is then placed upon a watch glass
(avoid borosilicate glass) and heated upon a steam bath.
Quinalizarine Test (7 .0-S.4); 200 parts H 2 Ge0 3 to 1 of H 3B0 3 prevent
the test; chromate, nitrate, and fluoride interfere to some extent.
Treat 2 volumes of test solution with IS volumes of lS-F H 2 S0 4 , Mix
and cool; then add 1 volume of a solution of 10 mg quinalizarine in 10 ml
water and 90 ml lS-F H 2 S0 4 , Allow to stand for five minutes, and then
compare the blue coloration with the color of a blank test.

No. 13: Aluminum, 26.9815

Potassium-Aluminum Alum (3.0-6.7; Fe-ic and Cr-ic: 1.3). Ferric iron,
chromic ion, gallium, and thallic ion also give alums; Ba, Sr, W, Hg,
and Te should be absent.
Sedel Oe'Jium Ohloride. In a mortar, grind 10 mg KAI(S04)2 . 12 H 20
with 0.1 g CsCI. Then grind 10 mg of the obtained mixture with another
portion of 0.1 g CsCI, and repeat this four more times t.o obtain a cesium
chloride containing 0.0001 % alum.
P.06 Group III A: Boron-Thallium Group 353

Place the test droplet upon a slide and evaporate just to dryness.
Dissolve the residue in a volume of 2-F HNO s which will give an approx-
imately 1 % solution of aluminum. Add a reasonably large crystal of KHSO"
and cover with a small watch glass (lO-mm diameter) to retard evaporation.
Inspect after three minutes. KAl(S04)2' 12 H 20 separates as isotropic
octahedra.
Alum has a strong tendency to form supersaturated solutions. Thus
if crystallization did not start by itself, try to get it going by scratching
with a glass needle. If this does not help, first mix the drop and then add
some seeded cesium chloride. The less soluble cesium alum separates
immediately if some aluminum is present. The crystals are small and less
regularly shaped than those of the potassium alum. If, however, sufficient
time is allowed, they grow to satisfactory size and shape (isotropic octa-
hedra).
To confirm the presence of aluminum, remove the mother liquor with
a capillary pipet, and treat the alum crystals with a large drop of 12-F NHs.
Aluminum alum becomes white and opaque; gallium alum dissolves;
ferric (and indium) alums turn brown and chromic alum, distinctly green.
Duval.
Test with Alizarin S, Sodium AlizarinsuHonate (5.0-8.0; Be, Sc,
lanthanides, Zr, Th, Cr, uranyl, Mn, Fe, Co, Ni, Cu, Zn: 2.0). Ti, Ga, and
Bi interfere and should be absent.
Transfer 1 volume of test solution to spot test paper, add an equal
volume of 0.2 % aqueous solution of alizarin S, and expose the wet spot
to fumes of NHa. When the spot has turned violet, immerse the paper
in 2-F acetic acid. A red or brownish red fleck indicates aluminum.
Performance of a blank test is advisable.

No. 31: Gallium, 69.72

Flame Test. The chloride is quite volatile and gives a violet color to
the flame. The spectral lines are weak. See P. 38 and Table IX.
Hydrous Oxide, Sulfide, Chloride. The commonly met most stable ions,
Ga+++ and In+++, behave very much like the aluminum ion, but there
are differences that ma.ke separations possible. Confirmatory tests may
be ma.de conclusive by combining them with separations.
The hy.:lrous oxides resemble Al(OH)a, but that of gallium is soluble
in ammonia at room temperature. The chlorides of gallium and indium
collect in the ether layer if their solution in 6-F HCI is extracted with
diethyl ether. Aluminum rema.ins in the aqueous layer and precipitates
as AlOIa' 6 H 20 when the mixture of ether and 'aqueous layer is saturated
with gaseous HCl. Sulfides of gallium and indium can be precipitated
from aqueous solution: Ga 2Sa, white, and In 2 Sa, yellow (under special
conditions, a white sulfur compound of In seems to form).
Benedetti-Pichler, Identification 23
354 Confirmatory Tests P.56

Potassium-Gallium Alum (3.3-6.3). Fe-ic, Cr-ic, AI, In, Tl-ic give

similar crystals. Only K2Ga(S04)2' 12 H 2 0 dissolves in NH 3 •
Proceed as directed under Aluminum, above (873).
Gallium Ferrocyanide (Estimate: 4-5). Aluminum and chromic ion
are not precipitated. If performed in a capillary cone, the test will detect
probably less than 1 ng Ga.
Take 3 volumes of solution of the gallium salt (hydrous oxide) in 4-F HCI
and add 1 volume 0.3-F K 4Fe(CN)6 solution. The Ga 4 [Fe(CN)6Js separates
as a white or bluish white precipitate.
Induction of the Manganese Ferricyanide Precipitation (3.9-5.6). The
test may be applied without fear of interference in presence of alkalies,
alkaline earths, Be, Y, lanthanides, Zr, Th, Nb, W, Ni, Rh, Pd, Ir, Pt,
Zn, Cd, Hg-2, Al, In, TI-3, Sn, As, Sb, Bi, and Te. The following, however,
must be absent since they give the same test: Sc, Ce-4, Ti, V, Cr, Mo, U,
Fe, Co, Cu, TI-1, and Se.
Reagent. Treat 120 ml of a 0.5% solution of MnC1 2 • 4 H 20 in 7-F HCl
with 30 ml 0.25-F K4Fe(CN)6 and 1 ml 0.05-F KBr0 3 •
Procedure. Place 1 drop of the reagent upon a spot plate and add a
drop of the solution to be tested. Presence of Ga induces the oxidation
to ferricyanide and subsequent precipitation of reddish brown manganese
ferricyanide (121, 136).

No. 49: Indium, 114.82

Flame Test. Violet coloration of Bunsen flame, see P. 38 and Table IX.
Indium SuHide (4.3-5.6). The test is characteristic when performed
after separation from interfering elements, i. e., all elements that give a
sulfide precipitate in 6-F acetic acid. When performed in a capillary cone,
the test would have a respectable identification limit of 0.5 ng In.
Dissolve the solid test material (hydrous oxide) in 6-F acetic acid
and saturate with hydrogen sulfide. Allow to stand for ten minutes.
The In 2 Sa separates as distinctly yellow flocks.
Test with Hexamethylenetetramine and Ammonium Thiocyanate (3.5-6.5;
alkalies, alkaline earths, Cr, W, Mn, Ni, Cu, Zn, Cd, Al, As: 2.0; Sc, Fe-3,
Sn, Pb: 1.0). Cobalt gives a precipitate of like appearance and must
be absent.
Upon a slide, treat a droplet of the weakly acid test solution with a
crystal of hexamethylenetetramine and one of NH 4CNS. If indium is
present, pink crystals (hexagons, crosses, and rosettes) separate. To start
the crystallization with dilute solutions, it is suggested (136) to rub with
a glass needle after warming the solution somewhat. Finally, it might.
be advisable to confirm the presence of indium further by conversion
to the yellow sulfide.
P.57 Group IV A: Carbon-Lead Group 355

No. 81: Thallium, 204.37

Flame Test: intensely green coloration, see P.38 and Table IX.
Redox Reactions. The salts of the univalent thallous ion resemble partly
those of potassium (or cesium rather) and partly those of lead. Thallous
ion is not oxidized by boiling with nitric acid, but aqua regia, chlorine,
and bromine readily convert it to trivalent thallic ion. Sulfur dioxide or
HI reduce the TI+++ to TI+.
Thallous Chloride (Estimate: 2.7-5.7). The test is specific if performed
as follows.
Transfer a droplet of the test solution to a slide and evaporate just to
dryness. Place upon the residue a droplet of a mixture of I volume 16-F
HNOa and 4 volumes 12-F HCI; again evaporate to dryness. Place a
droplet of O.l-F HCI upon the residue. After a few minutes, transfer
the clear solution to the center of another slide and expose it to an atmos-
phere of sulfur dioxide (which may be obtained by adding 0.5 ml conc. HCl
to a bottle containing about 10 g metabisulfite). Crystals of TICI will
separate as cubes (from dilute solutions), crosses, star-like clusters, and
hexagons. These are radiantly white in reflected light, but nearly black
in transmitted light because of their high refractive index. The crystals
dissolve again if the test drop is exposed to bromine fumes.
Thallous Bromide (Estimate: 4.0-7.0; Ag, Pb: 3.0 or better). The
limit of identification is improved by the addition of KI and may become
of the order of 10 ng. The use of KI, however, introduces the danger of
the precipitation of Cu 2I 2, HgI2' BiIa which may be present in solution
if they are not excluded by a separation preceding the test.
In a centrifuge cone or in a capillary, treat the test solution with bromine
and HBr (or bromide and HCI). Use the centrifuge to remove the precipitate.
Treat the clear solution (on a slide) with S02 or (in a cone) by adding
granules of NaHSOa or small portions of meta-bisulfite, Na 2S 205' The
thallous bromide will separate as a white precipit,ate. Finally, when an
excess of S02 is assured by the odor, add some 1-F KI solution. The white
TlBr is converted to yellow TIl. Furthermore, a precipitate may appear
now in the clear solution with small amounts of thallium which would
not be precipitated by bromide (870).

P.67 Group IV A: Carbon-Lead Group

No.6: Carbon, 12.01115
Elemental Carbon
Complete lack of black particles or spots excludes the presence of
elemental carbon. The black color and glowing upon heating in air suggest
its presence but do not prove it. To obtain proof, the object is best heated
in a current of oxygen which is finally passed through a solution of barium
23·
356 Confirmatory Tests P.67
or calcium hydroxide for the collection of the CO 2 formed, see under
carbonate, below.
Coal, charcoal, soot, and various types of "carbon blacks" burn readily
to leave more or less of light colored ash (inorganic material of various
kind). Graphite, which rarely occurs in black hexagonal plates but rather
forms soft (brown, gray, or black) flakes or glistening and iridescent scales,
burns slow and leaves much or little ash depending upon purity. Diamond
is oxidized very slowly, even at high temperature, and leaves no or very
little ash. See also CHAMOT and MASON (US).
Of course, also organic compounds burn to give CO 2, but they are not
necessarily black, and they usually give also a condensate of water when
burnt in oxygen.
Oarbides
Carbides which are decomposed by water or acid will have been detected
during the preliminary tests by the liberation of methane or acetylene, P. 36.
Refractive carbides, nitrides, and borides are indicated by their hardness,
P.17. They too may bum when strongly heated in oxygen (cubic boron
nitride, borazon, much slower than diamond) and leave an ash for identifica-
tion. In general, it will be best to fuse them with either NaOH or a mixture
of NaOH and Na 20 2; the flux will contain carbonate, and for testing see
below.
Oarbon Monoxide
See P. 37. - Carbon monoxide may be confirmed by first passing the
gas through Ascarite for the absorption of any CO 2, burning the gas in
pure oxygen, and testing for CO 2 and water in the combustion gases.
Only the former shall be formed.
Concerning an extremely sensitive test for the quick detection of CO
in air see SHEPHERD (440).
Oarbonate
Carbonate is generally recognized by the liberation of CO 2 when treated
with a reasonably strong acid. The CO 2 is identified by its action upon
barium or calcium hydroxide, see P. 36.
If lead, barium, strontium, or calcium are expected, one may prefer
to use 4-F HNO a in place of the dilute sulfuric acid for the liberation of
the carbon dioxide. Potassium permanganate may be added to this acid
in order to prevent the liberation of S02 (from sulfites, thiosulfates),
H 2S (from sulfides, thionates), HN3 (from azides), and N0 2 (from nitrites).
Oxydation by permanganate will, however, give CO 2 from cyanides,
formates, and oxalates; it should not be added if these are present.
FEIGL (121) suggests addition of HgCl 2 (binds H 2S and HCN), AgNO s (binds
H 2 S, HN3 and HCN), zirconium salt (binds HF), and (or) hydrogen peroxide
(oxidation of .S02 and nitrite).
P.57 Group IV A: Carbon-Lead Group 357

The testing of the carbonate precipitate for ready solubility in dilute

acid (difference from fluoride) and for absence of sulfite should not be
omitted, if possible, P.36.
Testing the pH of a carbonate solution will suffice to distinguish between
carbonate and bicarbonate.
Formic Acid
Reduction to Formaldehyde and Test with Chromotropic Acid (4.7-4.7).
Hexamethylenetetramine interferes with the test, and glucose gives some
formic acid (121, 137).
In a small test tube, treat I ml of the test solution with I ml 2-F HCI
and a granule of magnesium metal. Wait until the liberation of gas stops
and then pour 0.1 ml of the reaction mixture into another test tube. Treat
this portion with 0.5 ml of a solution of 0.1 g chromotropic acid (naphthalene-
1,S-disulfonic acid) in a mixture of 40 ml water and 60 ml lS-F H 2 S04 •
Place for ten minutes into a water bath of 60° C. If formic acid was present,
the solution assumes a violet color.
Acetic Acid and Acetate
Sodium-Uranyl Acetate (2.0-4.3). Specific test which is, however,
prevented by presence of much ammonium ion, free alkali hydroxide,
or free mineral acids. A satisfactory test solution may be obtained by
evaporating the material to dryness, if necessary after making it alkaline,
distilling the residue with strong H sP0 4, and collecting the vapors in
O.l-F KOH.
Slightly acidify the test solution with formic acid and add a drop of
a 10% solution of uranyl formate in saturated aqueous sodium formate.
The characteristic yellow tetrahedra of the sodium-uranyl acetate are
described under Sodium, P.44.
Conversion to Indigo. L. I. = 60 f-lg CHsCOOH. Copper salts interfere,
and the sensitivity suffers in presence of higher fatty acids. Chromate
and Mn0 2 have no adverse effect.
Mix some solid sample with calcium carbonate, or mix a drop of acid
solution with CaCO s and evaporate to dryness. Then transfer the solid
mix into a glass tube, 5-mm i. d., S cm long, and seal shut at one end.
Place over the opening of the tube a square of filter paper which has been
moistened with a saturated solution of o-nitrobenzaldehyde in 2-F NaOH.
Cautiously heat the lower part of the tube to drive the vapors of acetone
which form toward the moist paper. The acetone reacts to give indigo
which colors the paper blue or green depending upon the amount formed.
If there is a doubt concerning the color, remove the paper from the tube
and moisten it with 3-F HCl. This discharges the original yellow color
of the paper so that the blue of the indigo may show up clearly (121).
358 Confirmatory Tests P.57
It should be possible to perform the test with much smaller quantities
and a fiber or grain of porous tile inside a capillary in place of the paper.
Oxalic Acid and Oxalate
Test with Xanthocobaltic Chloride (2.7-5.7). Fluoride, thiosulfate, and
dithionate give crystals of different appearance; other anions produce
"amorphous" precipitates.
The test solution should be "neutral". On the slide, treat a droplet
of the test solution with a droplet of 3 % aqueous solution of nitropent-
amminocobaltic chloride. Oxalate gives dark yellow prisms, X forms,
and H forms.
Formation of Aniline Blue. L. I. = 5 f-lg (COOH)2' Formic, acetic,
propionic, glycolic, glyoxylic, tartaric, citric, succinic, dihydroxymaleic,
benzoic, and phthalic acid do not give the test.
Use a solid particle, the evaporation residue of the test solution, or
a precipitate obtained with CaCl 2 from neutral or ammoniacal solution,
which may also contain sulfate, sulfite, fluoride, tartrate, etc.
Place the dry solid substance into a small test tube. Add a like volume
of diphenylamine and some 85% (15-F) H 3P0 4 • Heat over a microflame
to obtain conversion to the triphenylmethane dye known as aniline blue.
The mixture may turn blue, but the color fades on cooling. Finally, take
up the cold reaction mixture in ethanol. The dye dissolves to give a brilliant
blue color. FEIGL and FREHDEN (867).
Tartaric Acid and Tartrate
Potassium-Hydrogen Tartrate (Estimate: 3.0-6.0). Boric acid should
not be present in significant amounts.
Reagent. Dissolve 15 g KCI in 100 ml 2-F formic acid and then add
6-F KOH until the pH of the solution is 3. Test with indicator (Hydrion)
paper.
Transfer a droplet of the test solution to a slide. (If it is not approxi-
mately neutral, add first very small amounts of KOH until just alkaline
and then expose to fumes of HCI until acid.) Evaporate just to dryness
by placing the slide upon the steam bath. After cooling, cover the residue
with a droplet of the reagent solution and cover with a l-cm watch glass
to prevent evaporation. If crystallization does not start by itself within
a few minutes,try to start it by rubbing with a glass needle. KOOC·
. CHOH· CHOH· COOH separates in hemihedric forms of the ortho-
rhombic class, long hexagons, pentagons, half hexagons?, prisms or plates
with heavily shaded faces. All crystals should show parallel extinction.
Remove the mother liquor with filter paper or a capillary pipet and
use the crystals for the test with gallic acid.
Heating with SuHuric Acid and Gallic Acid. L. I. = 2 f-lg. Formaldehyde,
glyoxylic acids, glycolic, glyceric, and tartronic acids give colored products
P.57 Group IV A: Carbon-Lead Group 359

and interfere. No colors are given by fatty acids, oxalic, malic, succinic,
cinnamic, citric, and salicylic acids. Since oxidants might interfere with
the test, it is best to first isolate the tartaric acid as potassium-hydrogen
t,artrate or as calcium tartrate.
Dissolve some of the solid sample, of a tartrate precipitate, or of the
evaporation residue from a test solution in 1 droplet to 0.5 ml of a solution
of 10 mg gallic acid, C6H 2 (OH)sCOOH, in 100 ml 18-F H 2 S0 4 , Heat
upon the slide or in a small test tube to 120 to 150 0 C. Depending upon
the amount of tartrate present, a yellowish green to blue color will develop.
EEGRIWE (1156).
Citric Acid and Citrates
Calcium Citrate (Estimate: 4.0-4.0). All anions may interfere, that
give insoluble calcium salts.
To 1 ml test solution add 3 drops of 1 % (O.I-F) CaCI 2 • The solution
will stay clear if it is acid or neutral. Add 1 to 3 drops 2-FNaOH. Aflocculent
precipitate of tertiary calcium citrate may form. This should dissolve
when solid NH 4 CI, in small portions, is dissolved in the reaction mixture.
When the clear solution is boiled, a crystalline calcium citrate should
precipitate (56). It is obvious that the test may be carried out on a smaller scale.
Conversion to Pentabromacetone (4.0-4.0). Acetone and all substances
that furnish acetone may give the test.
Use a solution of the free acid or of any citrate dissolved in I-F sulfuric
or nitric (but not hydrochloric) acid. To 1 ml of the solution add 2 to
3 drops of 0.02-F KMn0 4 and place just the part of the tube filled with
solution into a water bath of not more than 40 C. The citric acid,
0

HOOC· C(OH) : (CH 2 • COOH)2' is oxidized to acetonedicarboxylic acid,

CO: (CH 2 · COOH)2' whicp. gives pentabromacetone, CHBr 2 ' CO· CBrs,
more readily than the acetone itself, that would be obtained by longer
heating or higher temperature. Formation of the acetone would be un-
desirable also because of its volatility. Consequently remove the test
tube from the bath as soon as the reaction mixture shows a turbidity
(Mn0 2 • x H 20) or turns brown. Add 1 or 2 drops of ammonium oxalate
solution and 0.5 ml I-F H 2S0 4 , which will render the mixture clear and
colorless. Now add a few drops of bromine water. If citric acid was present,
white, crystalline pentabromacetone will precipitate. STAHRE (1136).
Hydrocyanic Acid and Cyanide
Silver Cyanide (4.2-7.2). The test, which has been described already
in P. 36, is highly selective. Hydrazoic acid and acetylene interfere, but
these will often be ruled out by the nature of the material under investigation.
To avo~d the slow liberation of HCN by the action of strong acids upon
ferrocyanide, ferricyanide, or other stable cyanide complexes, one may
treat the test material with NaHCO s or KH 2P0 4 solution instead of using
360 Confirmatory Tests P.57

a free acid. Much HCN is produced by the hydrolysis of alkali cyanides

so that a strong test is obtained if the droplet of silver nitrate solution is
exposed to the air from the bottle containing solid alkali cyanide.
According to BRUNSWIK (1086) proceed as follows. Liberate HCN on
the floor of the gas reaction cell and use a drop of 0.5 #1 1 % AgNO a solution
on the cover glass. The separation of a white precipitate is usually seen
with the unaided eye. When this happens, exchange the cover slip for
another one with a fresh drop of AgNO a solution to collect more of .the
AgCN. Inspect the first precipitate. As a rule, well developed crystals
will not be found. Allow the test drop to evaporate. Place on the residue
a cover slip with a drop of 8-F HNO a hanging on the underside, which
is just large enough to barely fill the whole space between slide and cover
slip. Cautiously heat over the microflame until bubbles of steam become
visible between slide and cover slip. Immediately stop the heating and
place the slide upon a cork for slow cooling. Using a magnification of
80 to 100 diameters, and not too strong illumination, search the whole
preparation systematically for the fine needles of AgCN. Be certain to
try the test first with controls.
Mercuric cyanide is not decomposed by adding dilute acid. Thus, in
presence of mercury, the test solution must be subjected to a preliminary
treatment. Add dilute NaOH until just alkaline, then add Na 28 solution
in small increments until the precipitation of black sulfides is complete.
Add Cd80.. until all sulfide ion is precipitated, centrifuge, and use the
clear solution for the test.
Conversion to Thiocyanate (6.0-7.3). The test is also given by mercuric
cyanide, and it may be performed with a precipitate of AgCN obtain~d in
the preceding test. The test is specific for cyanide, but it is assumed that
thiocyanate is not present in the test solution.
Treat the neutral or alkaline sample solution (solid particle or evaporation
residue) with a little yellow ammonium sulfide and evaporate to dryness
upon the steam bath. Extract the residue with a small volume of cold
2-F HCI, remove the clear solution from any insoluble residue, and treat
it with a very small amount of 1-F FeCla• The thiocyanate formed by the
reaction CN- + 8 2 = -+ CN8- + 8= gives the familar blood red coloration
of the ferric thiocyanate complex.
Oyanic Acid and Oyanate
HCNO is an unstable liquid of penetrating, disagreable odor, which
immediately decomposes in aqueous solution:
HCNO +2 H 20 -+ HCO~H2 +H 20 -+ NH,HCO s
The salts are stable, but when their solution in water is acidified,
the acid decomposes as indicated, and CO 2is liberated by the decomposition
of the ammonium bicarbonate.
P.57 Group IV A: Carbon-Lead Group 361

There are no good "tests" for cyanate ion, and its presence is found,
more or less by inference from the facts that BaCl 2 gives no precipitate
with an aqueous solution of cyanate and that AgNOs gives a white, curdy
precipitate which dissolves readily in dilute HNO s with the liberation
of CO 2 , whereafter a test for NH, may be obtained with the acidified solution.

Thiocyanic Acid and Thiocyanates

Thiocyanic acid, HCNS, is a colorless, unstable liquid of penetrating
odor. In aqueous solution, HCNS is reasonably stable and ionized like
a strong acid.
Ferric Thiocyanate (5.0-6.3). Azide must be absent since it gives the
same coloration. Iodide interferes because of the liberation of 12 by the
ferric ion. Sulfide and thiosulfate give a precipitate of sulfur and require
a larger amount of reagent than usually provided. The same holds for
ferrocyanide, ferricyanide, and the anions that give ferric complexes:
phosphate, fluoride, oxalate, and tartrate. The interference of the Prussian
blue precipitate may be overcome by centrifuging.
In a microcone, treat the test solution with an equal volume of 6-F HOI
and then with 1 % (0.1-F) FeCIs. The red color may be observed in the
coloriscopic capillary.
Ferrocyanide
The anhydrous acid is a white crystalline solid, stable in dry air. The
aqueous solution is slightly acidic and decomposes with the separation
of Prussian blue.
Prussian Blue (5.0-6.3; thiocyanate: 3.5; iodide: 3.3).
Use filterpaper impregnated with 1 % (0.1-F) FeCls solution and dried.
Place a drop of the acid (HOI) test solution upon the filter paper. A blue
circle or ring indicates ferro cyanide ; farther out, there may be a red
concentric ring indicating thiocyanate .
. If only a deep red fleck of thiocyanate or a deep brown fleck of 12
is seen, add a drop of saturated thiosulfate solution. The Prussian blue
becomes distinctly visible against the background of the white paper (121).
Ferricyanide
The anhydrous acid is a brown crystalline solid; the aqueous solution
is strongly acid.
In the Prussian blue test (above), ferricyanide gives only a white or
bluish white precipitate; a. blue precipitate is obtained with FeSO,
solution, so-called "Turnbull's blue" which may be identical with the
Prussian blue.
A number of tests based upon the oxidizing action of ferricyanide are
listed by FEIGL (121).
362 Confirmatory Tests P.57

Test with Liel and Hexamethylenetetramine (4.0-7.0). Specific for

ferricyanide, but Ca and Mg should be absent since they give precipitates
with hexamethylenetetramine in presence of ferricyanide and ferrocyanide.
Transfer a droplet of the neutral test solution to the slide and place
next to it a droplet of a solution of 1.5 g hexamethylenetetramine in a 9 %
aqueous solution of lithium chloride (1 g LiCI +
10 ml water). Connect
the drops by drawing a channel with a glass needle. If ferricyanide is
present, yellow octahedra and 4- and 6-pointed stars form where the two
solutions merge. Ferrocyanide gives colorless crystals of different shape.

No. 14: Silicon, 28.086

Bead Test. Many silicates leave an insoluble "skeleton" floating in the
phosphate bead, but various other insoluble oxides behave in the same
manner. See P. 39 ..
Sodium Fluosilicate (4.7-7.7). The test is specific and is given by
elemental silicon and apparently also by all of its compounds. Borate
prevents the test if present as more than a minor constituent of the sample
and must be removed which may be done by adding I8-F H 2S0 4 and
repeatedly evaporating with methanol. If one is afraid of a possible inter-
ference by germanium, this latter could be eliminated by first evaporating
with 9-F HBr and some bromine, P. 68.
It should be considered that not all forms of Si0 2 are attacked by HF
with equal ease. Quartz and vitreous silica react slowly, and it is necessary
to grind them (in a steel mortar) to a fine powder as this is done with
silicates. Materials that may contain sulfide or carbonate should be roasted
first in air and then ignited for the removal of CO 2 , If heavy metals are
present, this should be done on a sheet of nickel (nickel crucible); if not,
one will do it in the platinum crucible in which the test is to be performed.
Obviously, apparatus of glass, silica, or any kind of silicate must be
avoided. The slide should consist of clear plastic; in an emergency, a
square of clear cellophane, cemented upon a glass slide, will do. Plastic
apparatus has been described by HAHN (624).
Use a platinum crucible of I-ml capacity and provided with a platinum
lid. If a solution is to be tested, evaporate it in the crucible. A solid sample,
of not more than I-mg weight, is simply placed upon the bottom of the
crucible. Apply a small drop (1 ",1) of water to the underside of the crucible
cover. Then treat the sample in the crucible with 1 to 3 mg NH4F (that
should come from a plastic container) and 2 drops I8-F H 2 S0 4, Place
the cover upon the crucible without delay. For cooling, put a large drop
of water on top of the crucible cover. Set the crucible on the center of
a large watch glass and heat it in this manner on the steam bath. After
two or three minutes, lift off the cover and touch its underside to the
P.57 Group IV A: Carbon-Lead Group 363

surface of the plastic slide. This will transfer the drop of condensate to
the slide. Place a fresh drop of water on the underside of the cover and
return the latter upon the crucible for the collection of more distillate.
A series of distillates may be collected, and several of them should be
tested if silicon is found to be absent in the first distillate.
Test the drop on the plastic slide by adding, with a platinum wire or
a plastic tool, a kernel of sodium chloride. If silicon tetrafluoride has been
absorbed by the drop of water, 3 SiF4 + 3 H 20 ~ 2 H 2 SiF 6 + H 2 SiO a,
the characteristic crystals of N a 2 SiF 6 separate. They form hexagonal
plates, short hexagonal prisms, or six-pointed stars that have a light pink
color because of the Christiansen effect. Avoid strong illumination; other-
wise the crystals are difficult to see since their refractive index is quite
close to that of the solution.
If crystals of sodium fluosilicate are obtained, transfer the next drop
of distillate to an ash-free filter paper. Add a drop of molybdate reagent
(5 g ammonium molybdate dissolved in 100 ml cold watyr and poured
into 35 ml 6-F HNO a) and warm the moist spot gently (jet of steam or
infrared lamp). Add a drop of benzidine solution (50 mg benzidine or its
hydrochloride is dissolved in 10 ml glacial acetic acid and then diluted
with water to 100 ml) and then expose to ammonia fumes. A blue spot
(reduction of the silicomolybdate to molybdenum blue and oxidation of
the benzidine to benzidine blue) indicates the presence of silica. The blue
spot should be compared with a blank test.

No. 32: Germanium, 72.59

The silvery white, brittle metal is little affected by acids but dissolved
by 3 % H 2 0 2 • It is resistant against NaOH solution but reacts violently
with the fused hydroxide to form germanate. The compounds of tetravalent
germanium are more stable than those of the divalent. The white Ge0 2
is amphoteric; dissolution in NaOH solution gives germanates; treatment
with strong HF, HCI, HBr gives conversion to the volatile tetrahalides
which, with the exception of GeF 4, are rapidly hydrolized by water.
GeF 4 gives H 2 GeF 6' the sodium salt of which is identical in appearance
with Na 2 SiF 6. To complete the similarity, germanate gives molybdate
complexes in analogy to silicic acid, and the very weak germanic acid
forms a strongly acidic complex with glycerol, glucose, and mannite
similarly to boric acid. Obviously, the identification of germanium requires
separation from boron (distillation of methyl borate) and from silicon
(and also titanium and tin by distillation of GeCl 4 or GeBr4). Germanium
appears in the arsenic group of the hydrogen-sulfide scheme of separation
and may also be isolated as the white GeS 2 •
Rubidium Hexafluogermanate (123). The test may be used after
separation from Si, Ti, Zr.
364 Confirmatory Tests P./)7

With a platinum loop transfer some solution of the test substance in

HF to a plastic slide. With a platinum wire, put into the edge of the drop,
a grain of RbC!. The colorless hexagonal bipyramids are characteristic;.
also hexagonal plates form.
Mannite Complex (5.0-6.3; molybdate, arsenite, Sn, Sb-3, Te may be
present). Borate gives the same test.
On a spot plate, mix a drop of the weakly acid test solution with a drop
of 1 % phenolphthalein in ethanol and add NaOH until the mixture turns
light pink. Mix carefully. If the pink color persists, add a few small crystals
of mannite (hexanhexol) and mix again. If germanate is present, the
mixture should turn colorless. A blank is desirable, and a tiny square
of pH test paper (Hydrion paper) might be briefly soaked in the test drop
before adding the mannite and again after the addition.
Benzidine Blue (5.6-6.9). Sb-3, tellurate and tellurite reduce the
sensitivity somewhat. Reducing substances, Si, P, and As must be absent.
Concerning the reagents see above the last paragraph of the section Silicon.
On a spot plate, treat a drop of the test solution with a drop of molybdate
reagent, a drop of benzidine reagent, and a few drops of saturated aqueous
solution of sodium acetate for the adjustment of the pH. A blue coloration
(benzidine blue and molybdenum blue) indicates germanium.

No. 50: Tin, 118.69

Flame rrest (6.2-7.5) (121). Niobium and arsenic are the only two
elements that interfere; gold gives a green luminescence when present in
high concentration (136).
In a porcelain crucible, treat 5 drops of sample solution with 5 ml
12-F HCI and a small piece of metallic zinc. Dip into this mixture the
closed end of a test tube, filled with cold water, and then hold it into the
reducing zone of a non-luminous Bunsen flame. Presence of tin is indicated
by a mantle of blue flame around the test tube (1151, 1152).
FEIGL and KAPULITZAS (121) obtain the limit of identification quoted
above by taking a drop of test solution on a magnesia stick and evaporating
it at a low temperature (holding it near a Bunsen flame; an infrared lamp
might be used or the radiation from some hot object). A droplet of 12-F HCI
is applied to the residue which is then held into the reducing (inned)
portion of a (non-luminous?) microflame. A mantle of blue flame fOlms
around the magnesia stick. If there is more than 0.25 ftg of tin, the moistening
with HCI and heating may be repeated several times and continues to
give the test.
Rubidium Chlorostannate (4.5-8.0; Sb: 2.0).
Proceed as directed in Expt.41.
Reduction of Phosphomolybdic Acid (3.0-6.0; Cu: -0.6).
Proceed as directed in Expt.30.
P.58 Group V A: Nitrogen-Bismuth Group 365

No. 82: Lead, 207.19

Flame Test: see P.38 and Table IX.
Potassium-Copper-Lead Nitrite (4.5-8.5; Cu: 2.5; Hg-ic: 2.0). "Large"
quantities of Bi or Cd hinder the test.
Proceed as directed in Expt. 34. Addition of the cupric ion as cupric
sulfate causes precipitation of PbSO, which then dissolves slowly in the
reagent to give reasonably large crystals.
Lead Iodide Slide Test (3.0-6.7). All ions and combinations of such
that give insoluble iodides will interfere. The test is useful after the isolation
of the lead.
Proceed as directed in Expt.33.
Lead Iodide Spot Test (3.7-6.7; Bi: -1.7). See also the preceding test.
Perform the test as directed in Expt. 24.

P.DS Group V A: Nitrogen-Bismuth Group

No.7: Nitrogen, 14.0067
Ammonia and Ammonium Ion
Ammonia readily transfers from alkaline solutions via the gas phase
to acid solutions, and this may be used to separate the ammonia of the
ammonium ion from all cations except those that derive from volatile
(organic) bases. Consequently, it is customary to combine tests for
ammonium ion with the "distillation" of ammonia. As a rule, the reagent
could be directly applied to the test solution if it were known that interfering
substances are absent.
Ammonium Chloroplatinate (3.7-7.0 if performed in the gas reaction
cell). Only volatile bases can interfere.
If much material is available for testing, treat some in a test tube
with a small volume of 6-F NaOH (and some Na 2S if mercury is present)
and, if it seems desirable, warm the mixture. It may be possible to perceive
the characteristic odor of ammonia. Heavy white fumes are obtained
on inserting into the gas space of the test tube a loop with 12-F HCI.
A loop with a solution of chloroplatinic acid will contain a precipitate of
chloroplatinate that may be transferred to a slide and inspected under
the microscope. Other reagents may be brought into the gas space by
means of the loop, and test papers may be introduced suspended from
a hook.
For the detection of small amounts of ammonia follow the directions
of Expt. 46. As an alternative, the NH3 may be absorbed in a droplet
of 3-F HCI (or other desirable acid), whereafter portions of this "distillate"
may be tested by adding Na 2PtCl e or any other reagent or reagent paper
desired.
366 Confirmatory Tests P.58
Test with Nessler's Reagent (5.7-9.7). The alkali metals do not interfere.
Sulfate, sulfide, and metal ions including Hg interfere; amine bases give
the test.
Reagent. Dissolve 5 g HgI2 and 3.65 g KI in 100 ml water. The solution
is stable. Before use, mix some of it with an equal volume 3-F NaOH.
This alkaline solution does not keep.
With a capillary pipet, transfer 0.1 ,ttl of the alkaline reagent to a small
square of filter paper and expose the latter to the vapors given off by
the test solution (which has been treated with NaOH) in a gas reaction
cell or equivalent closed apparatus. An orange fleck indicates the presence
of NH s.
The reagent may be added directly to the neutral or slightly acid
solution of an ammonium salt; depending upon the amount of the latter,
an orange precipitate or coloration will be obtained: IHg-O-Hg-NH 2 •
Hydrazine
The free base, H2N -NH2' is a colorless liquid at room temperature
(m. pt. = 1.4° C, and b. pt. 113.5° C). It is soluble in water (pK = 5.5)
and gives salts with one and with two equivalents of acid. Hydrazine is
a strong reducing agent which precipitates metallic gold, silver, and
mercury from the aqueous solutions of their salts. Cupric ion is reduced
to the cuprous state in acid solution, and to metallic copper in alkaline
solution. The oxidation products of hydrazine are ammonia, hydrazoic
acid, and nitrogen depending upon the oxidant and the conditions of
the reaction; permanganate gives mostly ammonia; and hydrogen peroxide,
mostly hydrazoic acid.
Salicylaldazine (5.7-7.0). Ammonia, ammonium salts, urea, thiourea,
azide, nitrate, and nitrite do not interfere.
Reagent. Boil 1 g salicylaldehyde with a solution of 1 ml glacial acetic
acid in 60 ml water until the oil disappears. Cool and filter for the elimina-
tion of undissolved aldehyde. The solution is stable, but must be filtered
from time to time for the removal of separating aldehyde.
In a small test tube, treat a drop of the slightly acid test solution with
a drop of the reagent. A white turbidity or a precipitate, depending upon
the amount of hydrazine, appear after a short time.
Hydroxylamine
The free base, NH 20H, forms colorless orthorhombic crystals that
melt at 34 ° C. The freshly prepared aqueous solution (pK = 8.0) is odorless,
but it gradually decomposes to give water, N 2, and NH s. Salts are formed
with one equivalent of acid. Hydroxylamine is a strong reducing agent
and precipitates the metal from solutions of gold, silver, and mercury.
Alkaline copper solutions, however, are reduced only to cuprous oxide.
The hydroxylamine is oxidized to N 2' nitrogen oxides, or HN0 2, depending
P.58 Group V A: Nitrogen-Bismuth Group 367

upon the conditions. Especially in alkaline solutions, hydroxylamine may

also act as oxidant and become reduced to NH a.
2 Fe(OH)2 + NH 20H + H 20 -->- 2 Fe(OH)a + NH a.
The reaction may be used to detect hydroxylamine in the presence of
hydrazine.
Oxidation to Nitrous Acid and Griess Test (6.7-8.0). Hydrazine, azide,
and urea do not interfere. Nitrite and oxime must be absent.
Reagents. (a) Dissolve 1.3 g 12 in 100 ml glacial acetic acid; (b) dissolve
1 g sulfanilic acid in 75 ml water and 25 ml glacial acetic acid; (c) dissolve
0.3 g alpha-naphthylamine in 70 ml water and 30 ml glacial acetic acid.
On a spot plate, mix a drop of the test solution with a few milligrams
sodium acetate, 1 to 2 drops of the solution of sulfanilic acid, and iodine
(solution in glacial acetic acid) until the mixture remains brown. Allow
the mixture to stand for three minutes, and then add O.I-F Na 2S 20 a to
the disappearance of the iodine color. Finally add 1 drop of the naphthyl-
amine solution. The red color of the azo dye indicates the presence of
NH 20H. A blank test is desirable.
H ydrazoic Acid and Azides
The anhydrous acid, HNa, is at room temperature a colorless, mobile
liquid of penetrating odor (m. pt. = - 80°; b. pt. = 37° C). The aqueous
solution (pK = 4.6) is fairly stable, but slowly decomposes to nitrogen
and an ammonium salt when boiled with a mineral acid. KMn0 4 , HN0 2,
or 12 acting on HNa give H 20 and N 2. The salts of HNa, with the exception
of those of the alkalies and alkaline earths, are explosive.
Silver Azide (4.0-7.0). Interference by sulfide, sulfite, thiosulfate
may be avoided by first oxidizing the neutral or alkaline test solution
with H 20 2 (I2I)~
Proceed as outlined in P. 36. The use of a small gas reaction cell for
the distillation will improve the limit of identification. As an alternative,
the acid may be collected in a small droplet of O.I-F NaOH which may
be finally neutralized by adding dilute HNO a, exposure to NHa fumes,
and evaporation on the steam bath. Portions of the aqueous solution of
the residue may then be used for tests. See next paragraph.
Ferric Azide and Cupric Azide. On a spot plate, treat a droplet of
neutral azide solution with a droplet of O.I-F FeCla; the mixture is red.
The coloriscopic capillary may be used for observation.
Treat a droplet of neutral test solution with a droplet of 1 % (I-F)
CUS0 4 solution; a reddish brown precipitate of CuN 6 separates. The test
may be performed in a capillary.
Be certain to discard all azide precipitates promptly by rinsing them
down the drain. They become very dangerous if allowed to dry.
368 Confirmatory Tests P.58

Nitrous Acid and Nitrites

Nitrous acid has not been isolated in the pure state. The aqueous
solution (pK = 3.4) gradually decomposes to HN0 3 , NO, and water.
All normal salts are soluble in water.
Nitrous acid is an oxidant in acid solution toward HI, H 2S, S02' NH 4 +,
and urea, but it reduces permanganate, chromate, chlorate, bromate, and
iodate, etc. Metallic zinc and acid or aluminum and NaOH solution reduce
to ammonia (ammonium salt), but this also happens with nitric acid.
The reaction with ammonium ion (or urea) may be used to eliminate
nitrite from solutions before testing for nitrate: N0 2- + NH4 + ~ N 2 +
+ 2 H 20.
Diazotation Test (6.7-8.7; nitrate: 2.0). The test is specific.
Reagents. See above under Hydroxylamine.
On a spot plate, treat a drop of the test solution with one drop each
of the solutions of (first) sulfanilic acid and (then) alpha-naphthylamine.
The red color of the azo dye is evidence for the action of nitrous acid.
Indole Test (6.0-8.0; nitrate: 2.0). The test is specific.
In a small test tube, treat 1 drop of test solution with 10 drops of a
0.015% solution of indole in 95% ethanol and 5 drops 7.5-F H 2S0 4 •
A reddish violet coloration is given by the nitroso indol formed.
Nitric Acid and Nitrate
Most tests for nitrate are also given by nitrite. Thus, reduction to
ammonium ion by metallic zinc and dilute H 2S0 4 and subsequent test for
ammonium are useful to prove the absence of both nitrate and nitrite
(and ammonium), but does not show which of the two is present.
The test with diphenylamine is given by nitrite as well as by nitrate
and by many other oxidants.
Nitric Oxide-Ferrous Complex. L. I. = 1 fhg or less. Nitrite interferes,
but gives a warning of its presence.
In a test tube, treat 1 ml of the test solution with 5 drops 6-F H 2S0 4
and an equal volume of freshly prepared saturated aqueous FeS0 4 solution.
Mix and cool by running tap water over the outside of the tube. If the
solution turns violet to brown, nitrite is present. If the solution remains
pale green or colorless, add 1 or 2 m118-F H 2S0 4 so that it collects unmixed
below the aqueous solution. This is easily accomplished by holding the
tube at an angle of 60 degrees to the horizontal and allowing the acid to
flow slowly down the side of the tube. A violet or brown zone at the inter-
face of the two liquids is derived from the reaction, 3 Fe++ + N0 3 - +
+ 4 H + ~ 3 Fe+++ + 2 H 20 + NO, and the dissolution of the NO in the
excess of FeS0 4 solution with the formation of the labile compound
FeS0 4 • xNO.
P.58 Group V A: Nitrogen-Bismuth Group 369

It should be possible to perform the test in a capillary with very small

volumes of solution.
Janowski's Test· (4.0-6.0). Nitration gives m-dinitrobenzene which, in
turn, furnishes a violet compound with acetone and NaOH. The test is
specific for nitrate, but the question of absence or presence of nitrate in
a large amount of nitrite will remain academic until means are found to
prevent the formation of some nitric acid during the performance of
tests (56). Chloride does not interfere.
On the steam bath, evaporate to dryness 1 ml of the neutral or slightly
alkaline test solution. Dissolve the residue in 0.3 ml of a solution of
1 ml nitrobenzene in 10 ml 18-F H 2S0 4 , Transfer the solution to a test
tube and then heat it for three minutes in the steam bath. Slowly add
5 ml purest acetone and mix thoroughly. Finally add 3 ml 40% (l4-F)
NaOH without mixing. If nitrate was present, a violet coloration will
appear at the interface of acetone and aqueous solutions, become slowly
stronger, and spread through the acetone layer.

No. 15: Phosphorus, 30.9738

Of the four allotropic modifications, the analyst may, as a rule drop
from consideration the very poisonous white phosphorus (m. pt. = 44.10 C)
since it is very unstable in contact with air. Sensitive tests are available
for the use of the toxicologist (40, 55, 56, 186).
The non-poisonous red and black varieties melt above 300 C, but 0

start burning before that temperature is reached. All forms are oxidized
to phosphoric acid by nitric acid or aqua regia.
Phosphine and Phosphides
Phosphine forms when phosphides are treated with water or dilute
acid or when any of the elemental forms of phosphorus reacts with hydroxide.
P 4 + 3 NaOH + 3 H 20 -->- PH a + 3 NaH 2P0 2.
It also forms when hypophosphorous acid or phosphorous acid is reduced
with metallic zinc and dilute sulfuric acid.
On the identification see P. 36.
H ypophosphorous Acid and H ypophosphites
H aP0 2 (m. pt. = 26.5 C) is usually met as a colorless syrupy liquid.
0

It is a reasonably strong acid, but only one hydrogen ionizes, and only
one series of salts is known.
On ignition, the acid as well as its salts give phosphate and phosphine.
2 H aP0 2 -->- H aP0 4 + PHa,
2 Ca(H 2P0 2)2 -->- Ca 2P 20 7 + H 20 + PHa.
There is no reaction with dilute H 2S0 4 , Concentrated H 2 S0 4 is reduced
to S02 on warming.
Benedetti-Pichler, Identification 24
370 Confirmatory Tests P.58

H SP0 2 and the hypophosphites are powerful reducing agents. AgNO a

in approximately neutral solution precipitates white AgH 2P0 2 which is
reduced to metallic silver even at room temperature; at the same time,
H2 may be liberated. In acid solution, CuS0 4 is reduced to metallic copper,
which may be used for the separation of copper from cadmium; at 50° C
a dark red precipitate of copper hydride may form first, and then decompose
with the liberation of H2 and Cu ° when heated to 100 ° C.
Permanganate Test (5.2-5.2); HsPOa gives the same test.
In a test tube treat 1 ml of the neutral test solution with 2 drops 10 %
(2.5-F) NaOH and 1 drop 0.02-F KMn0 4 • With hypophosphite or phosphite
present, the solution assumes a greenish coloration within two to three
minutes. Compare with a blank (3, 1203).
Phosphorous Acid and Phosphites
(Ortho) phosphorous acid, HsPO s, forms deliquescent needles that melt
at 73.6° C and are readily soluble in water. The acid is quite strong, but
only two hydrogens seem to ionize so that only two series of salts are
known. The alkali salts, NaH 2PO S and Na 2HPO s, are soluble in water;
most other metals give water insoluble salts. HsPOa is a strong reductant;
in contact with air, it slowly changes to H SP0 4 • The ions of gold, silver,
mercury, and copper are reduced to the metals. HgCl 2 is slowly reduced
at room temperature to Hg 2C1 2; if there is enough phosphite and the solution
is heated, the reduction continues to the metal. Neutral test solutions
give a white precipitate of Ag 2HPO s which, in strong solutions, changes
at room temperature to AgO and H SP0 4 ; in dilute solutions this reduction
takes place only on heating.
On ignition, HsPOa and phosphites undergo autoreduction-oxidation
similar to hypophosphite.
4 HaPOa ~ + PHa,
3 H SP0 4
8 Na 2HPO a ~ 4 Na P0 + Na P 20 + H 20 + PHs.
S 4 4 7

There is no reaction with dilute H 2S0 4 ; concentrated H 2S0 4 is reduced

to S02 on heating.
Alkaline Earth Phosphite. Barium, strontium, and calcium salt solutions
precipitate white, water insoluble BaHPO s, SrHPO s, and CaHPO s,
respectively, all of which are soluble in acid. The corresponding hypo-
phosphites are soluble.
H ypophosphoric Acid and H ypophosphates
The free acid, H 4P 206' forms colorless needles that melt at 55 ° C. The
melt decomposes at 70° to give HP0 2 and H SP0 4 • All four hydrogens
may be replaced by metal, and as one would expect, the solution of Na 4P 206
is distinctly alkaline due to hydrolysis.
P.58 Group V A: Nitrogen-Bismuth Group 371

H,P 206 is not reduced by zinc and dilute sulfuric acid and it is also
decidedly more stable toward oxidants than H sP0 2 "and HsPOs. AgNO a
gives a white precipitate that does not become dark. The aqueous solution
is not oxidized by dilute chromic acid and only slowly by KMnO, at room
temperature.
Thorium Hypophosphate. Acidify the test solution with HCI and then
add an equal volume 6-F HCI and a few drops of a thorium salt solution.
Hypophosphate gives a white precipitate of ThP 2 0 6 •
Phosphoric Acid and Phosphates
All phosphoric acids and acid solutions of phosphates revert on boiling
of the aqueous solution to orthophosphoric acid, HaPO" the colorless
crystals of which melt at 42.4 0 C. At 213 0 C, the acid loses water and
gives pyrophosphoric acid, H,P 20 7, m. pt. = 61 0 C, which is converted
to metaphosphoric acid, HPOa, at red heat. The latter vaporizes on further
heating. In fact, there are three series of isopolyacids: the chain phosphates,
Hn+2PnOan+l; the ring or meta phosphates, Hm(POs)m; and the extremely
unstable, highly branched ultraphosphates. For n equal to 1, 2, 3, 4,
and 00, the chain polyphosphate formula gives the orthophosphoric acid,
HaPO,; the pyrophosphoric acid, H,P 20 7; the tri(poly)phosphate, HoPS0 10 ;
the tetraphosphate, H 6P,01S; and (HPOa)n which up to this time was
called metaphosphoric acid.
All of these isopolyacids are converted to HaPO, by evaporation to
just dryness (if aqua regia is present or an equivalent oxidant, all compounds
of phosphorus will be converted to HaPO,), dissolving the residue in I-F
HNOa, and boiling gently for ten to twenty minutes.
Just a few hints shall be given concerning the identification of particular
acids. Only orthophosphates give a yellow precipitate when treated with
acid molybdate reagent at 4 C; neither poly (chain) phosphates nor meta
0

(ring) phosphates do this. At pH 3 to 4, AgNO a gives a yellow precipitate

with orthophosphate, a white precipitate with polyphosphates (chain
phosphates), and no precipitate with meta phosphates (ring phosphates)
if the solution is not too concentrated. Controls are indicated, and this
holds also for the albumin coagulation test, for which the reagent is a
saturated solution of fresh egg white in distilled water. A very small
amount of the phosphate to be tested is added to 3 ml O.2-F acetic acid,
and this solution is mixed with 2 ml of the albumin solution. A turbidity
or flocculation is obtained only with chain phosphates of high molecular
weight [including, of course, (HPOa)n], but not with HaPO" H,P 207'
HsPSOlO' or ring metaphosphates (3).
The following tests are given by HaPO, and orthophosphates (and also
by all isopolyacids that have been converted to HaPO, by boiling of the
aqueous solution).
24*
372 Confirmatory Tests P.58
Heating of the Solid Salts. Assuming that the cation is stable at high
temperature, tertiary phosphates remain unchanged on heating, secondary
phosphates form pyrophosphates, and primary phosphates become glassy
long-chain polyphosphates.
Ammonium Phosphomolybdate (4.0-7.3). Arsenate and possibly silicate
and vanadate give the same test. Reducing agents must be absent, but
the test may be carried out in presence of all common cations and anions.
Molybdate Reagent. Dissolve 5 g ammonium molybdate in 100 ml cold
water and pour the solution into 35 ml 6-F HNOa.
Transfer one drop (50 ",1) of the test solution to a watchglass and evaporate
just to dryness on the steam bath. Moisten with 16-F HNO a and again
evaporate to dryness for the removal of halide. Repeat this once. Dissolve
the residue in one drop of 6-F HNOa and transfer the solution to a centrifuge
cone. Add 2 drops of molybdate reagent and place the tube for ten minutes
into water of about 65° C. If there is a yellow precipitate, transfer some
of it by means of a capillary to a slide (a coverslip may be applied) and
inspect under the microscope. (NH')3PO,' 12 MoO a · 2 H 20 forms highly
refractive isotropic disks, octahedra, and occasionally cubes and pentagon
dodecahedra.
Wash the precipitate in the centrifuge cone with some 2-F HNO a.
Dissolve the washed precipitate in a volume of 6-F NHa to obtain an about
1 % solution and use this solution for the following test with magnesium
acetate.
Magnesium-Ammonium Phosphate Hexahydrate (4.0-7.7). Arsenate
gives the same test. Cations that may be precipitated by ammonia should
be absent. Polyphosphates give "amorphous" precipitates. Phosphite
gives tiny six-pointed stars and rosettes, dendrites reminiscent of butterflies,
and elongated plates with parallel sides and irregularly acute ends.
Take a volume of test solution representing a few micrograms of
phosphate and transfer it to a slide. Evaporate just to dryness, and dissolve
the residue in 1 ",1 2-F HNO a. Expose the droplet to the fumes of concen-
trated NHa to render it strongly ammoniacal. Finally add a crystal of
magnesium acetate (about 0.1-",1 volume or 0.6-mm diameter) to the edge
of the droplet. The crystals of MgNH,PO,' 6 H 20 form clusters of feathery
dendrites if the precipitation occurs instantaneously. Slow growth gives
prismatic crystals. Characteristic X-shapes may nearly always be found.
The conversion to silver phosphate as described in Expt.40 permits
differentiation between phosphate and arsenate.
Benzidine Blue (6.0-7.3; arsenate: 3.0). Silicic and germanic acids
behave similarly. The interference of arsenate is suppressed by the procedure
of the test. H 20 2, fluoride, and oxalic acid interfere because of reactions
with the molybdate reagent (the first gives permolybdates, the latter
complexes).
P.58 Group V A: Nitrogen-Bismuth Group 373

Reagent8. a) Ammonium molybdate 8ee above under Ammonium

Phosphomolybdate. b) Dissolve 50 mg benzidine in 10 ml glacial acetic acid
and dilute with water to 100 ml.
Place a drop of the molybdate reagent upon filter paper (best No. 589
of Schleicher and Schuell) and place into a drying oven. When the spot;
has dried, put into its center a drop of the test solution and add 1 drop
of benzidine solution and 1 drop of saturated aqueous sodium acetate.
A blue fleck or ring will form depending upon the amount of phosphate.
Arsenic acid does not interfere since it gives the heteropolycomplex
only very slowly in the cold and the reaction with benzidine leading to
the mixture of molybdenum blue and benzidine blue is given by the complex
only.
No. 33: Ar8enic, 74.9216
Flame Test. See P. 38 and Table IX. If the substance is heated in
the upper reducing zone (d, Fig. 10), all arsenic compounds are reduced
to elemental arsenic which vaporizes (sublimes above 600 0 C). The vapor
burns in the oxidizing zone with a bluish white flame to As 20 3 , whereby
a gallic-like odor is given off. A cold glazed porcelain surface inserted
close above the heated sample collects a brownish black mirror of elemental
arsenic; if it is inserted some distance above the heated object, it collects
white As 20 a. The As mirror dissolves readily in a drop of sodium hypo-
+
chlorite solution, AS 4 10 NaCIO + +
6 H 20 ........ 4 HaAs04 10 NaCl. The
As 20 a deposit is changed to yellow AgaAsOa if moistened with AgN0 3
solution and exposed to fumes of NHa (excess of which dissolves the yellow
precipitate ).
The same phenomena of reduction and oxidation take place when
arsenic compounds are heated on charcoal, Expt. 17, and also the garlic-like
odor will be perceived.
Liberation and Detection of Arsine (4.4-2.7 to 7.4-5.7 depending upon
the reducing agent). The action of acid or alkali on metal, liberating
hydrogen, converts arsenic compounds to AsH a which accompanies the
hydrogen. Interferences depend upon the choice of metal and reagent
and upon the method used for the detection of arsine. Interfere may all
substances able of giving a volatile hydride: Sb, (Bi), Ge, P, reducible
compounds of phosphorus, sulfur, and selenium. Mercuric salts and
fluorides may interfere with the formation of AsH 3 • The tests are known
under the names Berzelius-Marsh and Gutzeit and are widely used in
toxicology and in the testing of food and all kinds of articles of daily use.
For details see the literature (3, 40, 56, 162, 186).
Using pipets so as not to wet the inside wall, place a few pieces of
metallic aluminum, 1 ml test solution, and 1 ml 2-F KOH on the bottom
of a test tube. Push into the upper third of the tube a wad of cotton that
.374 Confirmatory Tests P.58

has been treated with some saturated aqueous solution of lead acetate
(for the absorption of H 2S), and place upon the opening of the tube a small
disk of filter paper which has been dipped into 3 % aqueous HgCl 2 solution.
Warm the reaction mixture at the bottom of the tube. A yellow to brown
spot on the filter paper indicates the presence of trivalent arsenic. If
arsenate is present, it should be reduced by adding a drop of aqueous
S02 and warming before the test is started. Antimony is not reduced to
SbHa and does not interfere (4.0-4.0; V, Mo, Cd, Hg, Tl, Pb, Sn, Sb,
Se: 2.0; W, Pt, Cu, Ag, Au, TI, Bi: 0.0) and (3.0-3.0; W, Pt, Cu, Ag, Au,
Bi, Te: 2.0).
Bettendorff Test (4.7-7.7; Sb, Sn: 4.0). Platinum metals, Au, Ag,
Hg, Se, Te are also reduced to the elemental state and interfere with the test.
Proceed as directed in Expt. 50.
Magnesium- (Calcium-) Ammonium Arsenate Hexahydrate (4.0-7.3).
The crystals are isomorphous with those of the corresponding phosphate
precipitate. All substances interfere that give with NHs a precipitate.
Proceed as directed in Expt.40. If CaCl 2 is added in place of the
magnesium salt, the crystals of the calcium salt separate, CaNH4AsO 4 . 6 H 20.
Silver Arsenate and Arsenite (Estimate: 5.0-5.0). The test is useful
for the identification of the valence state, but requires absence of all other
ions that would give a precipitate with AgNO a in neutral solution.
In a test tube, treat I ml of the test solution with a few drops of AgNO s
solution and then add strong NHs dropwise until the solution is clear again.
Holding the test tube at an angle of 45 degrees, run 16-F HNO s down
the inside wall of the tube so that it collects in a layer on the bottom of
the test tube. The precipitate will appear in the neutral zone between
the acid and the ammoniacal solution: AgsAsO 4 is chocolate brown; AgsAsOs
and AgSP0 4 are yellow. For performance as a spot test, 8ee Expt.29.

No. 51: Antimony, 121.75

Flame Test. See P. 38 and Table IX. The mechanism is the same as
explained under Arsenic. The flame is a pale greenish white; the garlic-like
odor is missing. The antimony mirror is black; the deposit of Sb 20 s turns
black when moistened with AgNO s and exposed to NHs fumes: Sb 20 S +
+ 4 AgNOs + 4 NHa + 2 H 20 -+ Sb s0 5 + 4 Ag + 4 NH 4NO a. - The
antimony mirror is not affected by NaClO.
Thermoluminescence Test. L. I. = I ng Sb. See Expt. 60.
Stibine and Its Detection. Stibine forms under essentially the same
(}onditions as arsine, 8ee under Arsenic, above. In absence of the latter,
the test described there may be used to detect antimony. Some changes
are necessary, however. Granulated zinc is used in place of aluminum
turnings, and I ml of the test solution is treated with I m18- to 10-F H 2S04,
P.59 Group VI A: Oxygen-Polonium Group 375

The filter paper is treated with AgNO s solution and a black precipitate
indicates antimony, see also P. 36.
Cesium Iodoantimonite (4.0-8.0). All substances interfere that react
with iodide, i. e., Hg, Pb, TI, nitrite, etc. Bismuth gives the same test,
but may be readily recognized by adding stannite reagent.
Proceed as directed in Expt. 35.
Quinine and Cinchonine Iodoantimonite (4.8-6.3). The interferences
are the same as with Cesium Iodoantimonite.
Proceed as directed in Expt.26.

No. 83: Bismuth, 208.980

Flame Test. Heated in the reducing zone, bismuth compounds give a
pale greenish or bluish white flame. The mechanism is the same as explained
under Arsenic, above. A barely visible deposit of Bi 20 3 may be caught
upon the glazed underside of a porcelain dish (filled with cold water)
held above the oxidizing flame. Fumes of HI may be obtained by warming
a drop of the acid in a porcelain crucible. On exposing the condensate
to the fumes, it forms red HBiI 4 • On breathing upon the deposit, the red
color fades, but reappears when the m::>isture evaporates. Exposure to
fumes of NH3 converts to orange NH 4 BiI 4 •
Thermoluminescence Test. L. I. = 0.1 ng Bi. See Expt. 60.
Precipitatiou of Sulfide and Conversion to Chromate and Metallic Bismuth.
L. I. = 8 ng Bi. See Expt. 58.
Cesium Iodobismuthate (4.5-8.0). Sb gives the same test, and all
substances that react with iodide will interfere. Even in presence of antimony,
bismuth will be recogJ;lized by the treatment of the precipitate with stannite
reagent.
Proceed as outlined in Expt. 35.
Bismuth Cobalticyanide Pentahydrate (4.5-8.0; Tl: 1.3; Pb: 1.1; Sn-ous:
0.7; Cu, Ag, Cd, Zn, Hg-ic, As, Sb: 0.0). 1 % chloride ion in solution prevents
the test. Antimony does not give this test.
Proceed as outlined in Expt. 36.
Quinine and Cinchonine Iodobismuthate (5.5-7.0). Substances that
react with iodide interfere with the test.
Proceed as directed in Expt.26.

P.59 Group VI A: Oxygen-Polonium Group

No.8: Oxygen, 15.9994
More or less pure oxygen is recognized by the fact that it rekindles
a glowing splint. In mixtures with other gases it may be recognized by
the fact that it gives brown N0 2 when colorless NO is added to the gas
376 Confirmatory Tests P.59

mixture. It also gives a deep red coloration when the gas mixture is allowed
to act upon an alkaline solution of pyrocatechol and FeS0 4 (603).
Ozone
The presence of relatively small percentages of ozone in an atmosphere
is already indicated by the characteristic odor. Its presence may be
confirmed by allowing the gas to act upon a neutral solution of KI in water;
03 will liberate 12 and the solution becomes alkaline, which may be proven
with a pH paper. H 20 2 does not react with neutral iodide solution.
For additional proof, a slightly acid solution of titanium sulfate may
be exposed to the gas; it will remain colorless if 03 is present.
Oxidation of Metallic Silver. Heat some silver foil to about 250 0 C
and then expose it to the gas to be tested: steel blue spots with violet
edges develop immediately. Silver which has been polished with emery
paper (and thus contaminated) does give the test at room temperature.
Water
Water is most convincingly isolated as such by vaporization followed
by condensation. The condensate may then be identified by testing with
pH paper and determination of boiling point (Expt. 21) and then freezing
point, p. 173.
Decomposition of Potassium Tetraiodoplumbate. Place upon spot test
paper a drop of 20 % K 2PbI 4 in anhydrous acetone and dry in an oven
of 105 C. Then add a drop of the organic liquid that shall be tested.
0

A yellow spot of PbI 2 is obtained with 0.05% H 20 in methanol and with

0.1 % H 20 in acetone. Peroxides interfere since they produce a brown
coloration.
Solid substances might be treated with a suitable solvent (anhydrous
acetone) which would extract water held as "moisture" and adsorbed water.
Hydrogen Peroxide
Pure H 20 2 melts at - 2 C to a colorless, volatile, strongly acid, and
0

explosive liquid that strongly irritates the skin. As a rule, only more or
less dilute solutions in water are met. These act as oxidants more or less
comparable to the action of various per acids. Characteristic for solutions
of H 20 2 is the liberation of oxygen on adding catalysts like Mn0 2. In
addition, H 20 2 reacts at times as oxidant and at other times as reductant.
It liberates 12 from acidified (H 2S0 4 ) solution of KI; and it reduces KMn04
in dilute JI 2S0 4 to Mn-2 ion.
Dioxy-Disulfato-Titanic Acid. Fuse a small amount of Ti0 2in a porcelain
crucible with 15 times as much K 2S 20 7 and dissolve the melt in cold 2-F
H 2 S0 4 • On a spot plate, treat a drop of this solution with a drop of the
test solution. A yellow to orange coloration [H2Ti02(S04)2 or Ti0 2 · H 20 2
or H 2Ti0 4] indicates H 20 2.
P.59 Group VI A: Oxygen-Polonium Group 377

Reduction 01 Ferricyanide (6.0-6.0). Persulfate does not give the test.

Perborate and percarbonate give a positive test. Iodide and sulfite must
be absent.
In a centrifuge tube combine 3 drops 3 % AgNO a, 2 ml 8-F NH a, and
3 drops freshly prepared aqueous 6 % KaFe(CN)6' Mix and centrifuge.
Transfer 1 ml of the clear supernate to 1 ml of the test solution. H 2 0 2
produces a white precipitate.

No. 16: Sulfur, 32.064

Depending upon the allotropic modification present, elemental sulfur
melts at 120° ("amorphous"), 119.0° (monoclinic), or 112.8° C (ortho-
rhombic). The boiling point of the dark red liquid is 444.6° C. In contact
with air, the liquid sulfur ignites at 248° C and burns with a blue flame
to give S02 of characteristic odor.
Crystalline sulfur (both modifications) is soluble in CS 2; octahedral
crystals are obtained upon evaporation of the solution. All forms of sulfur
are insoluble in water, but dissolve in hot caustic alkali,
4S + 6 NaOH -- 2 Na 2S + Na S 20 a + 3 H 20
2

in solutions of sulfide, especially on warming,

Na2S + n S -- Na2Sn+1 where n = 1, 2, 3, 4
and in solutions of sulfite,
Na 2 SO a + S -- Na 2S 20 a.
Bromine and HCl, KCIO a and HCl, aqua regia, and hot concentrated HNO a
slowly oxidize sulfur to H 2 S0 4 ,
In qualitative testing, sulfur is frequently met as a condensate of
liquid or solidified droplets. The following two tests may be carried out
with parts of such a condensate.
Test for Elemental Sulfur (5.7-5.0). Selen and phosphorus do not
interfere; halogens, carbon disulfide, hydrocarbons, chloroform, etc. must
be absent.
Treat the dry test substance in a test tube with 5 ml pyridine, reagent-
grade and freshly distilled. All forms of sulfur are soluble in this solvent.
Filter into another test tube and treat the colorless filtrate with one tenth
of its volume (0.5 ml) of 4-F NaOH. Depending upon the amount of sulfur
present, the mixture turns blue, olive green, to reddish brown. Also
polysulfides give the test since they liberate sulfur.
Oxidation to Sulfate (Estimate, L.1. = 0.1 f-lg S). Elemental sulfur
as well as nearly all compounds of sulfur, organic and inorganic, give the test.
Transfer a tiny particle of the material to be tested to a slide and add
0.5 f-ll 1 % (O.I-F) CaC1 2 solution. Invert the slide and place it upon the
opening of a bottle with bromine water so that the test drop is exposed
378 Confirmatory Tests P.59
to the vapor. After five minutes, inspect the test drop under the microscope.
The characteristic sheaves of needles of CaS0 4 • 2 H 20 will appear along
the edge of the drop as evaporation takes place if only little sulfur is
present (150).
Hepar Test. This test, which may be about as sensitive as the preceding
one, is given by elemental sulfur and all compounds of sulfur, selenium,
and tellurium.
When performing a charcoal test by heating the test material with
Na 2C0 3 , it is customary to finally place some of the melt upon a silver
coin and to moisten it with a drop of water. Sulfur compounds are reduced
to sulfide during the heating upon charcoal. In contact with the silver,
the sulfide gives a brown to black spot,
2 Na 2S + 4 Ag + 2 H 20 + O2 - 2 Ag 2S + 4 NaOH.
A blank test is recommended. On the performance of the test with acid
soluble sulfides such as OiS, ZnS, lazurite (lapis lazuli), and synthetic
ultramarines see BO:STI~CK (877).
Hydrogen Sulfide and Sulfides
Hydrogen sulfide is a colorless, poisonous gas, the rotten-egg odor
of which may be recognized when 1 volume of H 2 S is present in 700000
volumes of air. Water saturated at 20° 0 and 1 atm. pressure is about
0.13-F with respect to H 2 S.
For the detection of H 2 S gas, one uses filter paper moistened with a
solution of lead acetate or lead plumbite [solution of Pb(OH)2 in NaOHJ.
The latter will give a black spot with 1 ng H 2S. If the paper is
treated with an alkaline solution of sodium nitrosopentacyanoferrate,
Na 2Fe(NO)(ONh, a transient purple color is obtained. See also P.36.
Water soluble sulfides give the Hepar test, above, without preceding
reduction on charcoal. Acidifying with HOI or dilute H 2S0 4 liberates H 2S
that may be recognized by the odor and the tests indicated above. Poly-
sulfides, when treated with acid, give a white precipitate of sulfur in
addition to the H 2S. Sulfides that are not attacked by acid may be briefly
fused with Na200a in a covered porcelain crucible; in spite of some oxi-
dation, the melt will contain enough soluble alkali sulfide to give a test
or to liberate H 2S upon adding dilute HOI.
When heated in absence of air, some sulfides sublime and some decompose
with the liberation of sulfur, e. g., FeS 2 , pyrite - FeS + S. In contact
with air, the sulfides of the heavy metals are oxidized to give S02 and
oxide or the metal.
Antimony Trisulfide (4.0-5.3). The test is speci:!'ic, especially when
H 2S is first liberated and allowed to act upon the test paper.
Place upon spot test paper a drop of a 5 % solution of potassium-
antim:myl tartrate [tartar em3tic, K(SbO)OaH40aJ and expose to the gas
P.59 Group VI A: Oxygen-Polonium Group 379

liberated from the acidified sample (or add a drop of the test solution
and expose to fumes of HCI). An orange fleck of Sb 2S3 indicates sulfide.
Methylene Blue (7.7-7.7). The test which has been recommended by
EMIL FISCHER (601) should be specific. The test solution may contain
free H 2S or a sulfide that liberates H 2S on adding acid.
To 1 ml of the test solution add 0.1 ml12-F HCI, about 1 mg dimethyl-
paraphenylenediamine sulfate, NH 2 · C 6H, . N(CH 3 )2 . H 2SO" and-when it
has dissolved-a drop of 1 % (O.l-F) aqueous FeCI3 • With very little sulfide,
the color of the methylene blue may require one hour to develop. If a red
coloration is obtained, it is due to the action of the reagent upon the FeCI 3
and may be discharged by adding some more HCl.
Sulfurous Acid and SUlfites
Sulfur dioxide is a colorless gas, very destructive to the vegetation,
the suffocating odor of which may be perceived with 1 volume of S02
in 200000 volumes of air. Its saturated solution in water of 20° C, 1 atm.,
is about 1.6 formal and contains mostly S02 and only a very small amount
of H 2S0 3 and its ions. H 2 S0 3 is a moderately strong acid and a good
reducing agent.
The alkali sulfites and the acid sulfites of the alkaline earths are soluble
in water. All other sulfites are dissolved by dilute solutions of strong acids
with the liberation of S02' The sulfite ion has the ability to give soluble
complexes with many of the heavy metals.
S02 is recognized by its odor, the precipitation of BaS0 3 , SrS0 3 , and
CaS0 3 from solutions of the corresponding hydroxides, and the reduction
of permanganate, chromate, and iodine, etc.; see also P.36. One should
be aware of the fact that especially CaS0 3 (not BaS03 ) dissolves readily
in an excess of S02 by forming bisulfite.
Heated in absence of air, alkali sulfites are converted by auto oxidation-
reduction to sulfide and sulfate; the other sulfites give S02 and the oxide
or metal.
Strontium Sulfite (Estimate: 4.5-5.5; thiosulfate: 3.0). All substances
interfere that give a precipitate with SrCl 2 in neutral aqueous solution.
In a centrifuge cone treat the neutral or slightly alkaline test so]ution
with 2% (O.l-F) aqueous SrCl 2 solution. If a white precipitate is obtained,
remove the supernate, wash the precipitate once with a little water, and
then treat it with little 6-F HCI. S02 is liberated if the precipitate was
SrS0 3 , and the solution should stay clear when heated upon the steam
bath (absence of thiosulfate).
Test with Nitroprusside and Zinc SuHate (4.2-5.5). The test is not
given by thiosulfate (121).
On a spot plate, treat a drop of cold saturated zinc sulfate solution
with one drop 0.25-F K,Fe(CN)6 and one drop 1 % (0.04-F) Na 2Fe(CO)(CN)6'
380 Confirmatory Tests P.59'

Add to the mixture containing a white precipitate of zinc ferro cyanide

a drop of the neutral test solution. The precipitate turns red if sulfite is
present.
Test for Sulfite in Presence of Sulfide and Thiosulfate (3.5-4.7; sulfide:
1.9; thiosulfate: 1.7). S02 is liberated after destruction of sulfide and
thiosulfate and detected by its action on zinc nitroprusside paste (121).
The test may be carried out with the use of test tube and glass loop, Fig. 75,
or in apparatus shown in Fig. 18.
Reagent. Add an excess of ZnCl 2to a boiling solution of Na 2Fe(NO)(CN)s .
. H 20. Filter and wash the precipitate, and store it moist in an amber bottle.
Mix a drop of the test solution with 2 drops of saturated aqueous HgCI 2:
+ Hg++ ~ HgS,
S~

S20a~ + Hg++ + H 20 ~ HgS + H 2S0 4,

After one minute add a drop of 3-F H 2S0 4 and test for escaping S02 with
a small amount of zinc nitroprusside paste. When it may be assumed
that the S02 has acted, expose the paste briefly to NH3 fumes. It will
assume a red color. Concerning a more sensitive reagent for this test
see SENISE (936).
Thiosuljuric Acid and Thiosuljates
The strong thiosulfuric acid probably forms when a solution of thio-
sulfate is acidified, but it decomposes quickly into S02 + H 20 + S.
In absence of air, the ignition of alkali thiosulfate gives sulfate, poly-
sulfide, sulfide, and sulfur which distills off. Most thiosulfates are soluble
in water; only those of Ba, Ag, and Pb are just slightly soluble. All thio-
sulfates are dissolved and decomposed by dilute acids with the liberation
of S02 and precipitation of sulfur. Thiosulfate ion has, furthermore, a
distinct tendency to give soluble complexes with heavy metals.
Like sulfite, thiosulfate reduces permanganate, chromate, and iodine
in acid solution.
Reduction of Ferric Chloride. Adding a few drops of FeCl3 solution
to the neutral or slightly acid test solution produces a violet coloration
that disappears after a short time to give a solution containing ferrous ion
and tetrathionate.
2 S20a~ + 2 Fe+++ ~ S406~ + 2 Fe++.
Test with Nitroprusside. Add test solution to a solution of K2Fe(NO)(CN)s
which has been treated with a few drops of KaFe(CN)6 solution and a few
droFs of 3-F NaOH. Thiosulfate gives a blue coloration. It darkens on
standing, heating, or adding of some more ferricyanide.
Tri-Ethylenediamminonickel Thiosulfate (3.0-4.3). Dithionates give a
similar precipitate; sulfide, sulfite, and tetrathionate do not interfere; the
plates of persul£ate cannot be mistaken for a thiosulfate precipitate.
P.59 Group VI A: Oxygen-Polonium Group 381

Reagent. Dissolve 4 g Ni(N0 3)2' 6 H 20 in 10 ml water and dropwise

add ethylenediamine until the solution is violet.
Place a droplet of the neutral or slightly alkaline test solution upon a
slide and add one or two droplets of the reagent solution. Ni(H 2N· C2H, .
. NH2)3S203 forms strongly anisotropic prisms and needles.
Thionic Acid8 and Thionate8
None of the acids H 2S.. 0 6 , where n = 2, 3, 4, 5, has been isolated in
the pure state. Even the aqueous solutions and the solutions of the salts
are unstable. Heating of a solution of dithionic acid gives H 2S 20 6 -..
-.. H 2SO, + S02' All other thionic compounds decompose with the
separation of some sulfur.
Sulfuric Acid and Sulfates
For the detection of SO, ion in acid insoluble sulfates, LUIS (957) heats
a particle of the solid substance with 85 % H 3PO, in a glass tube that is
drawn out to a capillary. The H 2SO, is distilled into the capillary and
identified by the tests given below. As little as 20 ng SO" given in the
form of insoluble sulfate, may be detected.
Calcium Sulfate Dihydrate, Gypsum (2.7 -6.2). The test seems to be specific.
Transfer a droplet of test solution to a slide and acidify it, if necessary,
by exposing it to the fumes of 12-F HCI. Next to the test drop, deposit
a like volume of 1 % (O.l-F) CaCl 2 solution and draw a channel to connect
the drops. Fine needles, sheaves of needles, rhombic plates, and arrow-head
(swallow-tail) twins of monoclinic CaS0 4 • 2 H 20 usually exhibit oblique
-extinction.
Barium Sulfate (Estimate: 6.0-7.3). The test is specific.
In a small centrifuge tube treat 50 pI of the nearly "neutral" test
solution with 15 pI 12-F HCI and 10 pI I-F BaCI 2. Mix, centrifuge, and
inspect the tip of the cone under magnification. A finely crystalline,
white precipitate indicates the presence of sulfate.
If other sulfur compounds and reducing substances are absent, the
test may be performed with the addition of KMnO, to make the solution
dark red before adding the BaCI 2. Because of coprecipitation, the sulfate
precipitate will appear violet and clearly visible after destruction of the
permanganate excess in solution by adding H 20 2.
Per8ulfuric Acid and Per8ulfate8
Solutions of persulfates in water decompose slowly when cold and more
rapidly on warming.
2 S208~ + 2 H 20 -.. 4 HS0 4 - + O2,
Consequently the freshly prepared solution does not give a precipitate
with BaCI 2, but on standing for some time and more quickly on boiling,
BaS0 4 will precipitate.
382 Confirmatory Tests P.59'

Persulfates are strong oxidizing agents and precipitate from neutral

or slightly acid solutions of Mn, Co, Ni, and Pb the corresponding black
peroxides. Black silver peroxide, Ag 20 2 , is precipitated from AgNO~
solution. If a concentrated solution of ammonium persuHate is treated
with NHs and a little AgN0 3 , nitrogen is liberated and the solution becomes
boiling hot.
Oxidation of Iodide. Persulfates react with neutral solutions of KI to
liberate iodine. Percarbonates and perborates do not.
Oxidation of Benzidine (4.7-6.0; bromate: 3.0; iodate: 3.8). Alkali
peroxides, perborates, and H 20 2 do not give the test. Chromates, per-
manganates, ferricyanides, and hypohalogenites give the same test and
must be absent.
On a spot plate, treat a drop of the neutral (or acidified with acetic acid}
test solution with a drop of a 2 % solution of benzidine in 2-F acetic acid.
A blue coloration indicates presence of persulfate.

No. 34: Selenium, 78.96

Like sulfur, selenium occurs in several allotropic modifications: red
amorphous, melting at 50° C; crystalline, melting at 220° C; and metallic,
melting at 217.4° C. The boiling point is 688° C. The element and all
of its compounds are severely toxic.
Flame Test. When heated in the upper reducing zone of the Bunsen
flame, selenium and its compounds give a blue coloration, and an odor
of rotting radishes will be perceived. Red selenium will condense on a
cold porcelain surface brought into the blue flame. A drop of 18-F H 2 S0 4
placed upon the deposit will dissolve the selenium to form a green solution
from which red selenium will again separate upon adding water.
All selenium compounds will give the odor of rotten radishes and a blue
flame when heated with N a 2CO S upon the charcoal. The melt gives the
Hepar test, see under Sulfur, above.
Hydrogen Selenide and Selenides
H 2 Se is a colorless gas with properties and odor similar to H 2 S. It is
more soluble in water and a stronger acid than the sulfur compound, and
it forms acid selenides, neutral selenides, and polyselenides. The selenides
of the heavy metals are strongly colored, insoluble in water, and some
even insoluble in acid.
In general, selenides may be decomposed by dilute HCI or H 2 S0 4
with the liberation of H 2 Se which may be (a) lighted to give a blue flame
and a Se deposit on cold porcelain; (b) passed through a glass tube which
is heated in one spot to give a red selenium mirror; or (c) absorbed in water
from which red selenium will separate on standing or, faster, on bubbling
air through it.
P.59 Group VI A: Oxygen-Polonium Group 383

Selenous Acid and Selenites

Se0 2 forms white needles that sublime at 317 0 0 and dissolve in water
to give H 2SeO S which is very soluble in water and forms hexagonal prisms
that decompose on heating. H 2SeO s is a weak, dibasic acid; the salts are
mostly colorless and soluble in acid if insoluble in water. Characteristic
of all selenium compounds is the tendency to decompose, more or less
slowly, with the separation of the element. The identification and separation
is mostly based on some method for reduction to elemental Se.
Barium chloride precipitates from neutral solutions white BaSeO s
which is soluble in dilute HOI; cupric sulfate gives a greenish blue, crystalline
precipitate with neutral selenite solution, whereas selenate solution is not
precipitated. Hydrogen sulfide precipitates a yellow to orange mixture
of selenium and sulfur, soluble in ammonium sulfide and in alkali sulfide
and hydroxide solutions.
Reduction with Hydriodic Acid (4.4-5.9). Sn, Sb-3, and tellurate do
not interfere. Molybdate must be absent; Ge and arsenite prevent the test
when present in hundredfold excess. Selenates do not give the test.
Place upon filter paper one drop of concentrated HI (or 12-F HCl in
which some KI has been dissolved) and put a drop of the acid test solution
into the center of the moist spot. Add a large drop of 5 % Na 2S 20 S solution.
A reddish brown fleck that does not dissolve indicates selenite.
Reduction with SuHur Dioxide (5.2-5.2). As-3, Sb-3, Ge, Sn-4, Te do
not interfere; Sn-2 must be absent.
In a test tube treat 1 ml of the test solution with 1 ml 18-F H 2SO,
and saturate with S02 (which may be obtained by dropwise adding of
concentrated H 2 SO, to solid sodium sulfite). Boil 1 minute and then
allow to cool. A red coloration or precipitate indicates selenium (also
selenate gives this test).
Selenic Acid and Selenates
H 2 SeO, forms hexagonal prisms that melt at 58 C and are very soluble
0

in water.
BaCl 2 gives a white precipitate which is insoluble in dilute acids; it
dissolves, however, on boiling with HCI.
BaSeO, + 4 HCl -. BaCl 2 + H 2 SeO s + C1 2 •
HzS gives no precipitate with selenic acid, unless the solution is boiled
with HCl which reduces to selenous acid.

No. 52: Tellurium, 127.60

Tellurium is silvery white with a metallic luster and very brittle, m. pt. =
452 0 C, b. pt. = 1390 0 C. From solution, it is precipitated as a voluminous
brown powder. When heated in air, it will burn with a bluish green flame
to TeO z"
384 Confirmatory Tests P.59

Alkali tellurides are colorless, but their aqueous solutions quickly

become red in contact with air due to the formation of polytelluride.
The tellurides of the heavy metals are dark in color. Treatment with acid
gives H 2Te, a colorless gas with the odor of H 2 S, thatis very soluble in water;
in contact with air, the solution rapidly decomposes with deposition of
elemental tellurium. The gas burns with a bluish flame to Te0 2, and
deposits (mirrors) may be obtained as with H 2Se.
The white, crystalline dioxide sublimes at 450 0 C without melting.
It is slightly soluble in water which then contains the slightly soluble
H 2TeO a that cannot be prepared in pure form since it starts losing water
at 40° C to finally become Te0 2. The corresponding tellurites resemble
the normal sulfites and selenites.
The white solid H 2Te0 4 starts losing water at 160 C to end up as
0

yellow TeO a. The very weak acid is quite soluble in water which seems
to contain orthotelluric acid, HaTeO 6'
Halides, such as TeCl 2 and TeCI 4 , hydrolyze in aqueous solution.
Flame Test. Metallic tellurium is formed by heating tellurides in the
upper reducing zone of the Bunsen flame. A pale green coloration is obtained,
and a black deposit may be collected on the glazed surface of porcelain
or upon the outside surface of a test tube filled with cold water. The
deposit dissolves in a drop of concentrated H 2S0 4 to give a carmine red
solution. Adding water again precipitates black tellurium. See also P. 38
and Table IX.
Tellurium compounds fused with Na 2CO a upon charcoal give the
Hepar test, see under Sulfur, above.
Treatment with Hydrogen Sulfide precipitates from acid solutions of
Te-4 the reddish black TeS 2 which is readily soluble in ammonium sulfide,
etc. The same precipitate is obtained with hot solutions of Te-6, but cold
solutions of tellurate are not precipitated by H 2 S, which provides a simple
means of separating heavy metals from tellurium.
Acidifying with Hydrochloric Acid precipitates quite difficultly soluble
H 2TeO s from solutions of tellurites. Solutions of tellurate give no precipitate
in the cold; on boiling, the soluble HaTe06 is converted to the slightly
soluble H 2TeO s with liberation of chlorine.
Reduction with Sulfur Dioxide. From solutions of Te-4 in dilute HCI,
Te is precipitated completely as a black powder. From 12-F HCI, tellurium
is not precipitated by S02' not even on boiling; this provides a means of
separating Se and Te.
Cesium Chlorotellurite and Iodotellurite (3.0-6.5). Selenium does not
interfere, but interferences must be expected from Ag, Pt, Sn, Pb, Sb, Bi,
and possibly others.
Tellurate may be converted to tellurite by boiling with strong HCI.
Acidify selenite solution with HCI and evaporate upon the steam bath
P.60 Group VII A: Halogen Group 385

just to dryness. Dissolve the residue in 3-F HCl to obtain a solution

containing not less than 1 % Te. Transfer 0.5 to l,ul of this solution to
a slide and introduce a large grain of CsCI into the center of the drop.
There will be a dense, granular precipitate close around the reagent, but
beyond this zone, moderately large, well developed, light yellow octahedra
of Cs 2TeCl 6 may separate. Still further out, thin triangles and hexagons
may form. On adding KI to the drop, the crystals are converted to black
CS zTeI 6 •
Reduction with Stannite Reagent (4.6-5.9). Tellurite and tellurate give
the test. Selenite and selenate do not interfere; molybdate, Cu, Ag, Hg,
Sn-4, arsenite, Bi, and Sb-3 interfere and should be absent.
Reagent. Dissolve 1 g SnCI 2 • 2 H 20 in 1 ml 12-F HCI and dilute with
water to 10 m!.
Transfer 1 drop 6-F NaOH to the spot plate, and add first 1 drop of
the stannous chloride solution and then 1 drop of the alkaline test solution,
preferably a soda extract which would not contain the interfering metals.
Presence of Te is indicated by the separation of black flocks. With low
concentrations, only a gray coloration is obtained and performance of a
blank becomes necessary (121).

P.60 Group VII A: Halogen Group

No.9: Fluorine, 18.9984
Pure (HF)n is a colorless liquid that boils at 19.40 C to give an intensely
corrosive gas. The aqueous solution, hydrofluoric acid, is a weak acid
which is able to penetrate tissue deeply, without any immediately visible
discoloration of the skin, to inflict extrem9ly painful burns that heal slowly.
The vapor is as dangerous as the liquid.
The following tests apply to fluorides and hydrofluoric acid.
Calcium Fluoride (Estimate: 6.0-7.3); all anions interfere, that give
calcium salts insoluble in dilute acetic acid: borate, oxalate, silicate,
phosphate, and arsenate.
Treat a drop of the neutral test solution (neutralized soda extract)
with a drop of 4-F acetic acid and a drop of 1 % (O.I-F) CaCl 2 solution.
CaF 2 separates as a white, slimy precipitate that is difficult to filter, but
may be collected and washed with the use of a centrifuge tube and used
for the alizarin test below.
Etch Test According to Mannheimer. Heat a rod of lead glass in the
reducing flame until it turns black by the separation of lead on the surface.
Then draw out a supply of fine threads.
Using the technique of the fiber tests, Expt. 57, insert the end of such
a thread into a test drop obtained by treating a sample (solid or liquid)
on a platinum sheet (foil) or on a plastic sheet with an equal volume of
Benedetti-Pichler, Identification 25
386 Confirmatory Tests P.60

9-F H 2 S0 4 , Move the supporting sheet from time to time to obtain a stirring
action. After five minutes remove the thread from the test drop, rinse it
in water, and mount it in cedarwood oil between slide and cover glass for
microscopic inspection. HF removes the black surface layer. Observe
the boundary line where the thread ermerged from the solution. Silicate
and borate in the sample will interfere with the test.
Sodium Fluosilicate (4.0-7.0; phosphate, sulfate, chloride, hypochlorite,
chlorate, bromide, hypobromite, bromate, iodide, iodate, periodate: 2.0).
Borate hinders the test and should be absent.
The test may be carried out in a 1-ml platinum crucible with lid, in
a small, 1-ml, porcelain crucible covered with a slide, in a test tube, Fig. 75,
or in apparatus shown in Fig. 18.
If a solution is to be tested, transfer a sample of it to the "pot of the
still" and evaporate it there to dryness. If the sample is a solid, simply
transfer it to the same location. Add a small amount of finely powdered
glass. Before adding a drop of lS-F H 2 S0 4 , get ready the loop with a
drop of water in it or the cover of the crucible with a droplet of water
hanging on its underside. Add the acid, get the water drop into position
(cover the crucible), and warm the "pot" (steam bath or jet of steam).
After two to three minutes transfer the drop of absorbing water to a slide
and continue the distillation with a fresh drop of water in place, compare
P.57 under Silicon.
The drop of water may contain a white precipitate resulting from:
"pot", Si0 2 + 4 F- + 4 H+ SiF + 2 H 0,
-)0 4 2

drop, 3 SiF + 3 H 0
4 2H Si0 + 2 H SiF
-)0 2 3 2 6'

A heavy precipitate indicates also a high concentration of fluosilicic acid;

in such instances dilute the drop of "distillate" with four volumes of water
and use only a fraction for the following test.
If the drop seems clear, do not dilute it, just place a grain of NaCl (of
appropriate size) into the edge of the drop. Light pink hexagonal plates,
short prisms, and six-pointed stars of Na 2 SiF 6 indicate the presence of
fluoride. If the test remains negative and the sodium chloride does not
dissolve in the test drop, the latter may be too concentrated and adding
some water may start the crystallization. See also P.57 under Silicon.
Use the distillate also for the test with molybdate and benzidine described
under Silicon. The presence of silica is indirect proof for the presence of
fluoride in the sample.
The test will not discover small amounts of fluoride in silicates. To
this end, it will be necessary to perform a Na 2C0 3 fusion and to test for
the fluoride in the aqueous extract, P.72.
Action Upon the Alizarin Lake of Zirconium (4.0-5.3; borate, chloride,
chlorate, perchlorate, bromide, bromate, iodide, periodate: 2.0). Ferri-
P.60 Group VII A: Halogen Group 387

cyamde, oxalate, nitrate, phosphate, thiosulfate, sulfate, hypochlorite,

and hypobromite interfere with the test and must be absent.
Reagent. To 0.1 % ZrOCl 2 solution in 5-F HCl add a 0.2 % solution of
alizarin S (alizarin-3-sulfonic acid) in ethanol until an excess of the dye is
indicated by the fact that ether, shaken with a sample of the mixture,
becomes yellow. Heat the solution for ten minutes on the steam bath.
Dip filter paper into the solution and then dry it.
Place upon the reagent paper first a drop of 8-F acetic acid and then
add a drop of the test solution. If fluoride is present, a yellow spot appears
on the violet paper. The reaction may be hastened by exposing the spot
to a jet of steam.
Red Zr-lake + 6 F- - ZrF 6 = + yellow dye.
The test appears useful in the presence of large amounts of borate.
The interferences by nitrate and the halogen anions may be removed by
first precipitating CaF2 (and possibly BaSO,) and performing the test by
mixing the precipitate with a small droplet of the violet reagent solution (121).
Red Zr-Iake + 3 CaF 2 -ZrF6= + 3 Ca++ + yellow dye.
Additional Tests of interest have been recently described by BALLOZO
and WEISZ (941) and by FINE and WYNNE' (776). Plastic apparatus has
been recently described by HAHN (624).

No. 17: Chlorine, 35.453

Flame Test. Organic and probably most inorganic compounds of
chlorine, bromine, and iodine (but not fluorine) give a blue or green colora-
tion to the Bunsen flame if heated on a copper spatula, see Expt. 18.
The test is not specific since also cyanides, thiocyanates, and various
organic nitrogen compounds produce the phenomenon. If inorganic
substances .are present, one has to add all the interferences by elements
giving green and blue flame colorations. If the sample does not give a
green or blue flame when heated upon the oxidized copper spatula, one
may assume that compounds of chlorine, bromine, and iodine are probably
absent. The Beilstein test may be performed with very small amounts
of material (919).
Elemental Chlorine
The gas is recognized by its yellowish green color, its characteristic
odor, its bleaching action upon litmus and indigo, and its ability to displace
bromine from bromides and iodine from iodides. These tests are also given
by water that contains chlorine dissolved.
Eosine. L. I. = 1 pg Cl2 (estimate); bromine gives the same test.
Reagents. (a) 0.1 % solution of fluorescein in 50% ethanol which has
been made slightly alkaline by adding KOH; (b) mix part of solution (a)
with an equal volume O.05-F KBr in O.1-F aqueous KOH.
25*
388 Confirmatory Tests P.60

Cut two strips of filter paper, about 5 mm X 5 cm. Dip the end of one
strip into reagent solution (a), and the end of the other strip into reagent
solution (b). Expose the yellow ends of both strips to the gas. The presence
of chlorine is proven if only one of them turns pink. If both become pink,
bromine is present and chlorine may be present.

Hydrochloric Acid and Chlorides

Liberation of Hydrogen Chloride (3.3-6.3).
Most solid chlorides give HCI gas if warmed with concentrated H 2 S0 4
or H aP0 4 • See P. 37. The liberated HCI is permitted to act upon AgNO s
solution (test tube and glass loop or gas reaction cell, etc.) and the precipi-
tated AgCI is recrystallized from NH a•
Liberation of Chlorine (5.3-5.3).
Using suitable apparatus (test tube and glass hook, gas reaction cell,
etc.), liberate chlorine by treating the solid material or the test solution
with 6-F H 2 S0 4 and a grain of Mn0 2 • Warm the mixture and test the gas
with both fluorescein reagent papers described above under Elemental
Chlorine.
Chromyl Chloride and Silver Dichromate (Estimate: 4.0-5.3; chlorate,
perchlorate, bromide, bromate, iodide, iodate, periodate: 2.0). Fluoride,
nitrite, and nitrate must be absent. Fluorine gives a volatile chromium
compound, and nitrite and nitrate may interfere by converting the chloride
to chlorine and nitrosyl chloride.
Place the material to be tested on the bottom of a I-ml porcelain crucible
(glass cup of similar dimensions). If a solution is to be tested, evaporate
a sample of it in the crucible. Add about 1 mg finely powdered K 2Cr 20 7
to the test material. Place upon the crucible a microscope slide with a
droplet of l,ul 4-F HNO a hanging on the underside. Lift the slide for a
moment, and from a pipet, add one large drop 18-F H 2S0 4 to the bottom
of the crucible. Heavy, brownish red fumes of Cr0 2Cl 2 (b. pt. = 117.6° C)
may be seen in the crucible. Warm the bottom of the crucible by placing
it on a steam bath or cautiously playing a microflame on it. After one or
two minutes remove the slide and replace it by another one. The droplet
of acid may have become orange yellow, and there may be a dark deposit
with a metallic luster around the droplet.
Place a kernel of AgN0 3 into the drop. If crystals of Ag 2Cr 20 7 do not
form (see Expt. 25), use a glass needle to mop up with the drop the whole
circular area that was exposed to the atmosphere in the crucible. The
crystals may appear then or on evaporation of the drop. The presence
of chromate (from Cr0 2Cl 2+ +
2 H 20 -'>' H 2Cr0 4 2 HCl) on the slide proves
the presence of chloride in the test material. The curdy precipitate of silver
halide does not contribute to the argument since it may be AgBr.
P.60 Group VII A: Halogen Group 389

Silver Chloride (Estimate: 8.0-9.3). All anions interfere, that give a

silver salt insoluble in acid.
Onto a black spot plate place one drop of the neutral or slightly acid
test solution and two (separate) drops of 1% (0.05-F) AgNO a in 3-F HNOa•
Use strong lateral light and make certain that all drops are clear. Then
combine the drop of test solution with one of the drops of reagent. Keep
the second reagent drop for a blank. A white turbidity or precipitate in
the test drop indicates halide. The turbidity should clear up when the
test drop is exposed to fumes of ammonia.
To improve the sensitivity, the test could be performed under the
microscope with the use of darkfield illumination.
To eliminate the interference of iodide, bromide, and thiocyanate,
WEISZ (923) suggests the following procedure. Treat the test drop with
I drop of a 2 % solution of 8-hydroxyquinoline in 4-F acetic acid, I drop
of a mixture of 2 volumes 6 % H 20 2 with I volume 4-F acetic acid, and
a small drop of 8-F HNOa. Heat four minutes to get complete reaction
of the liberated halogen with the mane, but do not evaporate to dryness.
Transfer the drop of the clear, honey yellow liquid to the spot plate and
add AgNO a• The turbidity of AgCI is still visible with dilutions I: 20000,
or with the adopted form of notation: (4.3-5.6; iodide, bromide, thio-
cyanate: 1.7).

Hypochlorous Acid and H ypochlorites

The very weak hypochlorous acid is assumed to be present in aqueous
solutions of chlorine monoxide, an orange yellow gas (b. pt. = 3.8 0 C)
which decomposes explosively into Cl 2 and O2 when heated.
2 HCIO ~ Cl 20 + H 0.
2

All hypochlorites are soluble in water and decomposed by acids,

including carbonic acid,
4 HCIO --+ 2 Cl 2 + O + 2 H 0:
2 2

The chlorine may be detected by the reagents given under Chlorine, above.
In general, the detection must be based upon the observation of the
oxidizing action and of the products of decomposition. Alkaline solutions
of hypochlorites bleach blue KI-starch paper, litmus paper and indigo
solution, but they do not reduce pernianganate. Auto oxidation-reduction
takes place when boiling solutions of hypochlorite,
3 CIO- --+ CIOa - + 2 CI-.
The removal of a product of the reaction will obviously aid it; thus adding
silver nitrate makes the reaction complete at room temperature,
3 CIO- + 2 Ag+ - CIO a- + 2 AgCI.
390 Confirmatory Tests P.60

After testing the supernatant solution for chlorate, it remains to prove

that chlorate was not present previous to adding AgN0 3 •
There is a possibility to test for hypochlorite in the presence of
chlorine (56), but the problem seems rather academic in nature, for the
two are inseparable: alkaline solutions of chlorine always form hypo-
chlorite, and acid solutions of hypochlorite usually contain some chlorine.
Ohloric Acid and Ohlorates
A 40 % chloric acid, HCI0 3 • 7 H 20, is a colorless and odorless liquid
which starts decomposing at 40 0 C. The acid and its salt are strong oxidants
and give explosive mixtures with combustible substances (S, P, C, organic
substances, etc.). All chlorates are soluble in water. .
Dilute Sulfuric Acid. Addition of H 2 S0 4 to the solution of a chlorate
causes slow decomposition:
3 HCI0 3 -+ HCI0 4 +H 20 +Cl 2 + 2 O 2,
KI-starch paper will turn blue in the acid solution.
Concentrated Sulfuric Acid. In a test tube, treat a tiny particle of the
test substance with 1 drop of 18-F H 2S0 4 , Chlorates react with the liberation
of orange yellow chlorine dioxide gas which colors the sulfuric acid yellow
and has a chlorine-like odor.
+
3 KCIO s 3 H 2S0 4 -+ 3 KHS0 4 + HCIO, + +
2 CI0 2 H 20.
The mixture is dangerous and may violently explode on warming. From
a capillary, add a few microliter of a saturated aqueous solution of aniline
sulfate. A blue coloration is given by chlorate but not by nitrate.
Reduction to Chloride. In a centrifuge tube, acidify 0.5 ml of the test
solution with 1 drop 4-F HNO s and then add dropwise 1 % (O.I-F) AgNO s
until it does no longer give a precipitate. Centrifuge and treat the clear
supernate with a kernel of reagent-grade bisulfite or metabisulfite. If
chlorate was present, it is reduced to chloride which gives a white precipitate
or turbidity of AgCl with the excess of AgNO s in solution.
Ignition of Chlorates. Oxygen is given off, and the residue contains chloride.
Perchloric Acid and Perchlorates
Anhydrous perchloric acid, HCIO" is a colorless liquid that boils at
about 130 0 C and decomposes first slowly and finally explodes. An acid
of about 70%, HCI0 4 • 2 H 20, is sufficiently stable for practical use as
dehydrating agent and oxidant. When hot, the action of the 70 % acid is
comparable to chromic-sulfuric acid; when cold, the acid as well as its
salts are remarkably stable. Acid solutions are not reduced by S02 and
do not react with KI-starch paper or bleach indigo solution. Dilute and
concentrated H 2SO, have no visible action.
With the exception of the salts of K, Rb, Cs, NH 4 , and organic bases,
all perchlorates are soluble in water.
P.60 Group VII A: Halogen Group 391

ltubidium Perchlorate-Permanganate (3.0-6.0). The test seems to be

specific for perchlorate.
Evaporate, if necessary, and then dissolve enough material to get a
0.1 to 1 % solution of perchlorate, which may be acid or neutral. Spread
1 ,ttl of it to an elliptical drop on a microscope slide. Introduce into one
end of the drop a.mere speck ofKMn0 4 and into the other, a grain of RbOl.
Mixed crystals of rubidium perchlorate and permanganate separate as
orthorhombic prisms, bipyramids, and combinations, and the color deepens
from pink to deep red depending upon the permanganate content.
Test with Methylene Blue and Zinc Sulfate (3.3-4.6). Only persulfate
gives the same test.
Treat 20 volumes of about neutral test solution with 5 volumes of
saturated zinc sulfate solution, 5 volumes of 20 % (2-F) NaN0 2, and 1 to
3 volumes of 0.03 % aqueous methylene blue. Mix. A reddish violet
coloration indicates perchlorate.
Ignition of Salts. At high temperature, chlorides are formed and oxygen
is given off.

No. 35: Bromine, 79.909

Bromine is a brown red, heavy (D = 3.12 gjml) liquid that boils at
58.8 0 0 and gives off heavy brown fumes at room temperature. The
characteristic odor resembles that of chlorine, and 1 volume of bromine
in 100000 volumes of air has still an irritating effect upon the mucuous
membrane of the nose. From the aqueous solution, bromine may be
extracted with OS2 or 0014 ; the extract is yellow to dark brown depending
upon the amount of bromine.
Flame Test. See Ohlorine, above.
Eosine (5.0-5.0 if the bromine is liberated from aqueous solution).
Use procedure and reagent (a) described above under Elemental Ohlorine.
2,4,6-Dibromo-l,3-diaminobenzene (4.0-4.0 if the bromine is liberated
from aqueous solution). The other halogens do not interfere.
Expose a droplet of 5 % m-phenylenediamine in 0.05-F H 2S04 to the
gas or vapor to be tested. If the droplet is suspended from a slide, micro-
scopic examination will not require a transfer. Bromine gives fine, colorless
needles which often combine to form sheaves or stars.
Hydrobromic Acid and Bromides
Hydrogen bromide is a heavy, colorless gas of suffocating odor which,
like HOI, gives heavy white clouds when getting in contact with NHa vapors.
It is very soluble in water. The solution, hydrobromic acid, turns soon
brown because of the action of atmospheric oxygen.
4 HBr +O 2 - 2 H 20 + Br 2•
392 Confirmatory Tests P.60

The bromides behave similarly to the chlorides. Adding chlorine water,

liberates bromine which may be extracted with CS 2 or CC1 4 or identified
with fluorescein or m-phenylenediamine as described above.
AgBr. Silver nitrate precipitates from neutral and acid solutions of
soluble bromides pale yellow AgBr which is less soluble than AgCl. It
dissolves only slightly in NH3 and gives crystals resembling those of AgCl
on evaporation of the NH 3.
Liberation of Hydrogen Bromide. Similar to chlorides, most solid
bromides give HBr when warmed with concentrated H 3P0 4 • If concentrated
H 2S0 4 is used, some bromine forms and the gas appears brown.
2 HBr + H 2S0 4 -+ Br2 + S02 + 2 H 20.
The liberated gas may be tested with AgN0 3 solution, see P. 37.
Oxidation to Bromine. L. I. = 0.5 fig Br. Cyanide interferes with the test.
Place the test substance or solution into a I-ml porcelain crucible.
Treat acid solutions with small portions of 6-F NaOH until they are alkaline.
Evaporate to dryness. If much iodide is present, add NaN0 2 and 3-F
acetic acid, and again evaporate to dryness. Dissolve the residue in 3-F HN0 3
and add a small crystal of KMnO,. Stir until the solution assumes the
color of permanganate. Cover the crucible with a slide having on the
underside a droplet of m-phenylenediamine solution, see above, or a little
square of filterpaper moistened with fluorescein solution (a), see under
Elemental Chlorine. Heat the bottom of the crucible to about 50° C with
a rnicroflame. The presence of bromide is proven if both bromine tests
are positive. Needles with the phenylenediamine but no red coloration
with fluorescein would indicate cyanide. Fluoride, chloride, and small
amounts of iodide do not interfere. Sulfide, sulfite, thiosulfate, and thio-
cyanate are destroyed by the permanganate.
Bromic Acid and Bromates
HBr0 3 is only known in the form of solutions which are colorless,
odorless, and strongly oxidizing. All normal bromates are more or less
soluble in water.
On adding AgN0 3 to not too dilute bromate solution (not less than 1 %),
AgBr0 3 crystallizes as colorless crosses and short, pointed, tetragonal
prisms. The crystals are moderately ~irefringent and appear almost black
in transmitted light because of their high refractive index.
All bromates decompose upon ignition. The alkali and alkaline earth
bromates give oxygen and bromide. The salts of heavy metals give oxygen
and bromine.
When a bromate solution is acidified with dilute H 2 S0 4 , the mixture
remains colorless for some time. As soon as the gradual decomposition
has furnished enough HBr to start the reaction 5 HBr +
HBr0 3 -+ 3 Br2 +
P.60 Group VII A: Halogen Group 393

+ 3 H 20, the solution turns suddenly brown and added CS 2 extracts bromine.
If bromide is present in the test solution, the separation of bromine takes
place immediately upon acidifying. Hence, if acidifying the cold test
solution does not give a yellow color to a drop of CS 2 present, but the
color appears immediately on adding bromide, this proves the presence
of bromate, especially if no other oxidant can be found. Furthermore,
if a bromide solution is treated with a droplet of CS 2 and dilute H 2 S0 4 ,
bromate must be absent if bromine does not separate immediately (40).

No. 53: Iodine, 126.9044

Iodine forms bluish black scales that melt at 114 0 C and begin to
vaporize already at room temperature, b. pt. = 184 0 C. The vapor is
violet and has a characteristic odor.
Iodine is slightly soluble in water and more soluble in KI solutions
because of the formation of triiodide ion 13 -. As a general rule, to which
there are exceptions, solvents containing oxygen dissolve iodine with a
yellow to brown color; solvents containing no oxygen, with a violet color.
To increase the sensitivity of the test, one extracts with a small drop of
solvent, CHCI3 ; absence of iodide improves the sensitivity.
A very delicate test for free iodine may be based upon the blue color
of the absorption compound of starch and triiodide ion since only one
starch grain need be exposed to show the color under the microscope.
The presence of iodide is necessary and improves the sensitivity of this test.
H ydriodic Acid and Iodides
Hydrogen iodide is a heavy, colorless gas of suffocating odor, that is
readily soluble in water. The aqueous solution is usually brown because
of the dissolved iodine.
4 HI + O2 ->- 2 12 + 2 H 20.
The iodides of the alkalies and the alkaline earths are colorless and
soluble in water. The iodides of the heavy metals are usually colored
(yellow to red); some of them are insoluble in water but readily dissolved
by excess iodide with the formation of soluble complexes.
Iodides are readily oxidized to iodine.
Flame Test: see above under Chlorine.
Silver Iodide. AgNOs precipitates from neutral and acid solutions
the iodide AgI which is practically insoluble in ammonia. The pure silver
iodide is pale yellow; a distinctly deep yellow color indicates that either
bromide or chloride are present and that traces of them have been co-
precipitated.
Dilute Sulfuric Acid. Iodine may be liberated at room temperature or
upon warming.
394 Confirmatory Tests P.60

Concentrated Acid. Concentrated H 2S0 4 oxidizes iodide to iodine,

whereby the sulfuric acid is reduced to S02' S, or H 2S depending upon
the amount of iodide present. Use of concentrated H 3P0 4 gives HI which
may be recognized by its action on AgN0 3 solution. See also P. 37.
Oxidation with Chlorine. Cyanide interferes.
In a test tube, treat 0.5 ml of the neutral or slightly acid test solution
with a drop of CS 2 and then add 1 drop of chlorine water and mix. In
presence of iodide, the carbon disulfide becomes violet. Continue the
dropwise addition of chlorine water with shaking after each addition.
The iodide is oxidized and the carbon disulfide becomes colorless (or
yellow if bromide is present also).
12 + 6 H 0 + 5 Cl
2 2 ~ 2 HI0 3 + 10 HCI.
Oxidation with Nitrite (5.8-6.1). Bromide and reducing agents interfere.
In a test tube, treat 0.5 ml test solution with 1 drop CCI 4 , 1 drop 0.5-F
H 2 S0 4 , and 1 drop O.l-F KN0 2 • Agitate thoroughly for 10 seconds. Iodine
is recognized by the violet to brown color of the solvent.
Palladous Iodide (4.7-6.0; fluoride, chloride, chlorate, perchlorate,
bromide, bromate, iodate, periodate: 2.0). Hypochlorite, hypobromite,
sulfide, cyanide, and thiocyanate interfere and must be absent; azide,
ferro- and ferricyanide interfere when present in excess.
On spot test paper treat 1 drop of test solution with 1 drop of 1 %
aqueous PdCl 2 solution. PdI 2 precipitates and gives a dark brown spot.
Also the precipitation of the very characteristic PbI 2 may be tried,
see Expt.33.
Iodie Acid and Iodates
HI0 3 is a white crystalline solid that can be dehydrated by slow heating
to llO° C. The white 120 5 formed decomposes at about 300° C to 12 and O2 •
The acid is readily soluble in water, and the saturated solution (at 20° C)
is about 14-F HI0 3 •
The alkali iodates are soluble in water; the other iodates are difficultly
soluble to insoluble, which distinguishes iodate from chlorate and bromate.
Action of Sulfuric Acid. Neither dilute nor concentrated sulfuric acid
decompose iodate if reducing substances are absent.
Silver Iodate. AgN0 3 precipitates white, curdy AgI0 3 which is difficultly
soluble in HN0 3 , but dissolves readily in NH 3 • From solutions that are
made slightly acid with HN0 3 , AgI0 3 may be obtained as pointed needles,
single and in radial clusters, which show strong birefringence and parallel
extinction. When performed upon the slide, L. 1. about = 1 #g.
Barium Iodate. The white precipitate of Ba(I0 3)2 dissolves only slowly
in dilute acids. When precipitated upon the slide, separation of curved
needles seems characteristic for the iodate, but also prisms, rhombs, crosses
are obtained, which are similar to the crystals of the more soluble Ba(BrOah.
P.61 Separations 395

Reduction to Iodine (3.9-5.4).

Place upon starch paper 1 drop of 5 % (0.5-F) KCNS solution and add
a drop of acid test solution. A blue stain indicates iodate (121).
6 10a- + 6 H+ + 5 CNS- + 2 H 0 2 -'>' 5 HS0 4 - + 5 HCN + 3 I 2•

Ignition of the Salts. All iodates are decomposed on ignition. Deflagra-

tion occurs on charcoal or in the presence of other oxidizable matter. This
is similar to the behavior of chlorates and bromates.
Periodic Acid and Periodates
The periodic acids, HI0 4 and HI0 4 • 2 H 20 or H sI0 6 , are described
as colorless, crystalline solids that are soluble in water and decompose at
about 120 0 C.
Most periodates are only slightly soluble in water, but they are readily
soluble in dilute HNO a• The composition of the precipitate obtained with
AgNO a depends upon the acidity of the solution; the color varies from
yellow to orange, red, and gray. Boiling the silver precipitate with water
causes it to become dark red.
Periodates are powerful oxidants. A reaction that is not given by
iodate is the oxidation of manganous salt to permanganate which may
be performed as follows.
Mix equal volumes of 2-F acetic acid and 10 % (0.5-F) solution of MnCI 2 •
On a spot plate, mix a drop of the mixture with a drop of the neutral test
solution. Appearance of the color of permanganate indicates the presence
of periodate. Chlorates, bromates, and iodates do not interfere with this
test. Persulfate should be absent. To intensify the color effect, the acetic
acid used for the reagent mixture may be saturated with "tetrabase" (121)
which is converted to a blue dye by the permanganate. In this instance,
the sensitivity is improved to (5.0-6.3).

P.61 Separations
If the orientation tests have shown that the material is of a complex
nature, i. e., probably contains several elements (cations or anions), the
use of a scheme of separation is indicated. If only little material is available,
the question arises whether or not a qualitative analysis with approximate
estimation of the quantities will be able to solve the problem; a quantitative
analysis or a series of quantitative determinations may better serve the
purpose.
Even with relatively simple substances, use of a scheme of separation
may become necessary if the orientation tests fail to give sufficient evidence
for definite identification. The performance will not be tedious if only
a few elements or radicals are present; only few analytical groups will
be represented, and these will contain only one or two members so that
further separation may not be necessary.
396 Separations P.62, 63

P.62 Systematic Schemes for the Detection of Cations

Adherence, in principle, to the classical hydrogen sulfide scheme is
recommended. For this scheme, the behavior of all elements is best known,
an advantage that should not be dismissed if the presence of uncommon
elements is possible. In addition, the hydrogen sulfide scheme permits
the use of gaseous reagents for the separation of the five principal groups.
These reagents may be obtained free from metallic impurities and they
may be again removed from the reaction mixture by the simple expedient
of evaporation and ignition. Finally, the very carefully tested scheme of
NOYES and BRAY (49), which includes the less common elements, is
essentially an extension of the hydrogen sulfide scheme.
There may be good and sufficient reason to eliminate rare elements
from consideration or to consider only a certain group of elements. In
such instances some other scheme of separation may be preferable. SWIFT'S
version of the classical procedure (53) permits semiquantitative estimation;
the scheme of WEST and MUKHERJI (473), the outline of which is readily
accessible, uses liquid-liquid extraction for the separation of groups which
are rather small and quite different from those of the classical scheme.
Whatever procedure may be used, the experimenter should keep in
mind that there is no scheme in existence-and probably never will be-that
has been tested for all possible combinations and all possible ratios. Even
the most carefully tested schemes may fail to reveal the "whole truth"
with certain materials even when the operator is on the lookout for signs.
of abnormal behavior. Ammonia and ammonium ion are not considered
by the schemes and must be found in a separate test, Expt. 46.

P.63 The Classical Scheme

The classical scheme is applied to the neutral or acid solution of the
material. Such a solution will fail to contain niobium and tantalum so
that these elements need not be considered. Wolfram may be present
if the solution contains arsenic or phosphorus.
The classical scheme is recommended if it is simple to obtain a neutral
or acid solution of the material under investigation or when only the
common metals need be considered and complete dissolution may be
obtained by means of a simple fusion procedure. The scheme of NOYES
and BRAY (49) is recommended if the material refuses to dis~:!Olve without
leaving a residue; it is also recommended for the safe detection of arsenic,
selenium, germanium, osmium, and ruthenium. For the analysis of complex
silicates, expert advice will be found in Applied Inorganic Analysis (7).
Characteristic for the classical scheme is the use of hydrogen sulfide,
ammonium sulfide, and ammonium carbonate for the separation of the
analytical groups. There are many modifications of it, but none of them
P.64 Systematic Schemes for the Detection of Cations 397

describes the reliable separation of all metallic elements, common and rare.
Consequently it becomes the task of the analyst to recognize the presence
of odd elements from unusual phenomena observed in the course of separa-
tion. This implies that he is familiar with the behavior of the constituents
considered by the scheme. The ability of correctly interpreting descriptions
of the phenomena is only a poor substitute for this familiarity unless the
precaution is taken to compare the questionable phenomenon with that
obtained with a material that has been prepared to have the composition
assumed for the fraction under investigation.
The following outline in the form of a flow sheet gives the key numbers
under which the directions and the observations (notes) are given.

P.64 Outline for the Separation of the Analytical Groups

Acidic Solution of the Sample
+ HCI~~---- -----------+ PPt. 1: Silver Group.
1 AgCI, Hg 2C12, PbCI2, TICI
Supernate 1:
+H 2 S - - - - Ppt. 2: H 2S Group.
I Extracted with (NH'.)2S",
i
I

1
Residue 3: Cu Group.
1
Extract 3: As Group.
Black: CuS, HgS, PbS, Black: (CuS), MoS3•
Bi2Sa, Re2 S 7 (!), RuS"" Re 2S 7, RuS"" IrsSa.
Rh 2 Sa, PdS, OSS4' PtS 2, Au 2 S2 ;
PtS 2; brown: SnS, Te;
golden: Au; orange: Sb 2S3.5, Se;
yellow or orange: CdS, yellow: AssS3.5• SnSs'
(Se) Se;
white: S. GeS s
Supernate 2:
+ NH3 + H ,S 2 or (NH4)2S --- ------> Ppt. 4: (NH )2S Group.
Black: FeS, CoS. NiS, U0 2 S. TI2S;
blue: Cr(OH)3' Nd(OH)3;
green: Cr(OH)3' Pr(OH)a, MnS;
tan: MnS;
yellow: Sa(OH)3' InsS3;
white: ZnS, Ga SS3, In2Sa, Al(OH)s'
Be(OH)s' (Sc, Y, La, Ce, Gd, Tb)·
. (OH)3' (Ti, Zr, Th)· (OH) ,
~ MgNH~P04 . 6 H 20, (Ca, Sr, Ba)s' (P04)s
Supernate 4:
+ (NH.)2COS -------~--~---+ Ppt. 5: Alkaline Earths Group.
(Ca, Sr, Ba) . COs
Supernate 5:
Mg and the alkalies.
398 Separations P.6l)

P.65
Procedure
The amounts of reagents suggested in the procedure are based upon
the assumption that one gram of solid sample has been used for the prepara-
tion of the solution to be used for the separation. If n gram of solid sample
have been taken, multiply all volumes and weights mentioned in the
procedure by n.
Sample Solution and Treatment with HCl. The sample solution is
prepared according to the experiences made in testing the solubility of
the material under investigation. The solution must be neutral or acidic,
and its volume should be between 10 ml and 50 ml. Prolonged boiling
with HOI or HBr may cause loss of As, Sb, Hg, Se, and Ge, which may
be prevented by using an efficient ref1ux condenser.
One might assume that a solution containing HOI will not contain
silver; since, however, AgOl is somewhat soluble in strong HOI (or HBr),
a side test should be made with such solutions to see whether or not diluting
with water gives a slight turbidity of silver halide (a heavy precipitate
upon diluting suggests the presence of Sb or Bi).
If HOI (HBr) has not been used in preparing the sample solution, it
is suggested to first add just one drop (0.05 ml) of 1-F HOI. A white
precipitate or turbidity suggests the presence of silver. If the solution
remains clear, heat for five minutes in a steam bath and then again cool
to room temperature; if no turbidity appears, silver is obviously absent.
Whatever the effects of the first drop may be, finally add 3 ml 12-F HOI
and mix.
Precipitation of the Hydrogen Sulfide Group. If the unknown has been
dissolved in nitric acid, it may be best to remove the excess of HN0 3 by
evaporation just to dryness and then adding 10 m16-F HOI before proceeding
to treatment with H 2S. Of course, HN0 3 and other oxidants interfering
with the H 2S treatment might be destroyed by addition of a suitable
reductant: S02' N 2H4HOl, or NH 20H· HOI.
As a rule, the supernate I from the silver group is strongly acid
corresponding to 3-F HOI or stronger. Saturate the strongly acid solution
at room temperature with H 2 S; heat to about 95° 0 (steam bath); saturate
with H 2 S while the mixture is cooling; dilute with 50 ml water; saturate
with H 2 S at room temperature; heat to about 95° 0; dilute with another
50 ml water and again saturate with H 2 S at room temperature; finally
repeat diluting with 50 ml water and saturating with gas until the hydrogen
ion concentration in the solution has dropped to 0.3-F (pH = 0.5). This
state is most conveniently established by adjusting the volume of the
mixture; if the latter contains a total of m mval (milliequivalents) of acid
P.65 Systematic Schemes for the Detection of Cations 399

(added in dissolution and precipitation of the silver group), the volume

should be approximately = m/0.3 mI, usually about 200 ml.
Flocculation indicates that precipitation is complete at the prevailing
acidity; more sulfide may come down when the mixture is further diluted,
but dilution should not go beyond pH 0.5 since this would lead to precipita-
tion of ZnS.
Test for complete precipitation by saturating the supernate or a sample
of it with H 2S and heating to about 95 ° C. There should be no more
precipitate.
! Wash the precipitate with three 10-ml portions of O.I-F HCI that has
been saturated with H 2S at room temperature. Combine the washings
with supernate 2.
Separation of the Copper and Arsenic Groups. Treat Ppt. 2 with 10 ml
6-F (NH4)2S1.2*; with occasional stirring digest for five minutes at 40° C;
separate the extract from the residue; wash residue 3 with three 10-mI
portions of 2-F (NH,)2S1.2 and reject the washings.
To obtain the sulfides of the arsenic group from extract 3, first dilute
with 30 ml water and then add with stirring (about 30 mI) 2-F H 2S04,
about 5 ml at a time, until the mixture is distinctly acid (pH 3 to 1; acid
to Congo red). Remove and reject the liquid, and wash the precipitate
once with 0.I-FH 2 S0 4 that has been saturated withH 2S at room temperature.
Precipitation of the Ammonium Sulfide Group. Treat supernate 2 with
2-ml portions of 3-F NHs (about 5 portions should suffice) until ammoniacal
as indicated by pH paper and odor. Heat to about 60° C and, with stirring,
slowly add 6 ml 2-F (NH4)2S, (Alternative: treat the ammoniacal mixture
with 8 ml 3-F NHa; saturate with H 2 S; warm to 60° C and add another
8 ml 3-F NH s.) Separate solution and precipitate, and wash the latter
with two 10-ml portions of O.I-F (NH4)2S, Combine the washings with
supernate 4.
Analysis of Supernate 4. Treat the supernate with 5-ml portions of
2-F HCI with mixing until acid. Separate from the precipitate which,
in addition to S, may contain NiS, V2S0 , and (or) WS s. On the steambath,
evaporate the clear solution to about 10 ml. Then add 5 ml 16-F HNO a
and evaporate to dryness. Finally heat gently (direct flame or heating
block) to vaporize any ammonium salt that might be left behind. After
cooling to room temperature, dissolve the residue in 10 mll-F HCI. Separate
from any insoluble matter and treat the clear solution with 2 ml 6-F NHa
and 20 ml I-F (NH4)2COa' Heat just to boiling, then allow to cool to room
temperature. Separate the solution from the precipitate, and wash the
latter once with 5 ml water. The washing may be combined with supernate 5.
* Saturate 50 mll2-F NH3 with H 2S, add 50 ml more 12-F NH3, and dissolve
0.6 g powdered sulfur.
400 Separations P.66

Analysis of Supernate o. A side test for magnesium may be carried

out with a small part of supernate 5. Before testing for alkalies, however,
it i8 preferable to remove the ammonium salts. To this end treat supernate 5
with 6 ml 12-F HCI and 3 ml 16-F HNOa• Evaporate to dryness on the
steam bath, and gently ignite the residue for the removal of any ammonium
salt left. After cooling, estimate the amount of residue left and then
dissolve it in the suitable volume of the desired solvent.

P.66 Observations and Notes

Precipitate 1: Silver Group. Silver chloride is somewhat soluble in
strong HCI, and very small amounts of silver may be missed when adding
a large amount of HCl at the start. On the other hand, silver and mercurous
mercury may be completely separated from lead by adding small portions
of dilute HCI until these ions are precipitated completely, whereas the
solubility product of lead has not yet been reached.
AgCI (curdy) and PbCl 2 (distinctly crystalline) are usually recognized
by their appearance.
Supernate 1. If barium and strontium are of interest, a tenth of
supernate 1 may be tested with the dropwise addition of 2-F H 2SO,.
A white precipitate may be BaSO" SI'SO" PbSO" or CaSO,' 2 H 20.
After recording the volume (compari80n with a known amount of BaSO,),
the precipitate is isolated, washed, and tested accordingly. PbSO, may
be extracted with a solution of ammonium acetate.
If thallium may be present, it is recommended to saturate supernate 1
with S02' Reduction to thallous ion is followed by precipitation of TICI
which separates as a fine, crystalline powder. Also reduced are permanganate,
chromate, and ferric ion as may be observed from the color changes.
BaSO" etc. may precipitate since SO, is formed. Before adding H 2S, the
excess of S02 is removed from the solution by boiling and bubbling CO 2
or air through it.
Precipitate 2: Hydrogen SuHide Group. The presence of oxidants
(S02) in supernate 1 will delay precipitation of sulfides since the H 2 S is
destroyed. Large amounts of white sulfur may separate, and the color of
the solution may change because of the reduction of colored ions (preceding
paragraph). Oxidation to sulfate may also cause the separation of sulfates
of the alkaline earths. It is undesirable to have so much strong oxidant
present that a violent reaction with liberation of heat and separation of
molten or plastic sulfur ensues. Normally, sulfur will appear, depending
upon its amount, as a turbidity or as a white milk which passes through
fi1ter paper and which may have to be boiled for some time to obtain
flocculation.
Some authors adjust the acidity by adding ammonia. This mURt not
be done, however, when the solution once contains HliS. Regardless of
P.66 Systematic Schemes fo~ the Detection of Cations 401

how efficient the mixing is, local excess of ammonia and alkaline reaction
will last long enough so that CoS and NiS precipitate locally. When the
solution is again uniformly mixed and acid throughout, these sulfides
may not dissolve again and remain in Ppt. 2.
The sulfides separate in the order of their solubilities. First separate
the sulfides of mercury and arsenic (yellow); mercury may first give a
dense white precipitate of Hg 2SCl 2 (1) which with more H 2S gives first
distinctly yellow HgsS2Cl 2 (1) that then changes through brown to black HgS.
The sulfides of mercury and arsenic separate even from 12-F HCI; the
orange sulfide of antimony (which may turn grayish black) is not precipitated
until the concentration of the acid is below 6-F. The black sulfides of the
common heavy metals separate from still less acid solutions, and the
yellow (from acid solutions more often orange) CdS will not appear before
the acid concentration has dropped to 0.5-F.
Lead "sulfide" may separate with a dull red color from solutions
containing much chloride.
The black MoSs is very insoluble but does not precipitate readily.
It is characteristic that saturating with H 2 S followed by heating gives
some black precipitate and a bluish supernate (probably a colloidal suspension
of MoSs). Repeated treatments with H 2S produce additional portions of
precipitate. To obtain complete precipitation, the solution must be saturated
with H 2 S and heated (steam bath) in a sealed vessel to prevent escape
of the gas. Similar trouble give platinum metals, of which 11' is most
difficult to precipitate. In the instance of ruthenium, the solution becomes
sky blue (sensitive and characteristic test) before the separation of the
brown sulfide starts.
Brown metallic gold is precipitated from hot solutions.
Residue 3: Copper Group. The residue will contain some white sulfur
and it may hold sulfates of the alkaline earths.
If it is intended to separate HgS from the other members of the group
by extraction with nitric acid, residue 3 must be washed free of chloride
ion which otherwise would react with the acid to give chlorine that dissolves
the HgS.
For dissolution in HNO s, residue 3 should be thoroughly mixed with
cold dilute acid and then the mixture warmed with continuous stirring
until dissolution occurs. This procedure favors the liberation of H 2S
without much oxidation of it and avoids the separation of plastic sulfur
that could envelop sulfides and later prevent their dissolution. The residual
HgS may become converted to white HgS(NOs)2 giving the impression
that only sulfur is left behind (mixed possibly with the sulfates of the
alkaline earths). It is suggested to dissolve the HgS (or the white mixed
sulfide) in a small volume of sulfide-hydroxide reagent (solution 2-F in
Benedetti-Pichler, Identification 26
402 Separations P.66

Na 2S and I-F in NaOH); this may leave a residue of CdS and (or) ZnS
(which give solid solutions in HgS) and also Au, PtS 2, RuS OIl , and Rh 2 Ss'
The sulfides of Os and Pd dissolve in HNO s'
Precipitate 4: Ammonium Sulfide Group. If the ammonia or the
ammonium sulfide contains carbonate or the precipitation is exposed to
the laboratory air for a significant time, alkaline earths will be precipitated
as carbonates and accompany this group even in the absence of phosphate.
To obtain subgroups, Ppt. 4 may be treated with 20 to 60 m] 0.5-F
ammonium citrate solution. The hydroxides and phosphates dissolve,
and the sulfides remain insoluble.
Supernate 4. If supernate 4 is brown, it probably contains some nickel
sulfide in colloidal dispersion. A brilliant violet red color indicates vanadium.
When the supernate is acidified, the following may separate: some white
sulfur, a small amount of black NiS, black V 2S4 or V 2 S5 , and light brown
WS a. The precipitation of V and W is expected at this point since they
are precipitated by neither H 2S nor (NH4)2S, but separate when the
solution is treated with the latter and then acidified.
Heating with HNO a and HCl destroys the ammonium salts so that
very little solid is left when the solution goes to dryness. If HNO a is applied
in excess (3 HCI + HNO a - 2 H 20 + NOCI + CI 2), the residue consists
mostly of nitrates which are not inclined to decrepitation when heated.
Precipitate 5: Alkaline Earths. For various reasons indicated above,
the alkaline earths may be carried down with ppts. 2 and 4 so that only
little or nothing of them may appear in this p]ace.
Supernate 5: Magnesium and the Alkalies. Magnesium need not interfere
when testing for Na and Li, but ammonium salts must be removed before
testing for K, Rb, and Cs.
Satisfactory proof for the presence of small amounts of potassium or
sodium is difficult since they may be derived from the reagents and
apparatus used in the preceding operations. They also may be lost by
coprecipitation, especially with precipitate 4. A simplified procedure is
desirable for the safe detection of alkalies.
If the material is only partly soluble in water, it is recommended to
separately analyze the aqueous extract. Otherwise the separation may
be simplified by treating supernate 2 with I ml 9-F H 2S0 4, evaporating,
and finally heating (300 to 400 0 C) until fumes of SOa are no longer given off.
The residue is extracted with 20 ml hot water, and the clear extract is
treated with a saturated solution of Ba(OH)2 until precipitation is complete
(one drop may suffice, but not more than 70 ml should be required).
After separation, the clear solution is precipitated with (NH4)2COa like
supernate 4 to remove barium, whereafter the filtrate (centrifugate) is
evaporated and ignited for the removal of ammonium salts. The clear
P.67 Systematic Schemes for the Detection of Cations 403

aqueous extract of the residue may contain the alkalies as carbonates

and will have to be acidified or neutralized depending upon the requirements
of the tests.
Insoluble materials (silicates) are best treated according to J. LAWRENOE
SMITH as described on p. 925 of Applied Inorganic Analysis (49). The
procedure of BERZELIUS-KRISHNAYYA, p. 930 (49), is preferable if significant
amounts of boric acid are present.
Whatever procedure is used, the purity of all reagents must be ~watched.
Apparatus is best of platinum, plastic, or clear vitreous silica.

P.67 AnalY8i8 of Metal8 and Alloy8 Attacked by Nitric Acid

Treatment with nitric acid is by no means obligatory. Some metals
and alloys do not dissolve in HNO a• Aluminum may be advantageously
dissolved in NaOH solution, whereby copper, iron, etc. remain as metals
in the residue.
In general, the treatment with HNO a has the advantage that the
"anion" of suHides, phosphides, arsenides, antimonides, and silicides is
oxidized to a corresponding oxygen acid instead of being lost as H 2 S,
AsH a, etc. as may happen when the sample is dissolved in HOI.
The following procedure is essentially that of A. A. NOYES (48).

Sample treated first with

HN0 3 and finally with HCI
!

Residue a: Si0 2, Si,

graphite, AgCl, PbCI 2 •
Treated with H2F2 and H 2 S04 Solution b Solution a
I1____ combined
_ _ _ _ _ _ _,

.j,
Residue b: PbS0 4
Precipitate c: Supernate c:
AgCI cations and P04

Procedure
Dissolution of Sample. Treat 0.5 g of filings, borings, or turnings
with 5 ml 6-F HNO a and heat on the steam bath until the reaction ceases.
If some metal remains undissolved, add I ml 16-F HNO a and continue
heating. Finally evaporate to 2 ml. Add 5 ml 12-F HOI and evaporate
slowly just to dryness. To render insoluble most of the silicic acid, heat
the residue to 120 a in a drying oven or heating block. Using a stirring
0

rod and a pestle, loosen the residue from the walls of the apparatus and
26·
404 Separations P.68
rub it to a fine powder. Add 5 ml 6-F HOI, cover, and allow to digest for
five minutes at 80° O. (If dissolution does not seem complete, add 2 ml
12-F HOI and evaporate slowly almost to dryness. Again treat with 5 ml
6-F HOI, as before.) Add 10 ml water, heat to just boiling, and separate
the solution from any residue while still boiling hot. Wash any residue
with 5 to 10 ml 2-F HOI and then with five 5-ml portions of hot water.
Reject all washings. Reserve solution a.
Residue a. Graphite will be recognized by its color; see also P.57
under carbon and silicon.
Estimate the amount of precipitate and then use the whole residue
for a test for silica, P. 57. When working on the gram scale, use a platinum
crucible of at least 1O-ml capacity. Use 2 ml water to transfer the bulk
of residue a to the platinum crucible. Evaporate the slurry to drYness
and then add 1 ml 18-F H 2 S0 4 and 0.3 ml (6 drops) 48% H 2F 2 • Test for
silica as outlined in P. 57. If the test is positive, add 2 ml 48 % H2F 2,
cover the crucible, and digest 15 minutes on the steam bath. Regardless
whether or not the test is positive, evaporate the mixture until the heavy
white fumes of S03 appear. Allow to cool and then pour the contents of
the crucible into 30 ml water (adding water to the acid is permissible
when working upon the microscale). Mix and boil gently for five minutes.
After cooling to room temperature, remove solution b and wash any
residue b with 0.5-F H 2 S04 ; reject the washings (Ag 2 S0 4 may crystallize
out when the mixture is finally cooled; inspection under the microscope
will identify it).
Residue b may be PbS0 4 and (or) carbon (graphite).
Solution b. Add to solution a. A precipitate of AgOI may separate.
Supernate c contains phosphate and the cations. Proceed as with the
supernate 1 of the classical scheme, P.64.

P.68 Separation 01 the Analytical Groups 01 Noyes and Bray

The procedure of A. A. NOYES and W. O. BRAY (49) is recommended
if (1) As, Ge, Se, Os, Ru may be present or are of special interest; (2) the
material resists dissolution; (3) the material is very complex in composition;
or (4) presence of the less common elements is probable and these are of
interest.
A complete outline of the scheme of NOYES and BRAY with emphasis
on later modifications suited for work on the milligram scale may be
found in Anorganische qualitative Mikroanalyse (162), and the numbering
used there for residues, extracts, solutions, and precipitates has been
adopted also for the following outline to facilitate finding the directions
for intergroup separations whenever they should be needed.
P.68 Systematic Schemes for the Detection of Cations 405

The following description assumes that 1 g of solid material (0.5 g of

metal or alloy) is taken for analysis. If n gram (0.5 n gram) is taken instead,
all weights and volumes of the directionsl are to be multiplied by n.
The scheme of NOYES and BRAY has been developed with the assumption
that the sample for analysis will not contain more than 0.5 g of any common
element and not more than 0.2 g of any rare element. Under these conditions,
it should be possible to detect 0.001 g of any element; frequently even
0.0001 g may be discovered, but in a few instances the limit is 0.003 g.
This implies limiting proportions of from 1 : 170 to 1 : 5000 (usually 1 : 500)
if a common element is present in the maximum amount of 0.5 g and of
from 1: 70 to 1: 2000 (usually 1: 200) if a rare element predominates
to the limit of 0.2 g.
The reasons for the selection of the methods of separation may be
found in the original papers of NOYES and BRAY (49), CROWELL et al. (420,
883), and BENEDETTI-PICHLER et al. (413, 415, 421, 866, 869, 870).
The material for analysis should be reduced to a fine powder to facilitate
attack of the solvents. In the instance of metals and alloys, borings and
turnings may be preferable to the finer filings since the last may be more
seriously contaminated by material from the tool. Of course, any lubricant
should be removed by washing with solvent and drying; steel particles
from tools may be removed from the dry material with a magnet, provided
that the sample itself is not attracted.
NOYES and BRAY separate into 16 analytical groups: I - selenium group,
II - wolfram group, III - tantalum group, IV - gold group, V - thallium
group, VI - tellurium group, VII - copper group, VIII - ammonium
hydroxide group, IX - ammonium sulfide group, X - nickel group,
XI - zirconium group, XII - aluminum group, XIII - chromium group,
XIV - lanthanide group, XV - alkaline earths group, and XVI - alkali
group. Fractions of groups VIII and IX are combined to give groups X
to XIV.
Aside from one exception (tantalum group), the intergroup separations
are not included in this outline since one may hope that it may be possible
to obtain confirmatory tests without preceding separations if only one or
two members of the group are present.
A glance at the flow sheets may be discouraging, but it should be
realized that even with complex materials not more than half of the
analytical groups will be represented; with relatively simple compounds,
only a few precipitates will be obtained, and most of the solutions will
be empty.

1 Anorganische qualitative Mikroanalyse gives ge (Gewichtseinheit = unit of

weight) and ve (Volumeinheit = unit of volume) which may be translated as
g (gram) and m!.
406 Separations P.68

Outline 01 Separation into Analytical Groups

Solid Sample + HBr + Br2
____ 1
1
---~----l

Residue AI: insolubles Extract AI: acid

+ H2F2 and heat, -+ SiF4 t BFa t soluble material
+ HOI0 4 and evaporate, -+ HF t I

1 I

Residue A 3: add to extract A 1

L ___
~
Residue A 3 + Extract A 1 distilled
I

1
Residue A 4: solution and undissolved
1
Distillate A 4:
matter + HNO a and distil - Selenium Group,
Se, Ge, As, (Sb, Sn)

Distillate A 5: OS04
t
Residue A 5 + HOI0 4 and distil Distillate A 6: Ru0 4

j 1----..
,
Solution A 7. Evaporate
to a small volume and

--------------
I
Residue A 6. Reflux with formic acid i cool
I ---~l

Orystallizate A 9: Supernate A 9: groups

I perchlorates of K, Rb, Os V to XVI and P0 4
~
Residue A 7. Heat with H2F2 ---- Extract A 8. Heat with
H 2 S04 to fumes of SOa'
add H 20, NH 3 , and
(NH 4 hS and heat in
pressure bottle
----~1
I Precipitate III:
t Tantalum Group,
Solution II 1. Ta 20 5, Nb 20 5, Ti0 2 ;
Add dilute H 2S0 4 Bi2Sa ; phosphates and

1---
I vanadates of Ti and Zr

-- ---- - Precipitate II 2:
t Wolfram Group,
Supernate II 2: WS a, MoS 3 , V2S 5 , TeSs,
P0 4, (W0 4 ) Sb 2Sa, SnS 2

Residue A 8
P.68 Systematic Schemes for the Detection of Cations 407

Outline Oontinued
Residue A 8: noble metals, AgBr, S,
fluorides of Ca, Th. Pb, Bi. Boil with
Na 2COa - - - - - - - - - - - - - > , Extract AI0: F, S04

1
Residue A 10: Acidify with HCIO 4 -.~..- - - - - - > Extract A 11 :
perchlorates of Ba, Sr,
Ca, Pb, (Cr)

Residue All HNO a + HCI ~---. -. -~-~ Extract A 12:

Gold Group, Au,
Hg, Pt, Pd, (Ir, Rh)

Residue A 12 extracted with NH3 . -~----.--.-.--7 Extract A 13 :

Ag(NH a)2-

Residue A 13: C, SiC, A1 20 3, Cr 20 3,

FeO . Cr20 3, Ti0 2, Sn0 2, Zr0 2 · Si0 2,
phosphates of Th and the lanthanides,
ThF~, some silicates, Pt metals and
alloys, alloys of Fe~Ni-Cr-Mo-W-Si.
Fused with K 2S20 7 and melt extracted Extract A 14: metals of
all groups
I
I
t
Residue A 14: sulfates of Ba, Sr, Ca,
Pb, Cr; oxides of Si, W, Ta, Nb, Sn;
refractive silicates, metals, and alloys.
+ H2F 2 and digested----~- --> Extract A 15: fluoride
complexes of W, Ta, Nb,
Si. It is treated like ex-
tract A 8
Residue A 15 is boiled with
Na 2C0 3 solution Extract A 16:
! F, S04 ions. Is rejected
,

i
t
Residue A 16 is extracted with HCl0 4 ~-~-.-~-+- Extract A 17: perchlor-
ates of Pb, Ba, Sr, Ca

Residue A 17: Sn0 2, Cr 2 (S04)3' some

silicates, Ir.
It is brought into solution by fusion
with Na 2 0 Z
408 Separations P.68

Treatment of Supernate A 9
Supernate A 9 + HBr - - - - -..... Precipitate VI:
Thallium Group, AgBr, TlBr,
I PbBr2

1
Supernate V 1 + H 2 S
!
Precipitates VI 1 + VI 2.
The combined sulfide ppts. are dis-
solved in HNOs + HCI, and the
solution is saturated with S02

1 j
Precipitate VI 3: Te Supernate VI 3 is extracted with
ether

----I
Ether
1
extract VI 4:
~
Aqueous Layer Vl4: Copper Group
Mo (Cu, Cd, Bi, Pb) and Ir and Rh

~
Supernate VI 2
is heated after adding ferric nitrate
and ammonium acetate ----- -- Precipitate VIn 1:
-~.....
Ammonium Hydroxide Group

J
Supernate VIn 1 + NHs + H 2S . -- - Precipitate IX 1:
Ammonium Sulfide Group

Supernate IX 1 is evaporated with

HCI and the solution of the residue
treated with (NH4)2COS Precipitate XV 1: Alkaline
Earths Group, Mg, Ca, Sr, Ba

Supernate XV 1: A1kali Group,

Li, Na, K, Rb, Cs
P.68 Systematic Schemes for the Detection of Cations 409

Resolution of the Ammonium Hydroxide and Sulfide Groups

Hydroxide Group: Ppt. V I II 1
It is dissolved in HCI and extracted with ether
~-'----- --------- -,
I
t
Water layer VIII 2 Ether extract VIII 2:
+ NaOH + Na20 2
!
FeCl3 and GaCl 3

Precipitate VIII 3:
Zr, (Ti), lanthanides, In, (Co,
Ni, Zn).
Combined with Ppt. IX 6 ---
t
Supernate VIII 3: AI, (Be), Cr, V,
W, (V), (Zn), PO •.
After adding supernate IX 4,
+NaHC03 and boiling--~ Precipitate VIII 4:
I Aluminum Group, AI, Be,
I
t Zn, (V, V)
Supernate VIII 4: Chromium
Group, Cr, V, W, V, P0 4

Sulfide Group: Ppt. IX 1

It is heated with HN03 and KCI0 3
I

Supernate IX 3 + NaOH + Na20 2 Precipitate IX 2 + 3:

Mn0 2 • X H 20

1--
I --- ._- --'.-.-------"
,~- ---.--~ Precipitate IX 4: Ni, Co,
lanthanides, (Zn) dissolved in
Supernate IX 4: U, Be, Zn, (Cr)
is added to supernate VIII 3
HCI, + acetate +
H 2S
I
1 - - - - - - - - - ------ _ _- - - 1

I I
t -I
Supernate IX 5 Precipitate IX 5:
is treated with NH3 Nickel Group, Ni, Co, (Zn)
i
1_- __________ _
j.
-~ Precipitate IX 6: lanthanides.
Supernate IX 6: After adding of Ppt. VIII 3,
(Co, Ni, Zn) is rejected + H2F2 <-~-----

Supernate VIII 5: Precipitate VIII 5:

Zirconium Group, Zr, Ti, In, Lanthanide Group, lantha-
(Zn, Co, Ni) nides, (In)
4lO Separations P.68

Analysis of the Tantalum Group

Precipitate II 1
is boiled with a solution of Na salicylate

1- i
1-
Residue III 1: Ta 20 S' Nb 20 s, Zr0 2, Extract III 1: Ti salicylate,
Zr(HP0 4 )2' (Bi 2Sa, other sulfides, V0 3 , P0 4 • After removal of
and Si0 2 from the apparatus). salicylic acid by extraction
For removal of Si0 2 and the oxida- with ether, the aqueous solu-
tion of sulfides and organic sub- tion is treated with NaOH
stance, H2F 2' HNO a, and H 2SO are
added and evaporated. The residue
is fused with K 2COa, and the melt
extracted with cold water
1------
1

1-
Precipitate III 3: Supernate III 3: VO a, P0 4,
Ti0 2 · x H 2 0, (Zr, V) (Mo, W)

Residue III 5: Zr0 2, Bi 20 a, (PbO,

1
Extract III 5: tantalate,
Fe20 a) is fused with K 2 S20 7 , and niobate, P0 4 , (Mo, W, V) is
the melt extracted with water. The diluted with water, saturated
extract is treated with H 2S with S02' and boiled
l--~----- --I I
1- J,
Precipitate III 6: Bi2S3 , (PbS, Supernate III 6:
Zr0 2, Ta 20 S' Nb 20 s, Pt from the ZrOS0 4
apparatus)
1------- - ---~--

,f ,f
Supernate III 7: P0 4 , (Mo, W, V) Precipitate III 7:
Ta 20 S and Nb 20 s

Procedure
Dissolution of Sample. Treat 1 g solid material (0.5 g of a metal or
alloy) with small portions of 9-F HBr until 10 ml have been added. Boil
10 minutes with refluxing. If dissolution is not complete, cool to room
temperature and add 0.5 ml Br 2. Heat 10 minutes on the steam bath and
add more bromine if its color should disappear. If dissolution is complete,
proceed immediately to distillation, see below. If a residue remains
undissolved, separate and wash it first with 2 ml 9-F HBr (combine with
extract A 1) and then thoroughly with hot water which is rejected.
P.68 Systematic Schemes for the Detection of Cations 411

Residue .A 1. Dry it in a drying oven or heating block. Add 2 ml

16-F HNO a and evaporate just to dryness. Add little water, stirr to obtain
a slurry, and transfer to a platinum crucible or dish. Evaporate to dryness.
Add 3 to 10 m127-F HF, cover, and digest on the steam bath for 15 minutes.
Rinse the cover into the crucible with 3 ml 9-F HCI0 4 and 2 ml 16-F HNO a.
Cautiously (infrared lamp) heat the contents of the crucible until white
fumes of HCI0 4 appear. Cover again, if necessary, and continue heating
to light fumes for 5 to 10 minutes; if necessary, add acid to keep the volume
at 2 ml. Finally evaporate to dryness and transfer the residue with 4 to
6 portions of 0.5 ml water to the distilling apparatus already containing
extract A 1.
Distillation of Extract .A 1 Combined with Residue .A 3. Place 5 ml
bromine water into the receiver and distill until about 3 ml of the charge
are left in the distilling apparatus.
Distillate .A 4. Treat with I-ml portions of 3-F NH 20H· HCl until
the solution is colorless. Then add 5 ml more of the reductant, close the
apparatus, and heat 5 minutes in the steam bath. Selenium precipitates;
the supernate is treated with a like volume of 12-F HCI and then with
H 2 S for the precipitation of the sulfides of As and Ge. The filtrate from
the sulfide precipitate should be added to the solution of the tungsten
(wolfram) group.
Distillation Residue .A 4. If Os and Ru cannot be present, add 4 ml
16-F HNO a and boil slowly until the volume is reduced to 3 ml. Add
5 ml 9-F HCI0 4 and again boil slowly until the volume is 3 ml; the boiling
of the concentrated HCI0 4 must last at least 2 minutes for the dehydration
of the acids of the wolfram and tantalum groups. Then proceed as with
residue A 6, below.
If Os or (and) Ru may be present, proceed as follows. Treat residue A 4
in the distilling apparatus with 5 ml 16-F HNO a. Place into the receiver
10 ml 6-F NaOH and cool it with ice. Distill slowly until the bromine
vapor is no longer noticeable in the delivery tube. Gradually add to the
cold distillate 1 ml Na 20 2 powder. Continue the distillation until only
4 ml of the charge are left in the still pot. On the milligram scale follow
the directions of RACHELE (413).
Distillate .A 5. Os is indicated by a yellow color of the distillate. Add
2 ml ethanol, allow to stand for 5 minutes, separate from any precipitate,
and saturate the clear solution with H 2 S: black osmium sulfide.
Residue .A 5. Treat the residue in the still with 5 ml 9-F HCI0 4 and
place 12 ml 6-F NaOH into the receiver which is cooled with ice water.
Distil until only 3 ml remain in the still pot. An orange to red color. of
the distillate indicates Ru (a red precipitate, indicating mercury, should
be separated out). If the distillate is colored, add 2 ml more of 9-F HCI0 4
412 Separations P.68

to the residue in the pot and distil into 5 ml 6-F NaOH until again only
3 ml remain in the pot. If it is colored, add the second distillate to the first.
Distillate A 6. Proceed as with distillate A 5; the precipitate is ruthenium
sulfide.
Residue A 6. If this seems advisable, transfer the residue from the
distillation to apparatus which facilitates the separation of solid from
liquid. Treat the residue with 10 ml 12-F HOOOH (which also may be
used for rinsing the still into the apparatus to which the residue has been
transferred). Heat and keep gently boiling for 15 minutes. (If dissolution
is complete and no precipitate forms, then absence of Ta, Nb, Au, Hg, Pt,
and Pd is proven; proceed as with solution A 7.) If a solid is present
after the indicated boiling period, remove the supernatant solution A 7
while it still hot. Wash the solid with three 4-ml portions of hot water,
and combine the washings with solution A 7.
Solution A 7. Evaporate to a volume of 10 ml, cool with tap water,
separate any crystalline solid from the supernate A 9, and wash it thoroughly
with a small amount of ice cold 3-F HOI0 4. Reject the washings. Save
supernate A 9 which contains the analytical groups V to XVI.
Crystallizate A 9 proves the presence of large amounts of K, Rb, or Os.
Reserve it, and add it to supernate XV I-or better XVI 2-so that it will
be ignited with ammonium salt (acidify with HOI before evaporating)
for the destruction of perchlorate: 3 KOI0 4 + 8 NH 401 ~ 3 KOI + 8 HOI +
+ 4 N + 12 H
2 2 0.
Residue A 7. Use a small amount of water to transfer the residue into
suitable platinum apparatus (or substitute). Evaporate the water used in
transfer. Add 5 to 10 ml 27-F HF and heat 3 to 10 minutes on the steam
bath. Allow to cool and then add a volume of water equal to 2.5 times
the volume of the mixture. Mix thoroughly with a platinum tool to bring
into solution any tin fluoride that may have separated. Separate extract A 8
and transfer it to suitable platinum apparatus. Thoroughly wash residue A 8
with water, but combine only the first (intentionally small) portion of
wash liquid with extract A 8. Reject the rest of the washings.
Extract A 8. Add 4 ml 9-F H 2S0 4 and evaporate to fumes of S03'
Use 10 ml water (and possible the reagent solutions to be added) to transfer
the concentrate to apparatus suitable for heating under pressure and
separation of the liquid and solid phases. There add 15-F NH3 until the
mixture is ammoniacal (about 6 ml will be needed) and then 10 ml 6-F
(NH4)2S which has been freshly prepared. Mix, close the apparatus, and
immerse it in boiling water. If a clear solution results, treat it as solution II 1.
If a small precipitate forms, cool and remove the supernate. If, however,
a large precipitate separates, add 10 ml more of 6-F (NH4}2S, close the
apparatus, and heat for 30 minutes in boiling water. Then allow to cool
and separate solution II 1 from precipitate II 1. Wash the precipitate
P.68 Systematic Schemes for the Detection of Cations 413

with 5- to 10-ml portions of hot water; combine the first washing with
solution II l.
Precipitate II 1. See p. 420.
Solution II 1. Under a fume hood, add the solution in small portions
and with continuous mixing to 40 ml 3-F H 2 S0 4 , The mixture must
remain acid when all of solution II 1 has been added. Finally stir for
2 minutes and then separate precipitate II 2 from the supernate. Wash
the precipitate with hot water and finally dry it by placing the apparatus
containing it into an oven or a heating block at 120 0 C.
Supernate II 2. It may contain significant amounts of wolfram only
when phosphate is present. Consequently test it first for phosphate, and
jf this is present, test it for wolfram.
Residue A 8. Use 50 ml l.5-F Na 2C03 for transferring the residue to
some apparatus (large test tube) which returns the condensate of the
steam to the solution. Heat to boiling, add 10 g solid Na 2CO a, and boil
gently for 15 minutes (so that the steam condenses in the upper part of
the tube). Separate extract A 10 from residue A 10 while hot, and wash
the residue thoroughly with hot water to remove all sulfate and fluoride.
Reject extract A 10 and the washings.
Residue A 10. Add 10 ml 3-F HC10 4 , heat a few minutes oil the steam
bath. and then separate the extract A 11 from the residue. Wash the
.latter thoroughly with hot water, but add only the first washing to the extract.
Extract A 11 may still contain some sodium derived from the Na 2C03
and, consequently, should not be combined with the main solution. Treat
the extract as directed for supernate V 1 and analyze it separately, but
combine the group precipitates with those of the of the main solution (VII,
VIII 1, IX 1, and XV 1).
Residue A 11. Add 4 to 12 ml of a mixture of 3 volumes I6-F HNOa
with 1 volume I2-F HCI and heat on the steam bath for 10 minutes.
Evap'orate to a volume of 0.5 to 1 ml, add 12 ml 0.025-F HCI, heat to.
boiling, and separate extract A 12 containing the gold group. Wash the
residue A 12 with three 3-ml portions of 0.025-F HCl. Combine the first
washing with the extract, and reject the rest.
Residue A 12. Add 10 mII5-F NHa, stir for 2 minutes, and then remove
extract A 13. Test one droplet of the extract for the presence of silver
as in Expt. 32. If crystals of AgCI form, wash residue A 13 with 6-F NHa
until free of silver, and combine the washings with extract A 13.
Extract A 13. Heat on the steam bath until nearly all NHa has been
expelled. Then add 1 ml 6-F HCI and, if not yet acid, some HNO a. Heat
5 minutes on the steam bath. Collect the AgCI and estimate the amount
of silver.
Residue A 13. If it is purely metallic, treat it as directed for residue A 17.
If it is partly or wholly non-metallic, fuse with 10 g K 2 S 2 0 7 in a crucible
414 Separations P.68

or test tube of vitreous silica (when working on the gram scale, introduce
small portions of the dried residue into the fused salt; when working on
a small scale transfer the residue as a slurry with water and evaporate
the latter before adding the solid salt). Heat for 20 minutes at incipient
red heat so that S03 escapes only slowly (on the small scale, it may be
necessary to repeatedly add H 2 S0 4 , see p. 112). Allow to cool and extract
the melt with 25 ml cold water (on the gram scale, transfer the solidified
melt to a mortar and crush it to a fine powder; on the small scale, try to
spread the solidifying melt to a thin layer and observe the subsequent
dissolution with a magnifier). Wait until the melt is completely broken
down by the action of the water, mix, and remove the clear extract A 14
as completely as possible. Wash the residue with 2 ml cold water and
combine this washing with the extract. Treat the residue with 5 to 10 ml
12-F HOI and heat on the steam bath until the attack seems to be complete.
Evaporate to a volume of 1 ml, add 5 ml water, mix, and transfer the'
clear solution to extract A 14. Wash any residue A 14 with cold water
and add the first two washings to extract A 14.
Extract A 14. Do not combine with the main solution (A 9) since
this would make impossible the testing for potassium. Precipitate with
H 2S as directed in P. 65 of the classical scheme. Treat the sulfide precipitate
as directed for the solid sample of the scheme of NOYES and BRAY, and
proceed with the filtrate from the precipitated sulfides as with supernate VI 2,
below. This implies going through the whole scheme of separations, but
simplifications will be possible. If the appearance of any group precipitate
is similar to the corresponding group precipitate of the main solution,
the two may be combined; if they are quite dissimilar, it will be advantageous.
to analyze them separately.
Residue A 14. Treat with H2F 2 as described for residue A 7.
Extract A 15. Treat as direeted for extract A 8.
Residue A 15. Treat as directed for residue A 8.
Residue A 16. Treat as directed for residue A 10.
Extract A 17. Proceed as directed for supernate V 1, below. Treat
the sulfide precipitate like precipitate VI 1 + 2 and the supernate, like
supernate IX 1.
Residue A 17. Perform a Na 20 2 fusion with apparatus of nickel, iron,
or platinum. Since all of these materials are attacked, use the one which
is least objectionable as an impurity. Platinum is strongly attacked, but
this may not matter (aside from the damage to the apparatus) if platinum
metals need not be considered in the analysis. For manipulation adopt
one of the alternatives mentioned for the treatment of residue A 13.
Fuse residue A 17 with 3 to 5 g Na 20 2 at a temperature just sufficiently
high to keep the melt in the liquid state (incipient red heat). Maintain
the temperature until the residue is completely dissolved, which may
P.68 Systematic Schemes for the Detection of Cations 415

take from 3 to 20 minutes. Allow to cool and then transfer with the use
of 30 to 35 ml water into glass apparatus suitable for the separation of
solid and liquid phase. Caution is required on the gram scale since adding
water liberates heat and may be accompanied by the violent expulsion
of steam and gas.
Continuously stirring, heat the resulting mixture for 10 minutes on
the steam bath. Then cool with tap water and neutralize with continuous
cooling to prevent the escape of OS04. Add 12-F HCl, first in portions
of 1 ml and finally in portions of 0.1 ml until the mixture is just acid.
Add 1 ml ethanol (to Ie:luce the OS04) and 2.5 ml 12-F HCI. Heat on
the steam bath and stir until dissolution is complete. (If there should be
an insoluble residue, separate it, add 5 ml 12-F HCl and 1 ml 16-F HNOa•
heat on the steam bath for 5 to 10 minutes, evaporate to dryness, and
transfer the residue-solid and liquid-with the use of 3 mll-F HCI to the
bulk of the solution. If platinum apparatus has been used, yellow Pt 20 S
may separate when the melt is dissolved and may be discarded.)
Saturate the solution of the melt (together with some insoluble residue)
with H 2 S, close the apparatus, and heat for one hour in the steam bath.
When the apparatus is opened after cooling, the odor of H 2 S must be
discernible (otherwise saturate and heat again). Finally add 50 ml water
and saturate the cold mixture with H 2 S. Separate and wash the solid
with hot water.
Treat the supernate as directed for supernate VI 2, and analyze the
sulfide precipitate like any solid sample, starting with the dissolution in
HBr and Br 2 •
Supernate A 9 (Main Solution). Add 2 ml 2-F HBr, free from bromine
(mix 1 ml 9-F BBr, 1 m112-F HCOOH, 2.5 ml water and heat a few minutes
in the steam bath), cool with tap water, stir a few minutes, and allow the
precipitate to settle. Add another 2-ml portion of the HBr and continue
this until precipitation is complete. Finally separate supernate V 1 from
the precipitate V I, wash the latter with 5 ml cold I-F HBr (free from
bromine), and add the washing to supernate VI.
Supernate V 1. Add 3 ml 3-F NH"Cl and evaporate to dryness on the
steam bath. Dissolve the residue by heating with 30 mll-F HCI. Transfer
the solution to suitable apparatus and saturate it with H 2 S while still warm.
Add 70 ml water, saturate with H 2 S, close gas-tight, and heat for 15 minutes
in the steam bath (immersion in boiling water). Allow the precipitate to
settle, cool, and again saturate the supernatant solution with H 2 S. Close
again and heat 10 more minutes in the steam bath. Remove supernate VI I,
wash precipitate VII once with 10 ml hot water, and add the washing
to supernate VI 1.
Supernate VII (not indicated on flow sheet). If Ir and Mo cannot
be present, treat it as directed for supernate VI 2. Otherwise evaporate
416 Separations P.68

to a volume of 10 mI. Add 20 ml 6-F HCI, heat 10 minutes on the steam

bath, cool to room temperature, saturate with H 2S, close the apparatus
gas-tight, and immerse for 30 minutes into a bath of boiling water. After
cooling, saturate again with H 2S, close, and heat for 30 more minutes.
Separate supernate VI2 from precipitate VI 2, and wash the latter with
5 to 10 ml hot water. Reject the washings.
Precipitates VIl and VI 2. Heat precipitate VI 1 with 20 to 40 ml
12-F HCI for five minutes on the steam bath. Add 4 to 8 ml 16-F HNO a
and keep at 70° C until dissolution is complete.
If a precipitate VI2 has been obtained, heat it in like manner first
with 5 to 10 ml 12-F HCI and then after addition of 1 to 2 ml 16-F HN0 3 •
Add the solution to that of precipitate VII, and add also two rinse portions
of 10 ml 12-F HCI.
Evaporate to dryness (steam bath) the solution of precipitate VI 1
or the combined solutions of precipitates VI 1 and VI 2. Dissolve the
residue in 10 to 20 ml 12-F HCI and saturate the solution with S02. (If
Se precipitates, separate it out and continue with the clear supernate.)
Add water (35 to 70 ml) to lower the HCI concentration to 2.7 formal.
Saturate with S02' close gas-tight, and heat for 15 minutes in the steam
bath. Open and, while hot, separate supernate VI 3 from precipitate VI 3.
Wash the precipitate with two 10-ml portions of hot water and combine
the washings with the supernate.
Supernate VI 3. Concerning the detection of rhenium in this solution
see CHRISTINA C. MILLER (724). For the separation of the other elements,
evaporate supernate VI3 to dryness on the steam bath. Add 1 ml 16-F
HN0 3, evaporate to dryness, and repeat this with another I-ml portion
of the acid to be certain of complete oxidation of molybdenum. Cool to
room temperature, dissolve the residue in 5 ml 6-F HCI, and extract with
three 50-ml portions of diethyl ether. Wash the combined ether extracts
once with 5 ml 6-F HCI, and add this washing to the aqueous solution
VI 4. ,
Supernate VI 2. Use one twentieth of the supernate for a side test
for Fe and P0 4. Boil to expel H 2S. Then add 0.5 ml 3-F HN0 3 and
evaporate nearly to dryness. After cooling, dissolve the residue in 1 ml
12-F HCI. Add 5 ml 6-F HCI and 6 ml diethyl ether for the extraction
of FeCI3 • - Allow the ether extract to evaporate. Treat the residue with
0.5 m16-F HCI, 10 ml water, and 2 mll-F KCNS. Use a light red coloration
for the estimation of the amount of iron. If the solution turns dark red,
add 2 ml 6-F NaOH and estimate the quantity from the volume of
Fe(OH)3. - Evaporate the aqueous layer nearly to dryness, then add 3 ml
water and lO ml of a mixture of equal volumes of 6-F HN0 3 and molybdate
reagent [dissolve 90 g (NH4)2Mo7024· 4 H 20 in 100 ml 6-F NH 3 , add
240 g NH 4N0 3, and dilute with water to 1 liter). Immerse for 15 minutes
P.68 Systematic Schemes for the Detection of Cations 417

into a water bath at 70° C. Estimate the amount of PO, from the volume
of the. yellow precipitate.
Treat the main part of supernate VI 2 as follows. Boil to remove
the dissolved H 2 S (passing a stream of air through the solution will hasten
the process). Add bromine until color or odor of the solution indicate
presence of a slight excess (this oxidizes pale green Fe++ and blue V 20,==).
Heat and pass air through the solution to expel the excess of bromine.
Add 4 ml 6-F NHa and 15 ml 3-F ammonium acetate (if phosphate is
absent, heat to near boiling-if no precipitate forms, the amounts of Fe,
AI, Ga, Ti, and Zr must be less than 0.002 g-and then cool again to room
temperature). Treat the cold solution (mixture) with I-ml portions of
I-F Fe(NO a)3 until its color becomes deep brown due to the hydrolysis
of an excess of ferric ion. Add 3 ml more of the ferric nitrate solution,
heat to incipient boiling, maintain this for 2 minutes, and then place
into a water bath at 90° C. If a precipitate does not settle out, add 2 ml
6-F NHa, and heat again to near boiling for 2 minutes. If this does not
cause settling, add 5 ml 0.5-F (NH')2HPO, and again heat for 2 minutes.
Keep above 90° C all the time. If the solution should assume a brown
color, add 1 ml 6-F NHa (test with litmus paper and add 6-F acetic acid,
if necessary, to restore the acid reaction) and heat again for 2 minutes
near boiling. Repeat the treatment with NHs if necessary.
Separate precipitate VillI from the supernate when the condition
is satisfied that the precipitate settles out and the solution does not show
a coloration indicating ferric ion. Separate solution and precipitate while
hot, and wash the precipitate with two 50-ml portions of boiling water.
Add the washings to supernate VIII 1.
Precipitate VIIIl. Without delay, treatwith5to 10mI6-FHClandevapor-
ate on the steam bath to a volume of 1 ml. Continue as directed on p. 418.
Supernate VITI 1. Treat with I-ml portions of 6-F NH3 until the mixture
has the odor of NH3, and then add 2 more milliliter 6-F NH s. Heat just
to boiling. Pass H 2 S into the mixture until it has the odor of H 2 S. Test
with litmus and, if necessary, add NH3 to restore alkaline reaction. Heat
to incipient boiling and remove the supernate. Boil the supernate for
half a minute (or longer if it has a dark color) and separate the clear
supernate IX 1 from any small amount of precipitate which may have
formed. Pass some H 2 S into 50 ml 0.06-F NHa (test with litmus to be
certain of the alkaline reaction), and use this solution for washing; with
the first portion transfer the small amount of precipitate obtained on
boiling the supernate to the main portion of precipitate IX 1. Combine
the first washing with supernate IX 1.
Precipitate IX 1. Without delay dissolve precipitate IX 1 in 10 ml 6-F
HN0 3, evaporate just to dryness on the steam bath, and reserve the residue
for further treatment as outlined below.
Benedetti-Pichler. Identification 27
418 Separations P.68

Supernate IX 1. Evaporate a small portion of the supernate on the

steam bath to dryness, add to the residue another small portion, evaporate
again, and continue until the whole centrifugate has been evaporated.
Treat the residue with 25 ml 12-F HOI, evaporate to dryness, and repeat
this treatment once. Extract the residue with three 50-ml portions of water.
Evaporate this extract in suitable platinum apparatus (dish, crucible, sheet)
to dryness on the steam bath (again it is advised to evaporate small portions).
Moisten the residue with 10 ml 6-F NH,OI and evaporate to dryness.
Heat for 10 minutes at 140° 0 (drying oven or heating block) and then
heat with a small flame until fumes are no longer coming off. Finally
heat to incipient dark red heat and quickly remove the flame. Extract
the residue with several 10-ml portions of water, and evaporate the clear
extracts to dryness. Dissolve the residue in 10 ml water and separate
out any turbidity. Treat the clear solution with 15 ml of carbonate reagent
(25 g freshly pulverized ammonium carbonate dissolved in 6-F NHs to
obtain a volume of 100 ml) and 15 ml 95% ethanol. If the precipitate is
large, add a second installment of 15 ml carbonate reagent and 15 mlethanol.
Finally stir, allow to stand for 10 minutes and aid crystallization by
occasional shaking or exposure to vibrations of high (but not supersonic)
frequency (413). Separate supernate XV 1 from precipitate XV 1, and
wash the latter with a very small volume of a mixture of equal parts of
carbonate reagent and ethanol.
Precipitate VIII 1. Treat the 1 ml of the solution of the precipitate
with 10 ml 6-F HOI and (regardless of any precipitate) 10 ml diethyl ether
(shake 40 ml ether, immediately before use, with 80 ml 6-F HOI to make
it suitable for this extraction). Repeat the extraction with two more
10-ml portions of the ether. The combined ether extracts are extract VIII 2.
A white precipitate forming on addition of ether indicates presence of a
large amount of Zr or W; a yellow coloration of the aqueous layer suggests Ti.
Water Layer VIII2. Evaporate to a volume of 1 ml. Add 2 ml 16-F
HNO s and evaporate nearly to dryness. Dissolve in 10 ml water and; if
necessary, transfer to suitable apparatus of either nickel, platinum, or
vitreous silica (the last is most convenient; Vycor or Pyrex glass may
be used if contamination with aluminum is not objectionable). Add small
portions of pure 6-F NaOH (tested for absence of AI if small amounts of
AI shall be detected in the sample) until the mixture is alkaline. If the
mixture is thick with precipitate, add 10 to 20 ml more water. 0001 the
apparatus with ice water, and treat its contents by slowly adding, with
continuous stirring, 1 to 3 ml solid Na 20 Z (free from AI and assuredly
undecomposed). Oontinuous liberation of oxygen indicates that enough
of the reagent has been added. Boil gently for 5 minutes, dilute with water
to 40 ml, heat to a brief boil, cool, and separate supernate VIII 3 from
precipitate VIII 3 (hardened filter paper may be used, but centrifuging
P.68 Systematic Schemes for the Detection of Oations 419

and decantation are preferable). Wash the precipitate thoroughly with

hot water.
Supernate VIII 3. Cool the supernate and add 0.5- to I-ml portions
of 6-F HN0 3 until the solution is acid. Set aside for combining with
supernate IX 4.
Precipitate VIII 3. Without delay add 5 to 10 ml 6-F HOI and set
aside for combining with precipitate IX 6.
Precipitate IX 1. Treat the residue from the evaporated solution of
precipitate IX 1 with 2 ml I6-F HNO a and again evaporate to dryness.
Treat the residue with 15 ml I6-F HNO a and 1 ml solid KOlOa, and stir
for 5 minutes while heating in the steam bath. Add another I-ml portion
of solid KOI0 3 , heat and stir for 2 minutes, and repeat this with a third
I-ml portion of the solid reagent. Remove supernate IX 2 from the
precipitate IX 2. Treat the latter with 10 ml I6-F HNO a and 1 ml solid
KOlOa, heat two minutes with stirring, separate, and transfer the supernate
to supernate IX 2.
Supernate IX 2 (not mentioned on the flow sheet). Evaporate just
to dryness, and dissolve the residue in 20 ml water. Separate from any
solid (Ppt. IX 3) to obtain the clear supernate IX 3.
Precipitates IX 2 and IX 3 are combined for the estimation of the
amount of Mn.
Supernate IX 3. Proceed exactly as directed for water layer VIII 1.
Obtained are supernate IX 4 and precipitate IX 4. Wash the latter with
three IO-ml portions of hot water and add the first washing to supernate IX 4.
Without delay treat precipitate IX 4 with 5 to 20 ml 6-F HCI and set it
aside for later use.
Supernate IX 4. 0001 and add 0.5- to I-ml portions of 6-F HNOa until
acid, and then add to supernate VIII 3.
Combined Supernates VIII 3 and IX 4. With mixing add 0.5-ml portions
of solid NaHOO a until Congo red paper remains red. Then add 2 g solid
NaHOO a, 40 ml water, and 2 ml 3 % H 20 2 • Close the apparatus air-tight,
and place it in boiling water for 20 minutes. Cool, open the apparatus,
and separate supernate VIII 4 from precipitate VIII 4. Wash the latter
with one 20-ml portion of cold water and combine the washing with
supernate VIII 4. Without delay treat precipitate VIII 4 with 5 to 15 ml
6-F HOI.
Precipitate IX 4. Evaporate it with the acid (which has been added)
nearly to dryness (steam bath). Treat the residue with 30 ml cold water,
and add I-ml portions of 6-F.NH3 until the mixture has the odor of ammonia.
Add 2 ml more of 6-F NHa and mix. (If no precipitate appears at this
stage, lanthanides are apparently not present in this solution which may
be treated immediately with H 2 S for the precipitation of 00, Ni, and Zn.)
Add 2-ml portions of 6-F acetic acid until the mixture is acid to litmus
27*
420 Separations P.68

(separate a precipitate "LA" which refuses to dissolve, wash it once with

hot water, treat it with 5 ml 6-F HCI, and reserve the mixture for combining
with precipitate IX 6). Mix into the clear solution 10 ml 3-F ammonium
acetate, heat to about 75° C, and pass into it for 5 minutes a slow stream
of H 2S. Mix and heat 5 minutes on the steam bath. Cool to room
temperature, saturate with H 2S, heat again to about 75° C, saturate
again with H 2 S, and then separate' supernate IX 5 from precipitate IX 5.
Wash the latter with cold water that has been saturated with H 2 S. Attend
immediately to the precipitate.
Precipitate IX 5. If the precipitate is pure white, it consists of ZnS
and is used for the estimation of the amount of Zn. Otherwise add 10 to
30 ml cold 1-F HCI, stir for 5 minutes, and without delay separate
supernate X 1 (containing Zn with little Co and Ni) from residue X 1
consisting of CoS and NiS.
Supernate IX 5. Heat on the steam bath and stir to remove the
dissolved H 2 S. Remove any precipitate of suliur, and treat the clear
solution with small portions of 6-F NH3 until the mixture has a slight odor
of NH 3. Finally add 2 ml more of 6-F NH 3, and separate supernate IX 6
from a precipitate IX 6. Washing is not necessary.
Precipitate IX 6. Add the acid solutions of precipitate VIII 3 and
of a precipitate "LA". Transfer the mixture to suitable apparatus of
platinum or plastic (also vitreous silica and Pyrex glass may serve with
the obvious limitations) and evaporate to dryness. Treat the residue
with 2 ml water, 0.5 ml 27 -F HF, 1 ml 6-F HCI, and finally again 9 ml water.
If a large residue is left, add O.l-ml portions of the HF (but not more
than a total of 0.5 ml) until there is no more dissolution (the volume of
the residue remains constant). Heat 5 minutes on the steam bath, and
then remove supernate VIII 5 from the precipitate VIII 5. Wash the
latter with two very small portions of a mixture of 10 ml water, 0.5 ml
27-F HF, and 1 ml 6-F HCI.
Without delay treat supernate VIII 5 as described in the literature.
(Add small portions of 6-F NH3 until alkaline, then add 0.3 ml portions
of 3-F HF until acid, add 0.3 ml more of the H2F 2' saturate with H 2 S,
allow to stand for 5 to 10 minutes, and separate In 2 S3 (CoS, NiS, ZnS)
from the supernate containing Ti and Zr.]
Analysis of the Tantalum Group
Precipitate II 1. Treat with 400 ml hot water which contains dissolved
15 g salicylic acid and 4.7 g Na 2C0 3. Boil the mixture for two hours (but
stop the boiling immediately if complete dissolution should occur), and
replace water which has vaporized to keep the volume constant. Separate
extract III 1, while boiling hot, from residue III 1. Wash the latter
thoroughly with boiling hot water..
P.68 Systematic Schemes for the Detection of Cations 421

Extract III 1. A slowly developing violet coloration is caused by the

presence of iron; 0.0003 g Ti give a yellow and 0.0025 gTi an orange coloration.
Evaporate to a volume of 40 to 60 ml. Add 6 ml 9-F H 2S0 4 , cool with
tap water, add 40 ml diethyl ether, mix, and separate the layers. Repeat
the extraction with two 20-ml portions of ether. Reject the ether extracts,
but treat the remaining aqueous layer as follows. Evaporate to a volume
of 15 to 20 ml. Add small portions of 6-F NaOH until the mixture is
alkaline, and then add 10 ml more 6-F NaOH. Mix and heat for 5 minutes
on the steam bath. Remove supernate III 3 while still hot, and wash
precipitate III 3 thoroughly with hot water.
Precipitate III 3 is suited for the estimation of the amount of Ti; it
will not contain more than 0.001 g Zr and possibly a trace of V.
Supernate ill 3. Use one fifth of it for estimating and confirming
phosphate by means of molybdate reagent, and use the rest of it for
confirming and estimating vanadium.
Residue III 1. Add water, make a slurry, and transfer it to suitable
platinum apparatus (dish, crucible, sheet). Evaporate the water and treat
with a mixture of 2 ml 9-F H 2S0 4 , 0.5 ml 16-F HNO s, and 1 m127-F HF.
Evaporate to .the appearance of fumes of SOs. Add 0.5 ml 16-F HNOs,
evaporate again to fumes of SOs, and repeat the treatment until all organic
matter is destroyed. Finally evaporate to dryness. Use the amount of
residue to get an estimate of the total amount of Ta, Nb, Zr, and Bi.
Then add 5 g of finely powdered anhydrous K 2CO S which contains 2 % KNO s.
Fuse and maintain the Hquid state for 5 minutes. After cooling, extract
the melt with 10 ml cold water. Stir until the melt is completely disintegrated,
and then remove extract III 5 and wash residue III 5 with several 5-ml
portions of cold 50% K 2CO S and finally with 5 ml water. Combine the
first washing with extract III 5.
Extract III 5. Dilute with water to a volume of 40 ml and saturate
with S02. Heat 30 minutes on the steam bath while passing S02 through
the solution. Remove supernate III 7 while hot, and wash precipitate III 7
with hot water.
Supernate 1117. Boil it for the removal of S02 before testing for
phosphate with molybdate reagent.
Precipitate III 7. The formation of the precipitate indicates the presence
of either Ta or (and) Nb. If necessary, transfer as a slurry with water to
suitable apparatus, evaporate the water, and dry by placing into a dJ·ying
oven. Add 5 g K 2S 20 7 , heat, and keep the melt in the liquid state for
5 to 10 minutes. After cooling, add 100 ml 0.2-F (NH4)2C204' heat on the
steam bath and stir until the melt is completely disintegrated.
To test the clear solution of the melt for Ta and Nb, add 5-ml portions
of 2 % tannin solution. Heat after each addition on the steam bath, and
then centrifuge (avoid to stir up any precipitate which has been already
422 Separations P.69

collected). If no precipitate is obtained, add some solid NH 4Cl and 2 ml

2-F NH a. Continue adding portions of tannin solution as long as a precipitate
forms.
The yellow Ta compound precipitates first (without requiring NH 4Cl
and NH a). Later (after adding the ammonia buffer) an orange mixture
of Ta and Nb compound separates and finally, the red Nb compound.
A small amount of Ti which, however, should not get into ppt. III 7,
gives a red ppt. on adding the first portion of tannin. MILLER (723) gives
advice concerning this difficulty.
Residue III 6. Perform a pyrosulfate fusion as directed for precipitate
III 7, above. After cooling, extract the melt with 20 ml water by heating
on the steam bath and stirring until the melt is completely disintegrated.
After cooling to room temperature, saturate with H 2S, remove super-
nate III 6, and wash precipitate III 6 with some water saturated with H 2 S.
Supernate III 6. Boil to remove H 2 S, and then test the clear solution
for Zr. [Add 2 ml I-F H 20 2 and 10 ml 0.5-F Na 2HP0 4 • Allow to stand
for 1 hour: 0.001 g Zr still gives a white ppt. of Zr(HP0 4)2.]
Precipitate ill 6. Of interest is only the presence of Bi. Not more
than 0.002 to 0.005 g of it may be expected in this place, but it may happen
that all of a small amount of Bi collects here. A small amount of black
precipitate may also be PtS 2 derived from the use of platinum apparatus.
To test a black precipitate III 6 for Bi, first estimate its amount and
then dissolve it by heating with 10 ml 2-F HNO a. Treat the clear solution
as directed for the solution of ppt. VI 5, see the literature.

P.69 Systematic Search for Anions

The search for cations is essentially ultimate or elemental analysis, i. e.,
an investigation of the absence or presence of elements without regard
to their binding. In the instance of anions, however, one customarily
tests for "molecular" species: 01-, CIO-, CIO a-, etc. This task would be
very difficult if a systematic separation were to include all known inorganic
anions, among others such as SnOa=, SnCI 6=, Mn(CN)6==' CO(N0 2 )6=-'
Co(NHs)(N0 2 )s-. There is no need for such a scheme, however, since
knowledge of elemental composition, of radicals present, and of the
appearance and physical properties usually permits drawing the necessary
conclusions.
If the tests with dilute and concentrated sUlfuric acid (P. 36 and P. 37)
have not yet been performed, they should be carried out now. Table XI
gives a list of the "acids" that should be considered. The inclusion of some
organic compounds and of the iron complexes may be justified by their
frequent occurrence and the possibility that their organic nature may
not be recognized when the substance is ignited. Italics indicate the acids
that may be recognized by testing with su]furic acid. The presence of Se,
P.69 Systematic Search for Anions 423

Table XI. Acids Oommonly Oonsidered in Testing tor Anions

HF
HOl HOW HOlO 2 HOlO a HClO 4
HBr HBrOa
HI HIO a HI0 4
H 2S H 2SO a H 2 S 20 3 H 2 S,.o6 H 2 SO 4 H 2 S2 OS
H 2Se H 2 SeOa H 2 Se0 4
H 2 Te H 2TeOa H 2Te0 4
HNa HN0 2 HNO a
HaP H(H 2P0 2 ) H 2 (HPO a) H 2 (H 2P 2OS ) H aP0 4 Poly
H3As HaAsOa HaAsO f
H 40 H(HOOO) H 2 (OOO) 2 H 200 a H(CHaCOO)
HON HONO HONS H4Fe(CN)6 H3Fe(CN)s
(HjSi) HzSiOa H 2SiF s
(H3 B ) HB0 2 HBF4
Italics indicate the acids that may be discovered when testing a sample
with dilute or concentrated sulfuric acid.
Boldface indicates those which are not precipitated by barium or silver
nitrate (or may escape detection by these tests) so that special tests are required.

Te, and As is discovered in the search for metals. In addition, the following
direct ultimate tests may be used to determine presence or absence of
nonmetals:
F: volatilization as SiF4 and precipitation of Na 2 SiF 6 , P.60.
01, Br, I: strong ignition of a solid sample with sufficient Na 200 a to
maintain alkalinity and then testing of the solution of the melt with HN0 3
and AgNO a. To exclude the interference of cyanide, recrystallize from NH a.
S: oxidation to sulfate or Hepar test, P.59.
P: oxidation by heating with conc. HN0 3, boiling the aqueous solution,
and molybdate test, P. 58.
Si: volatilization of SiF4 and precipitation of Na 2 SiF 6 , P.57.
B: tests given in P.56.
When finally the solubility of the material and its metallic constituents
are considered, a large number of possibilities may be eliminated from
the list of anions so that the remaining choice may be decided by a few
confirmatory tests (P. 43); these have to be selected so that the substances
known to be present and also those that might be present will not interfere.
To this end, it may be necessary to remove the metals (P. 71, 72) or to
perform some simple separations. Obviously, if the material under investiga-
tion is quite uniform in appearance and seems to represent only one substance,
the number of anions will probably not exceed three.
With complex materials, it may be desirable or necessary to do some
more preliminary testing (P. 70) before proceeding to the confirmatory tests.
In general, it is advisable to follow a slight modification of the reasoning
of A. A. NOYES (48) and to adjust the procedure depending upon the
424 Separations P.70

type of material: (I) materials that contain only alkali metals or NH"
P. 70; (II) such that are readily dissolved or decomposed by water or acids,
P. 71; and (III) refractive materials which are insoluble or only with
difficulty attacked by acids, P.72.

P.70 I. Only Alkali Metals or (and) Ammonium are Present

If the material is difficultly soluble in water, the choice becomes quite
limited; inspection under the microscope and a few tests should solve
the problem.
Otherwise dissolve 1 g of the material in 2 ml water. With the use of
4-F HN0 3 and 6-F NaOH (best a freshly prepared solution of pure reagent)
adjust the pH to about 5 to 9 and then dilute with water to 12 ml and
mix to obtain the sample solution for orientation and confirmatory tests.
Portions of this solution win serve for orientation and confirmatory tests
(if n grams are taken for work on a small scale, multiply by n all weights
and volumes given in the following directions, P. 70-72).
For orientation, it is suggested to use the classification with barium-
calcium nitrate and silver nitrate-and some .of the tests for oxidizing
and reducing agents given below. The observations should reduce the
number of possibilities to an extent permitting conclusion of the search
with a few confirmatory tests, P.43. A preliminary report by Ly (1263)
indicates that classification by barium-calcium and silver nitrate will
detect 0.0004 g (1 % of about 0.04 g total anion) or less of most of the
anions listed. It is suggested to separate the supernates from the precipitates
for additional testing; this will aid in saving sample.
Classification by Ba,rium-Calcium Nitrate
Dilute 1 ml of the test solution with 1 ml water. Add O.I-ml portions
of 6-F acetic acid until litmus turns red and .then as much more as has
already been added. Treat with 1 ml 0.5-F Ba(N0 3)2 and 3 ml 0.5-F
Oa(N0 3)2, heat nearly to boiling, and let stand for 10 minutes: precipitate (1)
and supernate (1).
Precipitate (1) may contain:
pale yellow BaCrO, soluble in 6-F HOI,
white S from polysulfide, thiosulfate, thionate; insoluble in acid,
BaSO, insoluble in acid,
BaSe0 4 and BaTe0 4 dissolved by boiling 12-F HOI with liberation
of chlorine,
OaF2 and BaSiF 6 (the latter distinctly crystalline) and both dissolving
slowly in 6-F HOI,
OaS0 3 and OaC 20, dissolving readily in 6-F HOI,
silicate which mayor may not dissolve in HOI.
P.70 Systematic Search for Anions 425

Supernate (1). Add 0.05-ml portions of 12-F NH3 until the odor of
NH3 indicates a slight excess. Warm and then allow to stand: precipitate (2)
and supernate (2).
Precipitate (2) may contain: iodate, sulfite, possibly thiosulfate, selenite,
phosphite, hypophosphate, phosphate (ortho or poly), arsenite, arsenate,
borate, and possibly some carbonate. It may be dissolved in 8-F HN0 3
and tested with AgNO a, below.
Supernate (2) may contain: any of the halogen ions excepting fluoride
(and only little iodate), persulfate, little sulfide or sulfite, azide, nitrite,
hypophosphite, and possibly hypophosphate. There is much acetate
present, that could give a crystalline precipitate with AgNOs, soluble in
hot water. The supernate may be tested with AgNO s, below.
Classification by Silver Nitrate
. To 1 ml of the test solution add 5 ml water, 1 ml I-F AgNO a, and
2 ml 6-F HNO s : precipitate (3) and supernate (3).
Precipitate (3) may contain:
black Ag 2S from sulfide, thiosulfate, thionate,
AgsP (possibly a small amount of it),
AgO from H aP0 2 or H aP0 3 ,
AgzO z from HZSzO g,
slate colored Ag aH 2IO e, from periodate,
dark red Ag2HsIOe from periodate,
orange AgIO, and AgaFe(ON)6,
yellow AgI,
pale yellow AgBr,
white AgOI, AgON, AgONS, Ag,Fe(ON)6' and possibly AgIO s.
If the precipitate is black, boil it briefly with 12-F HNO s and dilute
with four volumes of water to see if a light colored or white precipitate
persists.
Supernate (3): test 1 ml of it by adding 0.05 ml (drop) of I-F NaNOz.l
A white precipitate may be AgOI derived from hypochlorite or chlorate,
or AgBr from bromate.
Treat the rest of the supernate (3) by adding 0.05-ml portions of 6-F
NaOH to obtain a pH of 7 to 9. A precipitate may contain:
black AgO from phosphite or hypophosphite,
Ag Z0 2 from persulfate,
reddish brown Ag 2Cr0 4 or AgsAsO"
yellow AgsPO, or AgsAsOs,
white, turning brown to black: quickly AgzSzOs, AgHzPOz,
slowly AgzHPO a, AgB0 2,
1 Must be free from chloride; it suffices to treat 1 ml of the reagent with
1 drop AgNO s' to centrifuge, and to use the clear s.upernate.
426 Separations P.70

white AgIO a,
Ag 2 SO a, little and possibly derived from H 2S 20 a,
Ag 2 S 20 a,
Ag 2 SeOa,
Ag,P206' Ag 20 20" and AgB0 2.
Tests for Oxidants and Reductants
The absence of oxidants (reductants) cannot be considered conclusively
proved unless also the test for reductants (oxidants) has a negative outcome.
This still assumes that oxidant and reductant are not present in amounts
to produce complete cancellation of both their actions.
Test for Oxidants
a) NOYES (48): To 0.5 ml of test solution add 2 ml saturated solution
of MnOl 2 in 12-F HOI, and heat nearly to boiling. A dark brown to black
coloration may be caused by:
010-, OIOa-, N0 2-, NO a-, Fe(ON)6=-' OrO,=, MnO,-.
b) SWIFT (53): Treat 0.5 ml test solution with 0.5 ml 001" 0.5 ml 3-F
H 2SO" and five O.l-ml portions of a freshly prepared solution of 1 g KI
in 2.5 ml water, mixing thoroughly after each addition. A red or violet
color of the 001, may be caused by:
BrO a-, IOa-, 10,-, S208=' N0 2 -, Fe(ON)6=-' Or0 4 =, Mn0 4 -.
Ohlorate and nitrate give the coloration after standing for 30 minutes.
Tests for Reductants
c) MIDDLETON and WILLARD (44): Add 0.02-F KMnO, to l.5-F H 2S0 4
until a stable pink coloration is obtained. Mix 1 ml of this acid with
0.5 ml test solution. If the mixture becomes colorless, the following may
be present:
Br-, 1-, ONS-, Fe(ON)6==' S=, SOa=, S20a=, N0 2 -, AsO a=-, 0 20 4 =.
d) MIDDLETON and WILLARD (44): Mix 2 ml water, 2 ml 0.3-F Fe(NOa)a,
1 ml freshly prepared saturated solution of KaFe(ON)6' and 4 ml 3-F HOI.
Mix 2 ml of the freshly prepared reagent with 0.5 ml test solution and
compare the color of the mixture with that of the reagent. A blue or
green coloration may be brought about by:
1-, Fe(NON) 6==' S=, SOa=, S20a=, 0 20,=.
(A red color indicates thiocyanate.)
It should be understood that a test for nitrate is meaningless if nitric
acid has been used in the neutralization of the test solution. The search
is concluded by the performance of confirmatory tests, and it should not
be forgotten to test for the acids indicated by boldface in Table XI. If
much borate is present, it is recommended to test the Ba-Oa precipitate (1)
for fluoride.
P. 71, 72 Systematic Search for Anions I 427

P.71 II. Nonmetallic Materials Readily Dissolved or Decomposed

by Water or Acids
Into this class belong industrial products of all kinds as well as soluble
minerals and rocks.
It is assumed that the original material has been already tested for
sulfide and carbonate. The search for the other anions is carried out with
the soda extract which is prepared to eliminate most of the possibly
interfering cations. The procedure for the preparation of the soda extract
will have to be somewhat modified depending upon the nature of the
sample, see below. If part of the material is insoluble in acid, prepare
the soda extract from the acid solution and treat the insoluble residue
according to P. 72. Proceed with the soda extract as outlined in P. 70.
The Material is Solid or a Paste. Treat 1 g of the finely powdered
material with 10 m11.5-F Na 2CO a and heat on the steam bath for 10 minutes
with continuous stirring. Remove the clear supernate, dilute it to 12 ml,
and mix. It is the soda extract for the anion search. Wash the residue
once with water and save it for possible later use.
The Material is an Aqueous Solution of pH 5 to 10. Take a volume of it,
which corresponds to approximately 1 g solid residue, and evaporate to
about 2 ml. Add 10 ml 1.5-F Na~COa and heat for 5 minutes on the steam
bath with continuous stirring. Remove the clear supernate, dilute with
water to 12 ml, and mix; this is the soda extract.
The Material is an Acid Solution. Take a volume corresponding to
1 g solid residue and add small portions of freshly prepared 6-F NaOH
until alkaline. Evaporate to 2 ml and then add 10 ml 1.5-F Na 2CO a and
continue as directed in the preceding paragraph.
The Material is an Alkaline Solution. Take a volume corresponding
to 1 g solid residue and saturate with CO 2 (which may have to be washed
with bicarbonate and then water). Evaporate to 2 ml, add 10 ml 1.5-F
Na 2CO a, and continue as directed for an aqueous solution, above.

P.72 III. Nonmetallic Refractive Materials

This class comprises insoluble minerals and ores, ceramic products,
glass, slags, glazes, and abrasives. Their origin at high temperature excludes
many anions, and they have to be tested only for fluoride, chloride, nitride,
phosphate, sulfide, sulfate, carbide, carbonate, cyanide, boride, and borate.
Also the elements C, Si, S, Se, and Te should be considered, and obviously
oxides may be present.
Perform a fusion with 0.5 g of the material or with the acid insoluble
residue of 1 g of it. In platinum or nickel apparatus, fuse the finely pulverized
material with 10 to 20 ml of anhydrous Na 2CO a (or a 1: 1 mixture of
Na 2CO a and K 2CO S )' An electric oven or an alcohol blast flame will not
428 Final Review of Observations and Report P.73

introduce suliur. Keep in the molten state for 10 to 20 minutes. If dark

particles remain undissolved after this time, add 0.3 ml solid NaNOs and
again heat until the whole mass is liquid. After cooling, add 40 to 60 ml
water and boil until the whole melt has crumbled. The clear supernate
is the soda extract which is tested as outlined in P. 70. Test precipitate (I)
for fluoride and also consider the following tests.
Test for Silicate and Borate. Use one third of the soda extract obtained.
Acidify with HCl, evaporate to dryness on the steam bath, moisten the
residue with 12-F RCl, again evaporate to dryness, and heat the residue
at 130 0 C (oven or heating block) until completely dry. After cooling,
add 6 ml 6-F HCl and warm the mixture to about 60 C. A white residue
0

consists of partly dehydrated H 2 SiOs and may be used for estimation

of the amount and confirmatory tests. Remove the supernate and test
it for boric acid.

P. 73 Final Review of Observations and Report

For all practical purposes, the important part of the report is the final
conclusion. In the simplest instance, this will be a statement on presence
or absence of some substance. The conclusion may state the identity of
the material under investigation in more or less definite terms, or it may
only indicate the general type of matter.
The observations from which the final conclusion is drawn may be
made a part of the report. Their meaning increases in the same degree
as the final conclusion becomes indefinite or vague. If the facts indicate
a "new" substance which apparently has not yet been described, so that
identification or recognition is not possible, the report must by necessity
become a tabulation of the physical and chemical properties observed.
If the material has been recognized, the identification should explain
all observations such as state, shape, color, odor, hardness, density, behavior
on heating and toward reagents, etc. It may happen, however, that it
is impossible to explain an unusual color or odor by the observed facts
since the responsible trace of impurity or incidental contamination cannot
be detected by the common means of qualitative analysis. If explanation
of such deviations from the norm is essential, it will be necessary to turn
to the literature on the material involved, its investigation, and to methods
of trace determination. By all means, any features and facts not explained
by the identification should be emphasized in the report.
Checking positive findings by blanks and negative findings by suitably
performed controls becomes the more essential, the smaller the percentages
are, which have to be considered; in a search for traces, the performance
of blanks and controls may constitute the largest and most important
part of the task.
P. 73 Final Review of Observations and Report 429

It is occasionally stated that synthesis is the final proof of analysis.

There are, however, limitations to synthesis on the basis of a qualitative
analysis even when the latter provides good estimates of the amounts.
To mention just one example, the preparation of an alloy may not succeed
even when the correct elemental composition is duplicated. It may be
necessary to perform an extensive investigation to discover the treatment
of the alloy (tempering, quenching, cold and hot working, and surface
treatment) which produces the microscopic or submicroscopic structure
that gives the observed properties.
 Appendix
Test Solutions
To obtain test solutions containing 10 mg of the substance of interest
(metal, cation, anion) per milliliter solution dissolve the following weights
to obtain 100 ml solution (if work is done on a semimicro or smaller scale,
it is recommended to prepare only lO ml of the solutions).
Dissolve in water:
(test solutions for cations, metals): 6.1 g LiCl; 2.5 g NaCI; 2.6 g KNO a ;
1.4 g RbCI; 1.3 g CsCI; - - - 19 g BeS0 4 · 4 H 40; 11 g Mg(NO s)2' 6 H 20;
5.9 g Ca(NO S )2' 4 H 20; 2.4 g Sr(NO a)2; 1.9 g Ba(NO a)2; - - - 4.3 g Y(NOS)3'
.6 H 20; 3.1 g La(NOa)a' 6 H 20; 3.1 g Ce(NOa)a' 6 H 20; 3.0 g Nd(NOa)a'
.6 H 20; - - - 3.0 g Th(NO a)4' 12 H 20; - - - 7.7 g Cr(NOa)a' 9 H 20; 1.8 g
(NH4) sMo 7024 ·4 H 20; 1.8 g Na 2W0 4 · 2 H 20; 2.1 g U0 2(NOa)2' 6 H 20;
- - - 5.2 g Mn(NO a)2 • 6 H 20; - - - 7.3 g Fe (NOsh . 9 H 20; 5.0 g CO(NOa)2 .
. 6 H 20; 5.0 g Ni(NO a)2' 6 H 20; - - - 2.0 g RhCl a ; 1.3 g OS04, fume hood;
2.7 g HJltCl s ' 6 H 20; - - - 3.8 g CU(NO a)2' 3 H 20; 1.7 g AgNO a ; 2.0 g
HAuCI 4 · 3 H 20; - - - 4.6 g Zn(NOS)2' 6 H 20; 2.7 g Cd(NO a)2' 4 H 20;
1.4 g HgCI 2; - - - 14 g Al(NOs)s' 9 H 20; 1.3 g TINO s ; - - -1.1 g Pb(NO a)2;
- - - 1.5g As 20 S ; - - - l.4g Se0 2;
(test solutions for anions): 2.5 g Na 2B 4 0 7 · lO H 20; - - - 2.0 g Na 2CO a ·
. H 20; 1.9 gNaCN; 2 gKCNO; 1.7 gKCNS; 2.0 gK4Fe(CN)6' 3 H 2 0; 1.5 g
KaFe(CN)6; 1.5 gNaHCOO; 1.5g Na 2C20 4; 2.3g NaCH s ·COO·3H 20;
3.7 g Na 2SiO a · 9 H 20; - - - 1.5 g NaN a ; 1.8 g KN0 2; 1.7 g KNO a ; 1.6 g
NaH 2P0 2 • H 20; 2.7 g Na 2HPO a · 5 H 20; 2.0 g Na2HJl20S' 6 H 20; 1.4 g
KH 2P0 4 ; 1.5 g NaAs0 2; 2.3 g Na 2HAs0 4 • 7 H 20; - - - 8.0 g Na 2S· 9 H 20;
1.3g Na 2S 20 S ; 2.2g Na 2S 2 0 a ' 5H 20; 1.8g K 2S0 4 ; l.4g K 2S 20 S ; 2.5g
Na 2Se0 4 · lO H 20; - - - 2.0 g NH4F; 1.7 g NaCl; 1.5 g CaCI· CIO + 1.5 g
Na 2COd 1.3 g NaClO s ; 1.2 g NaCI0 4; 1.3 g NaBr; 1.3 g KBrO s ; 1.3 g KI;
1.2 g KIO a ; 1.1 g NaI0 4 •
Dissolve in 100 ml 3-F HOl:
16 ml 20% TiCld - - - 2.1 g RuCl a ; 1.7 g PdCI2; 1.6 g IrCla ; - - -
LOg Ge;3 1.9g SnCI2 • 2H 20; - - - 1.9g SbCla ; 1.3g BiOCI; - - - 1.3g
Te0 2 •4
1 Separate from the precipitated CaCOa.
2 Pass air through the solution until it is colorless.
3 First warm with 5 m16-F HCI and 1 ml 6-F HNO a, then dilute with 3-F HCI
to 100 ml.
4 First warm with 5 ml 6-F HCI until dissolved, and then dilute with 3-F HCI
to 100 ml.
 Preparation of Unknowns 431

Treat with 5 ml 6-F HNO s; when dissolved, dilute with water to 100 ml:
1.5 g SC 2 0 3 ; - - - 4.7 g Zr(NOs),' 5 H 20; - - - 1 g Ga; 1 gIn.
Dissolve in 100 ml 0.3-F NaOH:
2.3 g NH,V0 3 •

Preparation of Unknowns
As a rule, all information needed by the student is supplied in the
text of the experiment, and additional information restricting the field
of the search should not be given.
Experiment 42. The experiment is described for the milligram scale,
but it should also be carried out when working on either the gram or the
centigram scale. The directions are readily converted by multiplying all
volumes by 1000 or 10, respectively. The preparation of the unknown
is described with the experiment, but the volumes taken should not be
disclosed and should be varied from sample to sample. A relative agreement
of ± 0.3 between estimated and given amount should be considered
satisfactory.
Experiment 63. Directions are given with the experiment. The ratio
should be varied and it should not be disclosed before all reports have
been handed in.
Experiment 65. Test solution is diluted up to 10 times with the solvent
used in its preparation, see preceding section.
Experiment 66. The quantity to be given is discussed under Expt. 66.
Suitable unknowns are the various oxides, sulfides, carbonates, insoluble
halides and sulfates of Ag, Hg(l, 2), Pb, Bi, Cu, Cd, As(3, 5), Sb(3, 5),
and Sn(2, 4). A mixture of two modifications of a compound may be
given, but each sample should contain only one reasonably pure compound.
Experiment 67. Two different solid samples in the amount specified
in Expt.66. Recommended are water-insoluble compounds, oxides,
hydroxides, sulfides, sulfites, carbonates, phosphates of Fe(2, 3), AI,
Cr(3), Ni, Co, Mn, Zn, and the metals listed for Expt. 66. Minerals and
pigments fitting into this list are also recommended.
Experiment 68. Two solid-samples in the amount specified in Expt. 66.
Recommended are simple inorganic substances (hydrated salts), melting
below 300 0 C and having the m. pt. listed in the handbook used, and
reasonably pure organic solids, having a good melting point, which are
included in the identification tables of KOFLER (159, 160) or McCRONE (163).
These tables must be available for students' use in addition to some hand-
book of chemistry.
Experiment 69. Two solid samples as in Expt. 68. This time, it is
suggested to include abrasives (various forms of carbon, carbides, quartz,
432 Appendix

corund) and to select otherwise compounds which will suitably round

out the experience of the student.
Experiment 70. The study may be extended in any desired direction
by including inorganic and organic liquids, metals, alloys, complex
compounds, mixtures, substances which defy identification beyond recogni-
tion of the group of materials to which they belong, substances which have
not been described in the literature, and organized material derived from
living matter. It is understood that such work will require access to the
literature.

Reagents
The necessary reagents are listed in connection with the experiments
and directions. The concentrations are usually specified in terms of
formality according to the definition that a one-formal, I-F, solution
contains one gram formula weight (corresponding to the formula given)
per liter of solution.
Murrayite and Xyrax, obtainable from laboratory supply houses, may
be substituted for Oanada balsam. Dekadhese (Arthur H. Thomas 00.,
Philadelphia, Pa.) is recommended for cementing glass, varnishing labels, etc.

Table I
Color of Some Inorganic Substances
Black (gray): metal powders, B, 0, Si, As, Se, Te, 12 - - V 20 a, M0 20 a,
Mo0 2, U0 2, Mn 20 a, MnaO" Mn0 2, FeaO" FeO· Or 20 a, 0020a, OOaO"
Ni 20 a, NiaO" OuO, TI 20 a, Pb 20, and many oxidic minerals such as Ti0 2
(anatase, brookite, rutile), (Th, U)02 (thorianite), pitchblende, Fe 20 3
(martite), Si0 2 (quartz), Sn0 2 (cassiterite) - - V 2S 2, V2 Sa, MoS 2 , MoS a,
US 2 , U 2Sa, U0 2S, MnS 2, FeS, OoS, NiS, Ou 2 S, Ag 2S, Au 2S, Au 2Sa, Hg 2S,
HgS, TI 2S, T1 2S3 , PbS, Sb 2S3 , various sulfidic minerals - - FeOI3 , OuBr 2,
various triple salts, etc. - - permanganates - - boron carbide and the
refractory nitrides, carbides, and borides of Si, Ti, Zr, Hf, Mo, and W --
various silicates, obsidian, lava, slags - - opaque materials (of any surface
color) in transmitted light.
Dark brown: W0 2, Mn 20 3 , Mn 30" Fe 20 a · x H 20, 000, OdO, T1 20 3 ,
Pb0 2 , Bi 20 a - - MoS a, OoS.
Red: Se, P" Ou - - OrO a, Fe 20 3 , HgO, Pb 3 0, - - Od(S, Se), HgS,
Pb 2S01 2 (~), AS 2S 2, AS 2S3 - - chromates, compounds of Or(3) and 00(2),
iodides, Hg1 2.
Pink: As 2S 2 , MnS - - compounds of Or(3), Mn(2), 00(2), Nd, Eu, Er.
Orange (light brown): Ou, bronze - - Ou 20, HgO, PbO - - Fe(OH)a,
TI(OH)3' TIO(OH) - -: OdS, AS 2 S 2, As 2Sa - - compounds of Fe(3), Oe(4),
and of platinum metals.
 Substances Crystallizing in the Cubic System 433

Yellow: Au, brass, bronze, S, P 4, As - - Na 20 2, WOa, HgO, In 20 3,

PbO, Pb 20 a, Sb 20 5 , Bi 20 a, Bi 20 4 · 2 H 20 - - CdS, Hg 2SCl 2 (n, SnS 2.
AS 2 S3, AS 2S5 , FeS 2 (pyrite) - - Ag aP0 4 , AgAs0 2 , - - - Compounds of V,
uranyl, Fe(3), platinum metals, Au, TI(3}, Sm, Dy, Ho - - chromates,
ferrocyanides, ferricyanides, bromides, iodides.
Green: UaOs, MnO, NiO, (Zn, Co}O known as Rinman's green --
.MnS - - compounds of Cr(3), Fe(2}, Ni, Cu, Pr - - manganates.
Blue: AI 20 a · x CoO known as Thenard's blue, Nd 20 3 - - ultramarine,
Prussian blue - - compounds of Cr(3}, Co(2}, and Cu.
Violet and purple: 12 - - ultramarine - - compounds of Ti(3}, Cr(3),
Co, Nd - - permanganates .
.1lfetallic luster: Metals, Te - - FeTiO a (ilmenite), titanates, tantalates,
Mn0 2 · x H 20 (pyrolusite), Fe a0 4 (magnetite), (Fe, Mn, Zn}O . Fe 2 0 a (frankli-
nite), FeO . Cr 20 a (chromite), FeW0 4 (ferberite), (Fe, Mn}W04 (wolframite),
CuO (tenorite) - - most sulfide, selenide, telluride, arsenide, and stibnide
minerals (ores) - Ca(F, CI)2' Ca 4 (P04)a (apatite) - - mica - - tungsten
bronzes.

Table 2
Substances Crystallizing in the Cubic System
Element8
Li, Na, K - - - Ca - - - Ce - - - Ti, Zr, Th - - - V, Nb, tantalum:
Ta - - - Cr, Mo, W - - - ferrite: Fe; Co, Ni - - - Rh, rhodium gold:
(Rh-Au), palladium: (Pd-Pt-Ir), platinum: Pt, iridium: (Ir-Fe-Pt)
- - - copper: Cu, silver: Ag, gold: Au, perpezite: (Au - Pd) - - - amalgam:
(Ag-Hg) - - - Al - - - diamond: C; Si, Ge, Sn, Pb - - - P 4 (red), As 4 •

Inorganic Oompound8
Hydride8: LiH, NaH, KH, N 2H 4 • 2 HCI, PH 4Cl.
Halides, Oyanide8, Thiocyanate8: LiF, LiCI, LiBr, LiI; villiaumite:
NaF, halite: NaCI; NaBr, NaI, NaCN; KF, KHF 2, sylvite: KCI; KBr,
KI, KCN; RbF, RbCI, RbBr, RbI; CsF, CsCI, CsBr, CsI; sal ammoniac:
NH 4CI; NH 4 Br, NH 4I, NH 4CN - - - fluorite: CaF 2; CaCI 2, Ca(CN)2; SrF 2,
SrCI 2; BaF 2' BaCl 2 - - - yttrofluorite: CaF 2 . n YF 3' yttrocerite: CaF 2 .
. n (Y, Ce)F3 - - - TiI4 - - - VOBr - - - WBr 2, WCI 6 , UCl 4 - - - MnCl 2
- - - Fe(CNS)a' 3 H 20 - - - OsCI 3, PdCl 2 - - - Cu 2C1 2, Cu 2I 2; AgF,
cerargyrite: AgCI, bromyrite: AgBr, embolite: Ag(CI, Br) - - - ZnI 2;
CdF 2 , CdCI 2; HgF, HgF2' eglestonite: Hg 4CIO - - -ralstonite: 3 AI(F, OH}a'
. (Na 2Mg}F2' 2 H 20; TIF, TICI, TlBr, TIl - - - SiI 4; GeBr4' GeI 4 ; SnI4
- - - PSBr a·
Oxide8: K 2 0, Rb 20, Rb 20 2 - - - periclase: MgO; CaO, SrO, BaO - - -
Ce0 2 , Y 20 a, DY20a, Er 20 3, Yb 20 a - - - Th0 2 - - - NbO - - - W0 2;
Benedett,i·Pichler, Identification 28
434 Appendix

thorianite: (Th, U)02 - - - manganosite: MnO; Mn 20 3 - - - martite:

Fe203, magnetite: Fe30 .. ; CoO, Co 30 .. , bunsenite: NiO - - - cuprite:
Cu 20; CuO; Ag 20 - - - CdO - - - In 20 3 - - - cristobalite: Si0 2; SnO
- - - arsenolite: As 20 S ; Sb 20 3 ; Bi 20 3 •
Hydroxides and Acid8: Pb 30 2(OH)2, H 7P(Mo 20 7)a' 28 H 20, H 2TeO .. ·
. 2 H 20.
Sulfide8, Selenides, Telluride8: Li 2S, Na 2S4 - - - BeS, MgS, CaS, SrS,
BaS - - - alabandite: MuS, hauerite: MnS 2, MnSe - - - pyrite: FeS 2;
CoS 2, cobaltpyrite: (Co, Fe)S2' linnaeite: C0 3S4 , cobaltite: Co 2As 2S 2;
polydymite: Ni 3 S.. , gerstorffite: Ni 2As 2S 2; NiSe - - - laurite: (RU,OS)S2
- - - Cu 2S, berzelianite: Cu 2Se; argentite: Ag 2S, aquilarite: Ag 2 S-Ag 2 Se;
Ag 2Se, eucairite: Cu~g2Se2' naumannite: (Ag 2Pb )Se; hessite: Ag 2Te,
petzite: (Ag, AU)2Te - - - sphalerite: ZnS; ZnTe, CdTe, tiemannite:
HgSe - - - galena: PbS, clausthalite: PbSe, altaite: PbTe, stannite:
FeS . CU2S . SnS 2, ullmannite: NiSbS, bornite: Cu5FeS .. , tennantite:
5 Cu 2S . 2 ZnS . 2 As 2 SS ' tetrahedrite: 3 (Cu 2, Fe, Zn)S . Sb 2 S3, argyrodite:
AgS GeS 6, germanite: 5 Cu 2 S· 12 (CuFe)S· AS 2S3 · 2 GeS 2.
Nitrides, Amides, Pho8phide8, Ar8enide8: LiNH2 - - - arsenoferrite:
FeAs 2, smaltite: CoAs 2 - - - sperrylite: PtAs 2 - - - Cu5As 2 - - - Zn 3P 2'
Zn3As 2, Cd3 As 2 - - - BN (borazon).
Oarbide8, Silicide8: Be 2C, Mg 2Si - - - TiC - - - WC - - - Mn 2Si - - -
FeSi (octahedraH).
Boride8: CaB 6, BaB a•
Ohlorate8, Bromates, Iodates: NaCIOs, NaBr0 3, RbI0 3 - - - Co(CIOs)2 .
· 6 H 20, Co(Br0 3 )2' 6 H 20, Co(I0 3)2' 6 H 20, Ni(BrO S )2' 6 H 20 - - -
Cu(CIO s)2' 6 H 20, Cu(Br0 3 )2' 6 H 20 - - - Zn(BrO S)2' 6 H 20, Hg(Br0 3)2'
. 6 H 20.
Thio8ulfate8, Sulfate8: K 2S 20 3 - - - Cr 2 (SO .. )3· 18 H 20 - - - NiSO ..
- - - (the alums:) (K, Rb, Cs, NH 4, TI) (Al, In, Ga, TI, Fe, Cr, V, Rh) (SO .. )2·
.12 H 20; tschermigite: NH 4Al(SO .. )2· 12 H 20.
Nitrates, Hypopho8phite8, Pho8phate8, Pyropho8phate8, A r8enate8 , Anti-
monates: (phosphamic acid:) (OH)2PNH2 - - - 2 NaSb0 3 · 7 H 20, Na3SbS .. ·
.9 H 20, RbN0 3 - - - Ca(N0 3)2' Sr(NO S)2' nitrobarite: Ba(N0 3)2; Ba 3(P0 4 )2
- - - TiP 207 - - - Ni(H 2P0 2)2 . 6 H 20 - - - Ag3PO.. , Ag3AsO.. - - - TIN0 3.
Silicates: hackmanite, sodalite: 6 NaAlSi0 4 · 2 NaCI, noselite:
6 NaAlSiO .. · NaSO .. , lazurite: 3 NaAISiO.. · Na 2S, hauynite: 3 NaAlSiO .. ·
· CaS04 , analcime: NaAlH 2 Si2 0 7, chalcolamprite, endeiolite: Na..Ca~b2-
(F, OH)2Si09 (1), pollucite: CS 4Al 4H 2 (SiOs)g - - -helvite: 3 (Be, MH, Fe)SiO .. ·
· MnS, danalite: 3 (Be, Mn, Fe, Zn)SiO .. · ZnS, garnet: (Mg, Ca, Mn, Fe)a(Cr,
Fe, AlMSiO.. )3, beckelite: Ca3(Ce, La, DY)4Si3015, schorlomite: Ca 3(Fe, Ti)2'
.3 (Si, Ti)O.. - - -rowlandite: Fe(YFMY, Ce, LaMSi 20 7 )2' bodenbenderite:
Y -Al-Mn-etc. titanosilicate; greenalite: Fe 3H 2(Si0 3)4 . H 20 - - -zunyite
AI 2[Al(OH, F, CI)2]6(SiO.. )s - - - eulytite: Bi 4 (Si0 4 )3'
 Substances Crystallizing in the Hexagonal System 435

Borates, Aluminates: boracite: 6 MgO . MgCl z . 8 BZ03' spinel: Mg(AlOz)z,

picotite: (Mg, Fe)· (Al, Cr0 2)2; Ca(BOz)z' 2 H 20, rhodizite: (Li, Na, K)s'
. Be7Al6B14039 (1) - - - ceylonite: (Mg, Fe)O· (AI, Fe)20a, hercynite:FeO·
. Al 20 S - - - gahnite: Zn(Al0 2)z'
Salts of other Oxygen Acids: pyrochlore: (Na z, CaMNb, TiMOFh,
knopite, perovskite: (Ca, Ce, Fe)TiOs, dysanalyte: (Ca, FehNb z(Ti0 4)6 (1),
uhligite: Ca(Zr, Ti)206' (AI 20)Ti0 4, davidite: rare earths-V-Cr-U-Fe
titanate; samiresite: U -Pb-etc. niobate-titanate; ellworthite: Ca-U -Ti-
etc. niobate; betafite: hydrous Ca-U -etc. titanate-niobate; mendelyeevite:
Ca-U titano-niobate; zirkelite: Ca-Fe titano-thorio-zirconate, thorianite:
(Th, U)Oz - - - microlite: hydrous Na-Ca-rare earths niobotantalate-
fluoride; koppite: Ca ZTa 20 7 with Na, Ce, Fe, F, etc., neotantalite: hydrous
Mn-Fe niobotantalite, djalmaite: U niobotantalate - - - chromite:
Fe(Cr0 2)2; KsCrOs, K 2W 40 13 • 8 H 20, (NH4)zW40 13 • 8 H 20; (tungsten
bronzes:) Me",W0 3 - - - cleveite, nivenite, uraninite: complex uranates
with Y, lanthanides, Zr, Th, Pb, N, He, A; broggerite: U-Th-Pb uranate
- - - Sr(Mn0 4)2' 3 H 20 - - - franklinite: (Mn, Fe, Zn)(Fe0 2)z - - -
K 20s04 · 2 H 20 - - - lewisite: Ca5Ti 2(Sb0 4)6; mauzeliite: Ca-Pb titano-
antimonate.
Fluoborates, Fluoraluminates, Fluosilicates: cryolithionite: LiaNaa(AlF 6)2;
KBF 4' hieratite: K 2SiF 6; Rb 2SiF 6' Cs 2SiF 6, (NH 4)2SiF 6 - - - Ag 2SiF 6 •
. 4 H 20 - - - Tl 2SiF 6.2 H 20.
Ohloro-, Bromo-, Iodo-, and Oyanocomplexes: K zPdCI4, (K, NH4)2PdClo;
(K, NH4)20sCl6' K 20sBr 6; (K, NH4)21rC16; (K, Rb, Cs, NH 4)zPtCI 6,
(K, NH4)2Pt(Br, 1)0 - - - KAg(CN)2 - - - CsHgCIs - - - (K, NH4)2SnCl6,
(K, NH4)2PbCI6 - - - (NH4)2SeBrs'
Neutral Oomplex: Pd(NHs)z(OH)2'

Organic Oompounds
Calcium oxalate, anh., CaC 20 4 - - - carbon tetraiodide, C14, red, dec.
- - - cinchonine bisulfate, octahedral - - - mercuric fulminate, Hg(CNO)2
- - - morphine sulfate pentahydrate, 250 0 dec. - - - tri-iodoethane,
CH3 • CIa, yellow, octahedral, 95 0 dec.

Table 3
Substances Crystallizing in the Hexagonal System
Elements
Cs - - - Be, Mg - - - Y, Ce - - - Hf - - - Re - - - Ru, Os, ruthen-
osmiridium: Ru-Os-Ir alloy, iridosmine: Ir-Os with Pt, Rh, Ru, etc.
- - - zinc: Zn; Cd - - - graphite: C - - - P 4, arsenic: As, antimony:
Sb, bismuth: Bi - - - Se, tellurium: Te.
28·
436 Appendix

Inorganic C01npoundS
Halides: NH,F - - - nocerite: MgaCaaO2F S' chlormagnesite: MgCI 2;
MgBr2' 6H 20;tachyhydrite: 2 MgCI 2· CaC~ . 12H20; yttrocalcite: CasY 2F 16 ;
CaCl z ' 6 H 20, CaBr2 · 6 H 20, Ba1 2 · 6 H 20 - - - ZrF, - - - LaFs, CeF,'
· H 20, NdFa, fluocerite: (Ce, La, Dy)Fa - - - VCl 2 - - - UF a, UCls - - -
CrCla, CrBrs, CrCls ' 6 H 20, CrBrs ' 6 H 20 - - - FeCI 2, FeBr 2 , Fe1 2, FeCIs ;
CoF 2 · 5 HF· 6 H 20, CoFa, Co1 2 · 6 H 20; NiF 2 · 5 HF· 6 H 20 - - - AgI
- - - Cd1 2, kleinite: Hg,OaCI2 - - - AlF3' AlCla, AlOia' 6 H 20; TI 2Cla,
TICl3 - - - Si 21 6 ; Pbl z - - - P13 , P1 2Cla; As1 3 , Bila·
Oxides: Na 20 2 · 8 H 20 - - - BeO - - - Nd 20 a - - - Ti 20 S - - - Cr20 S
- - - hematite: Fe 20 S , hogbomite: chiefly (Al, Fe)203, MgO, Ti0 2 - - -
zincite: ZnO - - - corundum: Al 20 3 ; In 20 s, Tl 20 3 - - - amethyst, quartz,
tridymite: Si0 2.
Hydroxides, Acids: brucite: Mg(OH)2; Ca(OH)2 - - - H~004 - - -
Mn(OH)2 - - - Cd(OH)2 - - - TI(OH)a - - - H 2SeO a, H 2SeO,.
Sulfides, Selenides, Tellurides: molybdenite: MoS 2 - - - FeS, FeaS4,
pyrrhotite: FeS S 6 to Fe16S1?; millerite: NiS, melonite: NiTe 2 - - - covellite:
CuS - - - wurtzite: ZnS; ZnSe; greenockite: CdS; CdSe; cinnabar: HgS
- - - Al 2S3 - - - SnS 2 - - - AgaAsSa, pyrargyrite: AgsSbS 3; tetradymite:
Bi 2(S, Te)a·
Nitrides, Arsenides, Antimonides: NaNs (azide) - - - CrAs - - -
niccolite: NiAs; NiSb - - - Cu3 As - - - BN (white graphite); AlsN3'
Carbides: Be 2C - - - CeC2, NdC 2, SmC 2 - - - moissanite: SiC.
Chlorates, Bromates, Iodates, Perchlorates, Periodates: LiCIO,' 3 H 2 0;
NaCI0 3 , NaCIO,' H 20, NaIO,' 3 H 20; KBr0 3 - - - Ba(CIO')2' Ba(CIO,)2'
· 3 H 20 - - - Y(Br0 3h' 9 H 20, La(BrOa)a' 9 H 20, Ce(BrO a)3' 9 H 20,
Pr(BrOa)a' 9 H 20, Nd(BrOsh' 9 H 20, Sm(BrOa)a' 9 H 20, Dy(BrOa)a' 9 H 20
- - - Co(Cl04)2' 6 H 20; Ni(CI0 4)2' 5 H 20, Ni(CI0 4)2' 5 H 20, Ni(IO a)2'
. 4 H 20.
Sulfites, Dithionates, Sulfates, Selenates: Na 2SO a, Na 2S04, hanksite:
9 Na 2S0 4 · 2 Na 2CO S ' KCI; K 2S20 6 - - - MgSO,' 6 H 20; CaS 20 6 • 4 H 20,
SrS 20 6 • 4 H 20 - - - La 2(SO,h . 9 H 20, Yb 2(SeO,h . 8 H 20 - - - Fe 2(S04)3 .
. 9 H 20 - - - alunite: K 2Al 6 (OH)12(SO,), - - - Sn(SO')2' 2 H 20; PbS 20 6 •
. 4 H 20.
Nitrite8, Nitrates, Hypophosphites, Phosphates, Arsenites and Arsenates:
LiNO a, LisPO,' 12 H 20; nitratite: NaNOs; NasPO,' 12 H 20, NaaAsO,'
· 12 H 20; KNO a, KH 2P0 2, K 2HAsO, . H 20; RbNO a; CsN0 3 - - - Ca(N0 2 )2 .
· H 20, apatite: Cas(F, Cl)(PO,)a; Ba(N0 2)2' H 20 - - - buttgenbachite:
hydrous copper chloronitrate; Ag 2HPO, - - - AIPO,; stiepelmannite:
(Y, La, Ce, Pr, Yb)Al s(OH)6(PO,)a; florencite: Ce 2Al 6 (OH)12(PO,),; TINOa
- - - Pb(N0 3 )2' Pba(PO,)2' pyromorphite: 3 Pb 3(PO')2' PbCI 2 ; plumbo-
gummite: PbAla(OHMPO,)2' H 20; Pb(AsO a)2'
 Substances Crystallizing in the Hexagonal System 437

Carbonates: Na 2CO a · 7 H 20 - - - magnesite: MgCO a; calcite: CaCO a ;

dolomite: MgCa(CO a)2; codazzite: (Mg, Ca, Ce, Fe)CO a ; parisite: Ca(La,
Ce, DY)2F 2(CO a)a; BaCO a, cordylite: Ba(La, Ce, DY)2F 2(CO ah - - - basna-
site: (La, Ce, Dy)FCO a - - - Na 4U0 2 (CO a)a, andersonite: Na ZCaU0 2(CO a)a'
.6 H 20 - - - rhodochrosite: MnCO a - - - stichtite: Mg6Cr2(OH)16COa'
. 4 H 20 - - - siderite: FeCO a ; ankerite: MgCa 2Fe(CO a)4; CoCOa - - -
smithsonite: ZnCOa ; CdCO a - - - Pb a(OH)2(CO a)2'
Silicates, Stannates: eucryptite: LiAISi0 4 ; tourmaline: complex
Li-Na-Mg-Fe-Al borosilicate; Na 2 Si0 4 , nepheline: (Na, K)AlSi0 4 ;
gmelinite: (Na z, Ca)AI 2H 4(Si0 4)a' 4 H 20; chabazite: (Na 2, Ca)AI 2 (SiO a)4'
.6 H 20; cancrinite: NasCaAls(Si0 4 MCO a)2; bazzite: Na-Sc-rare earths-
Fe silicate; eucolite, eudialyte: Na-Ca-Ce-Zr-Fe basic chloride-silicate;
steenstrupine: hydrous Na-rare earths-Ti-Zr-Th-Mn-Fe-Alsilicate;
Na 2SnO a ; milarite: hydrous K-Be-Ca-AI silicate; Na 2 SnO a · 3 H 20;
KzSnO a · 3 H 20 - - - phenakite: Be 2Si0 4 ; beryl: Be aAI 2(SiO a)6; helidor:
Bea(Fe, Al)2(Si0 4 )3 (?); levynite: CaAI 2H 4 (Si0 4 h' 3 H 20; abukumalite:
Ca-Y phosphate-silicate; melanocerite: Ca-Y -La-Ce-Dy-Ta borate-
silicate; caryocerite: similar to melanocerite with Th added; SI'SiO a ;
benitoite: BaTi(Si0 3 )a; cappelenite: Ba-Y borate-silicate - - - buszite:
Pr-Eu-Er-Nb silicate - - - dioptase: CuSi0 3 • H 20 - - - willemite:
ZnSi0 4 •

°
Bo,rates,' NaB0 2; K 4B 20 7 · 5 H 20.
Salts ot Other Oxygen Acids,' geikielite: (Mg, Fe )Ti 3 ; pyrophanite:
MnTiO a ; senaite: (Mn, Fe, Pb)TiO a ; ilmenite: FeTiO a - - - Na 4V 20 7;
vanadinite: 3 Pb a(V0 4)2 • PbCl 2 - - - hatchettolite: U(Nb, Ta)Oa . H 2 0 - - -
ferritungstite: hydrous ferric tungstate - - - BaMn04 - - - mimetite:
3 Pb a(As0 4 )2' PbCI 2.
Fluosilicates, Fluogermanates,' Na 2 SiF 6; K 2SiF 6, K 2GeF 6 - - - tritomite:
Ca-Ba-Y-La-Ce-Dy-etc. fluosilicate - - - MnSiF 6 · 6 H 20 - --
CoSiF 6' 6 H 20; NiSiF 6 • 6 H 20 - - - ZnSiF 6' 6 H 20.
Chloro-, Bromo-, and Iodocomplexes: MgPdCI 6 • 6 H 20, NiPdCI 6 · 6 H 2 0;
Li 2PtCI 6 • 6 H 20, MgPtCI 6 · 6 H 20, MgPtBr 6 ' 12 H 20, MnPtC1 6 • 6 H 20,
MnPtBr 6 • 12 H 20, MnPtI 6 · 9 H 20, FePtC1 6 • 6 H 20, CoPt(CI, Br)6' 12 H 20,
CoPtI 6 • 9 H 20 - - - K 3 Cu(CN)4' KAg(CN)2' KaAg(CN)4' KAu(CN)2 - - -
MgSnCI 6 · 6 H 20, CoSnCI 6 · 6 H 20.

Organic Compounds
Aconitine hydrobromide, sint. 160° - - - barium acetate monohydrate
- - - benzil, (C 6H 5 =CO)2, 95° - - - benzoylcarbinol, C6H 5 • CO· CHaOH,
85-86° - - - camphor (d), 93-94° - - - cyclotetramethylenetetranitramine
IV, I at 279° - - - glycerol diphenylether, 80-810 - - - iodoform, yellow,
119° - - - methionine (l), 283° dec. - - - methylnaphthoquinone monoxime,
166-168° - - - mudarol, Cao H 47 0(OH), 176° - - - threonine (d), C4H 9 0 aN,
438 Appendix

225-227 0 dec. - - - triethylamine hydrobromide, 248 0 - - - triethyl-

phosphine sulfide, (C2Hs)3PS, 94 0 - - - triphenylacetic acid, dec. - - -
tryptophane (1), 289 0 - - -valdivin (in), glucoside, 230 0 dec. - - -ytterbium
acetate tetrahydrate, Yb(CH3' COO)3' 4 H 20.

Table 4
List of Common Inorganic Compounds in the Order of Their
Melting Points
Temperatures are given in degrees Celsius (centigrade); in general,
only those compounds are listed, which melt below 900. In addition,
the boiling points (b. pt.), transition points (tr.) and temperatures of
sublimation (subl.) or volatilization (vol.) are given. The color refers to
the solid substance.
Melting Point,
15.5 NaOH· 3.5 H 20, monoclinic
16.83 SOa, prisms, b. pt. 44.6
17 Na 2Mn0 4 • 10 H 20, green, monoclinic
17 MoF 6' crystals, b. pt. 35
19.2 H 2SnCI 6 • 6 H 20, deliquescent
19.9 Na 2Cr0 4 • 10 H 20, yellow, monoclinic, deliquescent
22 NaHS . 3 H 20, orthorhombic, decomposes
22 Pb(CH aCOO)2' 10 H 20, orthorhombic
22.5 P 20a, monoclinic, deliquescent, b. pt. 173
25 PBrsCla, brown needles
25 SrS 4 • 6 H 20, pink
25 TICl a, hexagonal plates, decomposes
25.5 Ru0 4 , yellow, orthorhombic
25.7 BaI 2 • 6 H 20, hexagonal
25.8 Mn(NO a)2' 6 H 20, rose, monoclinic, b. pt. 129.5
26 H 2Se0 4 • H 20, needles, b. pt. 205
26.1 GeBr 4 , gray, octahedra
26.5 H aP0 2 , decomposes
27 HAuBr 4 • 5 H 20, brown
27 FeBr 2 · 6 H 20, red, orthorhombic
27 FeBra · 6 H 20, red
27.2 Iel, red, cubic, b. pt. 97
28 Na 2HAsO,' 12 H 20, monoclinic, -12 H 20 at 100
28.5 Cs, silvery, hexagonal, b. pt. 670
29.75 Ga, gray, b. pt. 1983
29.88 LiNO a · 3 H 2 0
29.92 CaCI 2 • 6 H 20, trigonal, -6 H 20 at 200
30 N 20S' orthorhombic, b. pt. 47
List of Common Inorganic Compounds in the Order of Their Melting Points 439

30 Bi(NOa)a' 5 H 20, triclinic, decomposes at 30, -5 H 20 at 80

30 MnS0 4 • 4 H 20, pink, orthorhombic, monoclinic, -1 H 20
at 30, -4 H 20 at 450
31.0 SnBr4, orthorhombic, deliquescent, b. pt. 202
31 AsBr a, prisms, b. pt. 221
32.4 NaS0 4 · 10 H 20, monoclinic, -10 H 20 at 100
33 ICla, yellow to red-brown, orthorhombic, b. pt. 77, decomposes
33 KFe(S04)2' 12 H 20, colorless or violet, cubic
34 NH 20H, orthorhombic, deliquescent
34 Na 2COa · 10 H 20, monoclinic, -H 20 at 106
34 CdCI 2 • 2.5 H 20, monoclinic, tr. at 34
34.6 Na 2HP0 4 · 12 H 20, monoclinic, -12 H 20 at 180
35 Fe(N0 3 )3 • 6 H 20, orthorhombic, deliquescent, decomposes
35 H 2 S 20 7 , crystals, decomposes
35 Mg(CI0 3 )2' 6 H 20, deliquescent, boils and decomposes at 120
35.5 H 3As0 4 · 0.5 H 20, hygroscopic, -H 20 at 160
36 NH 4CN, cubic
36.4 Zn(N0 3 )2' 6 H 20, tetragonal, -6 H 20 at 105
36.5 Cr(N0 3 )3 • 9 H 20, purple prisms, decomposes at 100
37 FeCl3 • 6 H 20, brown, deliquescent, b. pt. 280
37 TICI3 • 4 H 2 0, needles, -4 H 20 at 100
37.5 La 2 (Br0 3 )6' 18 H 20, hexagonal, -14 H 20 at 100
37.7 SnCI 2 • 2 H 20, triclinic, decomposes
38 PSBr3 , yellow, cubic, decomposes at 175
38 H 4P 2°5' needles, decomposes at 130
38.2 CaBr 2 · 6 H 20, monoclinic, boils 149 to 150
38.5 Rb, silvery, b. pt. 700
39 TiBr4' amber, deliquescent, b. pt. 230
39 ZnS0 4 · 7 H 20, orthorhombic, transition at 39, -7 H 20 at 280
40 TlBr3 • 4 H 20, yellow, deliquescent
40 SeS0 3 , green prisms, decomposes with lib. of S02 at 40
40 NH 4Fe(S04)2' 12 H 20, violet, cubic
40 H 2TeO a, orthorhombic, monoclinic, gives off H 20 at 40
40 OS04, monoclinic, b. pt. 135
40 La(N0 3)3 . 6 H 20, triclinic, deliquescent, boils at 126
41 KF· 2 H 20, monoclinic prisms
41 Ba(CH3COO)2' H 20, triclinic prisms, -H 20 at 41
41.5 3 CdS0 4 . 8 H 20, monoclinic, transition at 41.5
41.7 SeOBr 2 , yellow crystals, b. pt. about 220
42 2 NaH 2POa . 5 H 20, monoclinic prisms, -5 H 20 at 100
42 Ca1 2 • 6 H 20, deliquescent, -H 20 at 160
42 ± BrI, dark gray crystals, b. pt. about 116
42.35 H aP0 4 , orthorhombic, -0.5 H 20 at 213
440 Appendix

42.7 Ca(NO S)2· 4 H 20, monoclinic, -H 20 at 42.7

43 BIs, plates, b. pt. 210
44 LiBr . 2 H 20, prisms
44.1 P 4, yellow, hexagonal, b. pt. 280
45 KI s, dark blue, monoclinic, deliquescent, decomposes at 225
45 to 46 Be(stearate)2' waxy
47 Fe(NOs)s· 9 H 20, pale violet, monoclinic, deliquescent, decom-
poses at 100
47 Na 2 SiO s · 9 H 20, orthorhombic, -6 H 20 at 100
48 NH 20H . HNO s, crystals, decomposes below 100
49 Ce(BrOS)3· 9 H 20, hexagonal, melts with decomposition
50 Sea, red powder, amorphous, b. pt. 688
50 HCI0 4 · H 20, needles, melts with decomposition
50 (SOS)2, silky needles
50 Na 2S· 9 H 20, deliquescent crystals, melt decomposes
50.7 NaBr· 2 H 20, monoclinic
51 CO(CO)4, orange crystals, decomposes at 52
53 Na 2HPO s · 5 H 20, orthorhombic, deliquescent, -5 H 20 at 120
53.3 NiS0 4 · 6 H 20, green monoclinic or blue tetragonal, transition
at 53.3, -6 H 20 at 280
55 H 4P 20 6 , vitreous, deliquescent, decomposes at 70
56 POBrs, plates, b. pt. 190
56.5 Pr 2(BrO S)6· 18 H 20, green, hexagonal, -14 H 20 at 100
56.7 Ni(NOs)2· 6 H 20, green, monoclinic, boiling at 136.7
57 CO(NOS)2· 6 H 20, red, monoclinic, melt decomposes
57 S20sC14, orthorhombic, melt decomposes, sublimate
57 NaB0 2 • 4 H 20, monoclinic
58.0 MnCI 2 · 4 H 20, rose, monoclinic, deliquescent, -H 20 at 106,
-4 H 20 at 200
58 H 2 Se0 4, hexagonal prisms, b. pt. 260
58 NaCHsCOO· 3 H 20, monoclinic, -3 H 20 at 120
58.5 NaHS0 4 · H 20, monoclinic, deliquescent, melt decomposes
59.4 Cd(NO s)2· 4 H 20, needles, -H 20 at 132
below 60 H 2 S 20 a, hygroscopic crystals
60 H 2PtCI 6 • 6 H 2 0, red to brown, deliquescent
60 Be(NO s)2· 4 H 20, deliquescent, decomposes at 100
60 Zn(ClO s)2· 6 H 20, monoclinic, deliquescent, melt decomposes
60 NaH 2P0 4 · 2 H 20, orthorhombic
60 KBr·IBr, orthorhombic, decomposes at 180
60 K 2 S . 5 H 20, orthorhombic, deliquescent, -3 H 20 at 150
60 KCI· ICI, monoclinic, decomposes at 215
60.2 U0 2(NOs)2· 6 H 20, yellow, orthorhombic, boils at lI8
60.5 Fe(NO s)2· 6 H 20, crystals
List of Common Inorganic Compounds in the Order of Their Melting Points 441

61 00(010 3 )2. 6 H 20, red, cubic, decomposes at 100

61 00(1° 3 )2. 6 H 20, red, cubic, melt decomposes
61 PIa, red, hexagonal, deliquescent, melt decomposes
61 NaAl(S04)2· 12 H 20, cubic
61 H 4P 2°7, needles
61 SrOI 2 · 6 H 20, orthorhombic, -4 H 20 at 61, -6 H 20 at 100
62.3 K, b. pt. 760
62.3 AI(BrOa)a· 9 H 2 0, hygroscopic, melt decomposes at 100
64 FeS0 4 • 7 H 20, green, monoclinic, -7 H 20 at 300
64.3 MnBr 2 · 4 H 20, red, monoclinic, deliquescent, melt decomposes
65 Ou(010a)2· 6 H 20, green, cubic, deliqnescent, decomposes at 100
65 N 2H4 . HNa deliquescent
65 LiOIO a · 0.5 H 2 0, tetragonal, deliquescent, -0.5 H 20 at 90
66.7 Nd 2 (BrO a)6· 18 H 2 0, red, hexagonal, -18 H 2 0 at 150
70 Se 212, steel gray crystals, melt decomposes at 100
70 AsI 5 , brown, monoclinic
70 K4Fe(ON)6 . 3 H 2 0, yellow, monoclinic, -3 H 20 at 70
70 H 2Mo0 4 · H 20, yellow, monoclinic, -H 20 at 70, -2 H 20
at 200
70 LiOHaOOO . 2 H 20, orthorhombic, melt decomposes
70 MgS0 4 • 7 H 20, orthorhombic, melts with decomposition
70 ZnS0 4 · 6 H 20, monoclinic, -5 H 20 at 70
70.7 N 2H 4 · HNOa, crystals, sublimate at 140
70 to 80 NaK· tartrate· 4 H 20, orthorhombic, -4 H 20 at 215
71.9 Sr(HOOO)2, orthorhombic
72 Au(NOa)a· HNO a . 3 H 20, yellow, triclinic, cubic, melt decom-
poses
73 N0 2HSO a, orthorhombic, melt decomposes
73 Al(N0 3 h . 9 H 20, orthorhombic, deliquescent, decomposes at
134
73 LiI· 3 H 20, monoclinic, -3 H 2 0 at 300
73.4 Na aP0 4 · 12 H 2 0, trigonal, -11 H 2 0 at 100
73.4 SbCl a, orthorhombic, deliquescent, b. pt. 220.2
74 HaPO a, decomposes at 200
74 Y(Br0 3 )3· 9 H 20, hexagonal prisms, -6 H 20 at 100
75 Sm 2 (Br0 3 )6· 18 H 20, yellow, hexagonal, -14 H 20 at 100,
-18 H 2 0 at 150
75 Na 2B 4 0 7 · 10 H 2 0, monoclinic, -10 H 2 0 at 200
75 SiS01 2 , prisms, b. pt. 185
75 Pb(OH 3000)2· 3 H 2 0, monoclinic, -3 H 20 at 75
75.5 GaOI 3 , deliquescent needles, b. pt. 215
75.5 NbF5 , monoclinic, b. pt. 229
77.9 Ba(OH)2· 8 H 20, monoclinic, -8 H 20 at 550
442 Appendix

78 H 7P(Mo z0 7 )6· 28 H 20, yellow, cubic, -25 H 20 at 140

78 Dy(BrOa)a· 9 H 20, yellow, hexagonal, -6 H 20 at 110
78 CaS 20 6 • 4 H 20, trigonal, -4 H 20 at 78, decomposes at 110
79 Sbls , dark brown, b. pt. 400.6
79 Hg(NO a)2· 0.5 H 20, melt decomposes
79 NaNH,HPO,· 4 H 20, monoclinic, melt decomposes
80 Sel" dark gray crystals, -4 I at 100
80 Ni(CIO a)2· 6 H 20, green crystals, melt decomposes
80 Mg(CHsCOO)2· 4 H 20, monoclinic prisms
80 Na,Ca(SO,)s· 2 H 20, -H 20 at 80
80 Cd(CIO a)2· 2 H 20, deliquescent
80 In(CIO,)s· 8 H 20, deliquescent, melt decomposes at 200
84.6Na 2Cr 20 7 • 2 H 20, orange, monoclinic, -2 H 20 at 84.6,
anhydrous salt melts at 356, decomposes at 400
85 N 2H,· 0.5 H 2SO" deliquescent plates
86 CoCla · 6 H 20, red, monoclinic, -6 H 20 at 110
86.3 NaaAsO, . 12 H 20, hexagonal
86.5 Al 2(SO,)s . 18 H 20, monoclinic, melt decomposes
88 SrBr 2 · 6 H 20, -6 H 20 above 180
88.5 Mg( stearate) 2
89 KCr(SO')2· 12 H 20, violet or green, cubic
89.3 Tb(NOa)s· 6 H 20, monoclinic
91 TlAl(SO')2· 12 H 20, cubic
91 Gd(NOa)s· 6.5 H 20, triclinic
92 KAl(SO')2· 12 H 20, cubic, 9 H 20 may be lost at 64.5
92.6 N 2H,· HCl, crystals
93 SiSBr 2 , plates
93.5 NH,Al(SO')2· 12 H 20, cubic, -10 H 20 at 120, -12 H 20
at 200
94 LiHCOO· H 20, orthorhombic, -H 20 at 94
95 LiCIOs · 3 H 20, hexagonal, -2 H 20 at 100, -3 H 20 at 150
95 Si 2Br 6, orthorhombic, b. pt. 240
95 Mg(NOs)2· 6 H 20, monoclinic, deliquescent, -5 H 20 at 330
96.6 SbBrs, orthorhombic, b. pt. 280
96.8 CoSO,· 7 H 20, red, monoclinic, -7 H 20 at 420
96.8 TaFs, tetragonal, b. pt. 230
97.5 Na, silvery, b. pt. 880
97.5 AlBra, trigonal, b. pt. 268
98 MoOF" pale blue, b. pt. 180
98 to 100 NiSO,· 7 H 20, green, orthorhombic, -6 H 20 at 103
99 RbAl(SO')2· 12 H 20, cubic
below 100 H 2PtBr 6 • 9 H 20, red, monoclinic, deliquescent, melt decom-
poses
List of Common Inorganic Compounds in the Order of Their Melting Points 443

below 100 PBrs, yellow, orthorhombic, decomposes at 106

below 100 LiNO a · H 20, needles, melt decomposes
100 Pt0 2 · H 20, black, -H 20 at 100
100 Co(OH)s, black, -1.5 H 20 at 100
100 CuCrzo, . 2 H 20, black, triclinic, -2 H 20 at 100
100 Zn(MnO,}z· 6 H 20, dark brown needles, -5 H 20 at 100,
decomposes at higher temperature
100 Pt0 2 · 2 H 20, brown, -2 H 20 at 100
100 Na 2PtCI,' 4 H 20, red, melt decomposes
100 PtCI,' 8 H 20, red, monoclinic, -4 H 20 at 100
100 Na 2PtC1 6 · 6 H 20, red, triclinic, -6 H 20 at 100
100 CoBr2 ' 6 H 20, red, deliquescent, -4 H 20 at 100, -6 H 20
at 130
100 Cr(NOa)a' 7.5 H 20, purple, monoclinic, melt decomposes
100 K~i(CN), . H 20, red to yellow, monoclinic, -H 20 at 100
100 CoaH 6 (AsOa),' H 20, rose, -H 20 at 100
100 FeFa · 4.5 H 20, yellow crystals, -3 H 20 at 100
100 Pt(OH)2' 2 H 20, yellow, -2 H 20 at 100
100 BaPt(CN)4' 4 H 20, yellow, monoclinic, -2 H 20 at 100,
-4 H 20 at 150
100 H 2W0 4, yellow, orthorhombic, -0.5 H 20 at 100, melts at 1473
100 1Bra' 4 H 20, olive green crystals, -3 H 20 at 100
100 NH,Cr(S04)2' 12 H 20, green or violet, cubic, melt decomposes
100 Cr(OH)a' 2 H 20, green, -2 H 20 at 100
100 Pr2(CO a)a' 8 H 20, green plates, -6 H 20 at 100
100 FeF2 , 8 H 20, green, -8 H 20 at 100
100 (NH,)aFe(CzO,)s, light green crystals, -3 H 20 at 100;
decomposes at 165
100 NasFe(C20,h' 5.5 H 20, green, monoclinic, -4 H 20 at 100,
-5.5 H 20 at 200
100 KaFe(CzO,)a' 3 H 20, monoclinic, -3H 20 at 100, decomposes
at 230
100 Cr 2(SO,)s' 15 H 20, violet, -10 H 20 at 100
100 Crpo, . 6 H 20, violet, triclinic, -3.5 H 20 at 100
100 Na 2SO" orthorhombic, tr. to monoclinic at 100, to hexagonal
at 500, melting at 884
100 (NH')2W,Ola' 8 H 20, cubic or (NH')6W,024' orthorhombic,
-H 20 at 100
100 Na 2CO S ' H 20, orthorhombic, -H 20 at 100
100 NaH 2PO,' H 20, orthorhombic, -H 20 at 100, decomposes
at 200
100 KSbO . tartrate· 0.5 H 20, orthorhombic, -H 20 at 100
100 K2TiF 6' H 20, monoclinic, -H 20 at 100
444 Appendix

100 Ca0 2 · 8 H 20, tetragonal, pearly, -H 20 at 100

100 Ca(H 2PO,)2' H 20, triclinic, -H 20 at 100, decomposes at 200
100 BaS 20 S . H 20, orthorhombic, decomposes at 100
100 Y 2(COS)s' 3 H 20,. -H 20 at 100, -3 H 20 at 130
100 ThF 4 • 4 H 20, powder, -H 20 at 100, -2 H 20 at 150
100 Li 2SiF 0 • 2 H 20, monoclinic, -2 H 20 at 100
100 Na 2WO, . 2 H 20, orthorhombic, -2 H 20 at 100
100 CaSO s ' 2 HzO, -2 H 20 at 100, decomposes at 650
100 BaCl z ' 2 H 20, monoclinic, -H 20 at 100
100 BaBr 2 · 2 H 20, monoclinic, -2 H 20 at 100 and decomposition
100 ZnSOa • 2.5 H 20, monoclinic, -2.5 H 20 at 100, decomposes
at 200
100 Zn(H 2PO,)2' 2 H 20, triclinic, melt decomposes
100 Zn s(AsO')2' 8 H 20, monoclinic, -2 H 2 0 at 100
100 Zn(salicylate)2' 3 H 20, needles, -2 H 20 at 100, -3 H 20
at 150
100 CdSeO, . 2 H 20, orthorhombic, decomposes at 100
100 Cd(H zP0 4 )2' 2 H 20, triclinic, decomposes at 100
100 MgHPO,' 7 H 20, monoclinic, -3 H 20 at 100
100 Mg 2P 20 7 • 3 H 20, amorphous, -3 H 20 at 100
100 Mg-tartrate· 5 H 20, monoclinic, -3 H 20 at 100 and decom-
position
100 MgCO s ' 3 H 20, orthorhombic, -H 20 at 100
100 Ca(lactate)2' 5 H 20, -3 H 20 at 100
100 Sr(CNS)2' 3 H 20, deliquescent, -3 H 20 at 100, decomposes
at 160
100 Y(NOs)s' 6 H 20, triclinic, -3 HzO at 100
100 Ce(NOsh' 6 H 20, deliquescent, -3 H 20 at 100, decomposes
at 200
100 1n2 (SO,)s' 8 H 20, crystals, -3 H 20 at 100, -8 H 20 at 260
100 SrS 20 S ' 5 H 20, monoclinic, -4 H 20 at 100
100 Yb(CHsCOOh' 4 H 20, hexagonal plates, -4 H 20 at 100
100 ZnF 2 • 4 H 20, orthorhombic, -4 H 20 at 100
100 1n(NOs)s . 4.5 H 20, deliquescent needles, -4.5 H 20 at 100
100 MgHAsO,· 7 H 20, monoclinic, - 5H 20 at 100
100 Na 2H 2P 20 o ' 6 H 20, monoclinic, -6 H 20 at 100, the anhydrous
salt melts at 250
100 KNaCO s ' 6 H 20, monoclinic, -6 H 20 at 100
100 Zn(BrOS)2' 6 H 20, -6 H 20 at 200
100 TI 2(SO,)s' 7 H 20, -6 H 20 at 100
100 Sr0 2 · 8 H 20, -8H 20 at 100
100 Ba0 2 · 8 H 20, pearly scales, -H 20 at 100
100 LisPO, . 12 H 20, trigonal
List of Common Inorganic Compounds in the Order of Their Melting Points 445

100 Cr 2(S04h' 18 H 20, violet, cubic, -12 H 20 at 100

above K 20s0 4 · 2 H 20, violet, cubic, -H 20 above 100
100
above P 20 4, orthorhombic, deliquescent, b. pt. 180
100
above LiH 2P0 4
100
above Ca(CI0 3)2' 2 H 20, monoclinic, deliquescent, -H 20 above 100
100
above Sr(N0 3)2' H 20, -H 20 above 100, decomposes at 240
100
102 P 40 6 S 4, tetragonal, deliquescent, b. pt. 295
103 CuS, blue, hexagonal, monoclinic, tr. at 103, decomposes at 220
104 N 2H 4 · 2 HN0 3, needles, melt decomposes
105 Zn3(P0 4)2' 4 H 20, orthorhombic, triclinic, tr. at 105, 140,
and 163
105 Sm 2(S04)3' 8 H 20, monoclinic, -3 H 20 at 105, -8 H:P at 450
108 CdS0 4 • H 20, monoclinic, tr. at 108
110 P 214' orange, triclinic
110 WOF 4, plates, b. pt. 187
110 TI· CH 3COO, silky needles
110 Li 2Cr 20 7 • 2 H 20, deliquescent, orange, -2 H 20 at 110
110 U0 2(CH3COO)2' 2 H 20, yellow, orthorhombic, -2 H 20 at 110
110 U0 2 (CI0 4 )2' 4 H 20, yellow, melt decomposes
110 CuCI 2 • 2 H 20, green, orthorhombic, -H 20 at 110 and
decomposition
110 CuS0 4 • 5 H 20, blue, triclinic, -4 H 20 at 1l0, -5 H 20 at 250
110 K 2HAs0 4 • H 20, trigonal, -H 20 at 110
110 PbH105 • H 20, amorphous, -H 20 at 110
110 K· benzoate· 3 H 20, -3 H 2 0 at 110
110 Mg(benzoate)2' 3 H 20, powder, -3 H 20 at 110
110 YCI3 · 6 H 20, orthorhombic, deliquescent, -5 H 20 at 110
110 Gd 2(C 20 4)a' 10 H 20, monoclinic, -6 H 20 at 110
110 Ce 2(C 20 4 h' 9 H 20, powder, -9 H 20 at 110
112.8 S8' pale yellow, orthorhombic, b. pt. 444.6
113.5 12, blue black, orthorhombic, b. pt. 184.35
114 NH 4 · CHaCOO, hygroscopic, melt decomposes
114 (NPCI 2)a, orthorhombic
114.5 CU(NO a)2 . 3 H 20, blue, deliquescent, - HNO a at 170
114 to 116 NH 4 · HCOO, monoclinic, deliquescent, decomposes at 180
115 TIO·OH, red brown crystals, -H 20 at 115
115 Cu(CH aCOO)2' H 20, dark green, monoclinic, boils at 240 with
decomposition
115 PrCla · 7 H 20, green crystals
115 TIHS0 4 , trimorphous, melt decomposes
115 Ba(N0 2)2' H 20, hexagonal, decomposes at 115
115 Ce 2 (CH 3COO)6 . 3 H 20, needles, -3 H 20 at 115 and decom-
position
446 Appendix

117 CsAl(SO,h' 12 H 20, cubic, melt decomposes

118 MgCl 2 . 6 H 20, monoclinic, deliquescent, melt decomposes
119.0S8' pale yellow, monoclinic, b. pt. 444.6
120 Bi 20j)' H 20, red, -H 20 at 120, -0 2 at 357
120 BaCr 20 7 • 2 H 20, yellow, -H 20 at 120
120 MgCrO,' 7 H 20, yellow, orthorhombic, -3 H 20 at 120
120 S, pale yellow, amorphous, b. pt. 444.6
120 Sr(CIOs)2' orthorhombic, melt decomposes
120 Sr(BrOsh' H 20, monoclinic, -H 20 at 120, decomposes at 240
120 Ba(CIOs)2' H 20, monoclinic, -H 20 at 120
120 Ca(salicylate)2' 2 H 20, cubic, -H 20 at 120
120 Sr(lactate}2' 3 H 20, -3 H 20 at 120
120 Al2F 6.7 H 20, -4 H 20 at 120, with boiling -6 H 20 at 250
120 MgS (PO')2' 8 H 20, monoclinic prisms, -5 H 20 at 120,
-8 H 20 at 400
120 Na 2MosO lO • 7 H 20, needles, -6 H 20 at 120
120.5 SiI" cubic, b. pt. 290
above 120 (NH,)~g(SO')2' monoclinic
above 120 CdHAsO,' H 20
above 120 Tl,P 2°7' monoclinic
123 NH,H 2PO s, deliquescent, melt decomposes at 150
124 NdCl a • 6 H 20, red, orthorhombic, -6 H 20 at 160
125 Na 2HAsO, . 7 H 20, monoclinic
127 HgI2' red tetragonal, tr. to yellow at 127, melts at 259, b. pt. 354
128 NiH" 2 HCOOH, cubic
128 CaSO,' 2 H 20, -1.5 H 20 at 128, -2 H 20 at 163
129 LiCIO s, melt decomposes at 270
129 KNO s, orthorhombic, tr. at 129, melts at 333, decomposes at 400
130 AsI 2, red, b. pt. 380
130 Li 2CrO, . 2 H 20, yellow, orthorhombic, -2 H 20 at 130
130 VFs' 3 H 20, dark green, orthorhombic, -3 H 20 at 130
130 Li 2SO, . H 20, monoclinic, -H 20 at 130
130 SrHAsO, . H 20, orthorhombic needles, -H 20 at 130,
-1.5 H 2 0 at 360
130 Ba(IO a)2' H 20, monoclinic, -H 20 at 130
130 3 PbO· H 20, cubic, -H 20 at 130
130 Cas(citrate)2' 4 H 20, needles, -2 H 20 at 130, -4 H 20 at 185
130 Cd(CH aCOO)2' 2 H 20, monoclinic, -H 20 at 130
130 H 2TeO,' 2 H 20, cubic, monoclinic, -2 H 20 at 130
130 Gd 2 (SeO,)s' 8 H 20, pearly, monoclinic, -8 H 20 at 130
132 NH,SOsNH 2, melt decomposes at 160
135 Co1 2 • 6 H 20, red, hexagonal, -6 H 20 at 135
140 P s0 81 6 , red crystals, melt decomposes
List of Common Inorganic Compounds in the Order of Their Melting Points 447·

140 Co(CH aCOO)2· 4 H 20, red-violet, monoclinic, -4 H 20 at 140

140 CuCI 2 · 2 CuO . 4 H 20, blue-green, -3 H 20 at 140
140 MnHAs0 4 • H 20, -H 20 at 140
140 K 2SnO a · 3 H 20, trigonal, -3 H 20 at 140
143 Co(C10 4 )2· 6 H 20, red, hexagonal
143.5 SnI4 , red, orthorhombic, cubic, b. pt. 340
144 GeI 4, orange, cubic, deliquescent, b. pt. 375
above 145 K 2S4 , red-brown, melt decomposes at 850
146 AsI 3 , red, hexagonal, b. pt. 403
146.9 NH 4HS0 4 , orthorhombic, boils at 490
147 Mg(CI0 4 )2· 6 H 20, hygroscopic, -6 H 20 at 250
148 KH(CH aCOO)2, deliquescent needles or plates, melt decomposes
at 200
149 Ni(CI0 4 )2· 5 H 20 or 6 H 20, blue-green, hexagonal
149.6 NH 4 CNS, monoclinic, melt decomposes at 170
150 TiI 4 , red, cubic, boils above 360
150 NbBr5 , purple-red, b. pt. 270
150 KaOsCI 6 • 3 H 20, red crystals, - 3 H 20 at 150
150 Mn(H 2P0 2)2· H 20, rose, monoclinic, -H 20 at 150 with
decomposition
150 Na 2PtOa · 3 H 20, yellow, -3 H 20 at 150
150 DY2(Cr0 4 )a· 10 H 20, yellow crystals, -3.5 H 20 at 150
150 AgsAsO a, yellow, melt decomposes
150 SC(NO a)3
150 Re0 4 , globules, melt decomposes
150 SrC 20 4 • H 20, -H 20 at 150
150 BaHAs04 • H 20, orthorhombic, monoclinic, -H 20 at 150,
-1.5 H 20 at 225
150 Na 3 • citrate· 2 H 20, -2 H 20 at 150
150 Na 2 SO a · 7 H 20, monoclinic, -7 H 20 at 150, decomposition
150 Ba3 (citrate)2· 7 H 20, -H 20 at 150
150 2 Na a · citrate· II H 20, orthorhombic, -II H 20 at 150
above 150 In(OHb gelatinous, -H 20 above 150
150 to 155 YbCI 3 • 6 H 20, green, orthorhombic, -6 H 20 at 180
151 NH 20H . HCI, monoclinic, melt decomposes
153 HgICI, red, orthorhombic, b. pt. 315
155 In, soft metal, b. pt. 1450
160 AuBra, dark brown, melt decomposes
160 YCl a · H 20, decomposes at 170 to 180
160 Cd(IO a)2· H 20, tr. at 160
160 Mn(CNS)a· 3 H 20, deliquescent, -3 H 20 at 160
161.4 RbNO a, hexagonal, cubic, triclinic, and orthorhombic; tr. at
161.4, 219, and 310, melts above 700
448 Appendix

163 BiOl 2 (1), black needles, decomposes at 300

163 OaSO" . 0.5 H 20, -0.5 H 20 at 163
164 GaOI 2, deliquescent crystals, b. pt. 535
165 MgBr 2 · 6 H 20, hexagonal, deliquescent, melt decomposes
167 Sbla, orange, orthorhombic, monoclinic, b. pt. 401
167.5 KHOOO, orthorhombic, melt decomposes
169.6 NH 4NO a, tetragonal, orthorhombic, monoclinic, cubic; decom-
poses at 210
170 SbOOI, monoclinic, melt decomposes
170 NH 20H . 0.5 H 2S0 4, monoclinic, melt decomposes
170 Ba(Br0 3 )2· H 20, monoclinic, -H 20 at 170, decomposes at 260
170 (NH4)2Pr2(S04),,· 8 H 20, -8 H 20 at 170
170.5 LiHS0 4, prisms
172.3 KONS, monoclinic, deliquescent, decomposes at 500
172.5 P"Sa, yellow, orthorhombic, b. pt. 408
174 BeOH(stearate), powder
175 TeOI 2, black crystals, b. pt. 324
175 Ag 2S, black, cubic, tr. at 175
175 Hg(OH)2' -H 20 at 175
177 FeI 2, gray, hexagonal
180 Li 2PtCI s · 6 H 20, orange-red, hexagonal, -6 H 20 at 180
180 OsBra, orthorhombic
180 3 K 2S 20 S • H 20, monoclinic, deliquescent, -H 20 at 180 and
decomposition
180 K"P 20 7 • 3 H 20, deliquescent, -2 H 20 at 180, -3 H 20 at 300
185 H 3 BO a, triclinic, melt decomposes
185 NaAlOl4.
186 Li, silvery metal, b. pt. 1336
190 SeBrOla, yellow-brown crystals
190 RbI 3 , orthorhombic
190 PBr 2N, orthorhombic
190 BeS0 4 • 4 H 20, tetragonal, -4 H 20 at 190, decomposes at 550
191 AlIs , brown crystals, b. pt. 382
194 MoOls , black crystals, b. pt. 268
194 NbOla, yellow needles, deliquescent, b. pt. 240.5
195 NH 40I· MgOI 2 • 6 H 20, orthorhombic, deliquescent, -4 H 20
at 195
195.5 OsBrI 2
197 Or0 3 , red, orthorhombic, melt decomposes
198 N 2H,,· 2 HOI, cubic
198 NH" . benzoate, orthorhombic
199 Inls, yellow deliquescent crystals
200 K 2Ru0 4 • H 20, black, orthorhombic, -H 20 at 200
List of Common Inorganic Compounds in the Order of Their Melting Points 449

200 MnHPO a · 3 H 20, pink, -H 20 at 200

200 MnHP0 4 • 3 H 20, pink, orthorhombic, -3 H 20 at 200 and
decomposition
200 CaCr0 4 • 2 H 20, yellow, monoclinic, -H 20 at 200
200 DyP0 4 • 5 H 20, yellow, -5 H 20 at 200
200 DY2(8e0 4 )a· 8 H 20, yellow needles, -8 H 20 at 200
200 NiBr 2 · 3 H 20, green, deliquescent, -3 H 20 at 200
200 Ni(CN)2· 4 H 20, green plates, -4 H 20 at 200 and decomposi-
tion
200 AuOH, violet, -H 20 at 200
200 NH 4H 2P0 2, orthorhombic, decomposes at 240
200 NaH 2P0 2 · H 20, monoclinic, -H 20 at 200
200 Li 2B 4 0 7 • 5 H 20, -2 H 20 at 200
200 Zn(I0 3)2· 2 H 20, -2 H 20 at 200
200 Rb 28 . 4 H 20, deliquescent crystals, -4 H 20 at 200
200 K 2Te0 4 • 5 H 20, orthorhombic, deliquescent, -H 20 and -0
at 200
200 Mg(Br0 3)2· 6 H 20, cubic, -6 H 20 at 200
200 Mg80 3 · 6 H 20, trigonal, -6 H 20 at 200
above 200 AgIOa, orthorhombic, melt decomposes
205 NH 280 3H, orthorhombic, melt decomposes
206 K 28 s , orange
206 TIN0 3, cubic, orthorhombic, trigonal, transitions at 75 and 145;
boils at 430
207.5 CsI3, orthorhombic
210 NaNH 2, olive green, b. pt. 400
210 KH80 4 , orthorhombic, monoclinic, melt decomposes
210 Cs 28s
210 Mg(I0 3)2· 4 H 20, monoclinic, -4 H 20 at 210
210 to 215 8nF2' monoclinic
211 WOCI 4 , red needles, b. pt. 227.5
212 AgN0 3, orthorhombic, decomposes at 444
215.5 8nBr 2 , yellow, orthorhombic, b. pt. 620
217 Cs 28 3 , orange, boils above 800
217 Ba(N0 2)2
217.4 8es, steel gray, b. pt. 688
218 Ag 2C0 3, yellow powder, melt decomposes
218 BiBr3' yellow, b. pt. 453
220 8es , gray, trigonal, b. pt. 684.8
220 Re 20 7 , yellow crystals, sublimation at 450
221 TaCIs, yellow prisms,· b. pt. 242
224 TeCI 4 , yellow, deliquescent, b. pt. 414
225 Rb 28s, red, orthorhombic, deliquescent
Benedetti·Pichler, Identification 29
450 Appendix

229 HgIBr, yellow, orthorhombic, b. pt. 360

230 AgCIO a, tetragonal, decomposes at 270
230 BiCla, b. pt. 447
Sn, transition from gray cubic at 18 to silvery tetragonal,
231.85
b. pt. 2260
236 LiCI0 4 , deliquescent, decomposes at 410
237 HgBr2' orthorhombic, b. pt. 322
237 Zn(CHaCOO)2· 2 H 20, monoclinic
240 TaBrs , yellow crystals, b. pt. 320
240 CU(IO a)2· H 20, blue, triclinic, -H 20 at 240, decomposes at 290
242 Zn(CH aCOO)2' monoclinic, sublimes in vacuum
244.7 NH 4 SO aF
246.8 SnCI 2 , orthorhombic, b. pt. 623
248 WCIs , dark green, deliquescent, b. pt. 275.6
248 NaCIO a, cubic, trigonal, melt decomposes
250 Au(OH)a, brown-black, -H 20 at 250
250 Co a(P0 4)2· 8 H 20, red powder, -8 H 20 at 250
250 Na 2H 2P 20 s
250 AgN a, prisms, explodes at 297
250 Si 2l s , hexagonal plates, melt decomposes
250 to 300 H 2U0 4 , yellow powder, -H 20 at 250 to 300
251.8 N a 2SS' yellow
252 K 2Sa, yellow-brown
253 NaHCOO, monoclinic
254 N 2H 4 • H 2S0 4 , orthorhombic
254 AuCI a, red, deliquescent, melt decomposes, sublimation at 265
256 KH 2P0 4, tetragonal, deliquescent, melt decomposes
256 Cd(CH aCOO)2, melt decomposes
259 Te1 4 , gray crystals
259 Hg12' yellow, orthorhombic, b. pt. 354
260 CuCr0 4 • 2 CuO· 2 H 20, yellow-brown, -2 H 20 at 260
261 LiNO a, trigonal
265 KMgCl a • 6 H 20, orthorhombic, deliquescent
266 W0 2C1 2, yellow tablets
267 As 2S2, red, transition at 267, melting at 307, b. pt. 565
271 Bi, b. pt. 1450
271 NaN0 2 , pale yellow, orthorhombic, decomposes at 320
272 CsOH, deliquescent
272 TI 2CO a, monoclinic
275 Na 2 S 4 , yellow, cubic
275 WCI s, dark blue, cubic, b. pt. 346.7
276 WBrs, violet brown needles, b. pt. 333
276 P 2SS' yellow, deliquescent, b. pt. 530
List of Common Inorganic Compounds in the Order of Their Melting Points 451

277 WOBr4 , black, deliquescent, b. pt. 327

277 HgCI 2, orthorhombic, b. pt. 304
280 CO(CN)2· 2 HzO, buff, -2 HzO at 280, decomposes at 300
280 Rb z0 4 , yellow, -0 at 500
280 Ba(NHz)z, gray-white
280 Pb(CHaCOO)z
above 280 K 20 4, orange-yellow
282 FeCl s, brown-black, hexagonal, deliquescent, b. pt. 315
283 ZnCl z, deliquescent, b. pt. 732
287 NaCNS, orthorhombic, deliquescent
288 KHzAsO 4' tetragonal
290 P 4S 6 , grayish yellow, b. pt. 490
292 KCHsCOO
292 SbFa, orthorhombic, sublimes
297 KN0 2, prisms, decomposes at 350
298 P aS 6 , yellow needles, boils at 4001
300 MnP0 4 · H 20, gray crystals, -H 20 at 300 and decomposition
300 AszS a, red or yellow, monoclinic, b. pt. 707
300 TI 20, yellow, deliquescent, -0 at 1865
300 U(S04)2· 4 H 20, green, orthorhombic, -4 H 20 at 300
300 Na 2(NO)2' melt decomposes
300 NaRe0 4 , plates
300 K 2 S20 7
300 RbOH, deliquescent
300 Al(OH)a, monoclinic, -H 20 at 300
300 Hg 2NCI· 3 NH 4 CI, reddish crystals
300 Na 6W 7024. 16 H 20, triclinic, -16 HzO at 300
above 300 NaPO s, glass, crystallizes above 300
303.5 TI, bluish white metal, b. pt. 1650
307 As 2SS ' red, monoclinic, b. pt. 565
308 NaNO a, trigonal, decomposes at 380
308 Ce(CHaCOO)a, melt decomposes
310 BzSa, white crystals
311.1 KSOaF
315 As 20 S ' vitreous
above 315 NaHS0 4 , triclinic, melt decomposes
318.4 NaOH, deliquescent, b. pt. 1390
320 Snl z, red, orthorhombic, monoclinic, b. pt. 720
320 AgCN, -(CN)z at 320
320.9 Cd, b. pt. 767
323 Mg(CHaCOO)z
324 NaCHaCOO, monoclinic
327.3 Pb, b. pt. 1620
29*
452 Appendix

333 KNO a, trigonal, decomposes at 400

334 TINa, pale yellow, tetragonal, explodes at 430
338 KNH 2, yellowish green, sublimes at 400
340 TI 2Se, dark gray leaflets
340 Se0 2 , tetragonal, sublimes at 317
above 340 TI(OH)a, brown, hexagonal
350 IrO z ' 2 HzO, dark blue, -2 HzO at 350
350 Cd(NOa)z
350 KReO" tetragonal
356 Na 2Cr20?, orange, decomposes at 400
357 !(AuCI" yellow, monoclinic
360 RUZ05' black crystals, -0.5 0 at 360
360 As 2 Se S ' dark brown
360 CuOH, yellow, -0.5 H 20 at 360
360 DY2(S04)S' 8 H 20, yellow crystals, -8 H 20 at 360
360 LiS03 F
368 KCIO a, monoclinic, decomposes at 400
370 KBrO s, trigonal, melt decomposes
373 PbBr2, orthorhombic, b. pt. 918
375 Sr(OH)2' deliquescent
375 Eu 2(SO,)a' 8 H 20, pale red, -8 H 20 at 375
380 KOH, orthorhombic, deliquescent, b. pt. 1320
380 TeBr4, orange prisms, b. pt. 421
380 Zr(SO')2' 4 HzO, orthorhombic, -4 H 20 at 380
381 NaBrOa, cubic
383 Ag,V 20?
384 NaBF" orthorhombic, melt decomposes slowly
385 CdI 2, brown, hexagonal, b. pt. 713
388 K 2W0 4 · 2 H 20, monoclinic, transitions at 388 and 555,
melting at 921
390 LiNH2' cubic, b. pt. 430
390 B 2Sa, white crystals
394 ZnBr2' orthorhombic, b. pt. 650
398 K 2Cr2 0?, orange, triclinic, decomposes at about 500
400 CS 20 3 , chocolate brown
400 Ba(CI0 4)2' 3 H 20, hexagonal, decomposes at 400
400 Th(S04)z' 9 H 20, monoclinic prisms, -9 H 20 at 400
400 to 500 T1 2C13 , yellow, hexagonal, melt decomposes
above 400 K 2PtBr 6' red, cubic, decomposes above 400
above 400 T1 2Se04, orthorhombic
402 PbI z, yellow, hexagonal, b. pt. 954
402 SrI 2, plates, melt decomposes
408 Ba(OH)2' monoclinic
List of Common Inorganic Compounds in the Order of Their Melting Points 453

412 Tl 2Te
414 CsNOs, hexagonal, melt decomposes
414 Ba(CIOs)2
419.5 Zn, silvery, b. pt. 907
420 CsBrOs
above 420 Zn SP 2 ' steel gray, cubic, b. pt. 1100
422 Cu 2C1 2, cubic, b. pt. 1366
424 TIVO s, dark gray crystals
430 Pt0 2, black
430 RbBrO s
430 TICI, cubic, b. pt. 806
434 AgBr, pale yellow, cubic, decomposes at 700
435 AgF, yellow, cubic, deliquescent
439 Bil s, red to black, hexagonal, boils at 500 with decomposition
440 TlI, yellow orthorhombic or red cubic, b. pt. 824
440 RU409, black crystals, -0 at 440
440 BeCI 2, deliquescent needles, b. pt. 547
445 LiOH, crystals, boils at about 925
446 Lil, cubic, deliquescent, b. pt. 1190
446 ZnI2' cubic, b. pt. 624
448 Tl 2S, bluish black, tetragonal, melt decomposes
450 FeS 2, yellow, transition from orthorhombic to cubic at 450
450 Y2(S04k 8H 20, monoclinic, -8H 20 at 450, decomposes at 700
452 Te, metallic, hexagonal, b. pt. 1390
454 Tl 4V 20 7
455 AgCI, cubic, b. pt. 1550
455 LiSO a, powder, melt decomposes
455 KHS, orthorhombic, deliquescent
460 Ce 2S a, dark red, amorphous, deliquescent, decomposes above 800
460 TlBr, pale yellow, cubic, b. pt. 815
470 Rb 20 a, black
471 K 2S, brown, deliquescent
471 K 2 S 2, orange
474.5 Cu 2 (CN)2' monoclinic, melt decomposes
482 NaC10 4, orthorhombic, melt decomposes
482 AgPO a
490 BeBr 2, deliquescent needles, sublimes at 450
498 CuBr 2, black, monoclinic, deliquescent
498 CuCI 2, brownish yellow powder, becomes Cu 2Cl 2 at 993
500 PdCI 2 , brown, cubic, melt decomposes
500 Fe(OH)3' brown, -1.5 H 20 at 500
500 Na 2 S04, monoclinic, transition to hexagonal at 500, melting
at 884
454 Appendix

501 PbCI 2, orthorhombic, b. pt. 954

501 TICI0 4, orthorhombic, melt decomposes
504 Cu 2Br 2, cubic, b. pt. 1345
505 Ba(CI0 4)2' hexagonal
510 BeI 2, needles, b. pt. 590
513 (NH4)2S04, orthorhombic, melt decomposes
520 CdF 2' cubic, boils above 1200
524 PbCI 2 · PbO, tetragonal, melt decomposes
529.5 KBF 4, cubic, orthorhombic, melt decomposes
547 LiBr, cubic, deliquescent, b. pt. 1265
550 Sb 2S a, black, orthorhombic
550 Zr(OH)4, gelatinous, -2 H 20 at 550
555 PtO, violet-black
560 KIO a, monoclinic
561 Ca(NO a)2' cubic
563.7 NaCN, cubic, b. pt. 1496
566 ThI 4, prisms, b. pt. 837
568 CdCI 2, cubic, b. pt. 960
570 HgF, yellow, cubic?
570 Sr(NOa)2, cubic
573 Bi2Te a
575 CaI 2, deliquescent plates, b. pt. 718
577 B 20 a, vitreous, boiling above 1500
580 CdBr 2, plates, b. pt. 963
580 Cd(OH)2' hexagonal, -H 20 at 580
582 KI0 4, tetragonal, -0 at 300
585 Ag 2P 2 0 7
588 TbCla, red
592 Ba(NO a)2' cubic, melt decomposes
593 P 4, violet, monoclinic
600 Rb 20 2, yellow, cubic
600 CS 20 4, yellow crystals, melt decomposes
605 Cu 2I 2 , reddish brown, cubic, b. pt. 1290
611 Sb 2 S 3, gray
612 Na 2Mo 20 7 , needles
614 LiCI, cubic, deliquescent, b. pt. 1360
618 Li 2C0 3, monoclinic, melt decomposes
621 CsI, cubic, b. pt. 1280
627.6 NaP0 3, transition at 500 from the insoluble to the soluble
modification
628 GdCI 3, monoclinic
629 Sb 2Te 3, gray
630 Ce 2(S04)3' 8 H 20, triclinic, -8 H 20 at 630
List of Common Inorganic Compounds in the Order of Their Melting Points 455

630.5 Sb, b. pt. 1380

632 TI 2S0 4 , orthorhombic, melt decomposes
634.5 KCN, cubic, deliquescent
636 CsBr, cubic, b. pt. 1300
642 RbI, cubic, b. pt. 1300
643 SrBr 2, needles, melt decomposes
645 Ce, steel gray, b. pt. 1400
645 Ni 2S, yellow crystals
645 HgF 2' cubic, melt decomposes
646 CsCI, cubic, deliquescent, b. pt. 1290
650 MnCI 2, rose, cubic, deliquescent, b. pt. 1190
650 SnSe 2, white or brown
651 Mg, b. pt. 1110
651 NaI, cubic, b. pt. 1300
652 Sb 20 a, cubic
652 Ag 2S0 4 , orthorhombic, melt decomposes
654 N a 4 V 207' hexagonal
655 DyCl a, yellow plates
656 Sb 20 a, orthorhombic, b. pt. 1570
660 AI, b. pt. 2057
680 LiH, cubic
682 RbBr, cubic, b. pt. 1340
683 CsF, cubic, b. pt. 1250
685 Bi 2S a, brown, orthorhombic, melt decomposes
below 686 YCl a, plates
686 SmCla, greenish yellow crystals
687 CuaSb, light gray
687 Na 2Mo0 4
692 Na 2W0 4 , orthorhombic
693 PbCI 2 · 2 PbO, colorless or yellow, orthorhombic
700 MnS0 4 , pink, decomposes at 850
700 MgBr2' deliquescent
704 Bi 20 a, yellow, cubic, orthorhombic or tetragonal, transition
at 704
710 BiSe a, black, orthorhombic, melt decomposes
712 MgCI 2, hexagonal, b. pt. 1412
715 RbCI, cubic, b. pt. 1390
715 MgaBi2' metallic
721 CdaAs 2, dark gray, cubic
723 KI, cubic, b. pt. 1330
728 RaBr 2 , monoclinic, sublimes at 900
730 KBr, cubic, b. pt. 1380
730 K 2GeF 6' hexagonal, b. pt. 835
456 Appendix

740 BaI a · 2 HaO, monoclinic, water lost at 539, decomposition

at 740
741 Na aB,07
755 NaBr, cubic, b. pt. 1390
759 TlaO a, brown-black, hexagonal, -20 at 875
760 RbF, cubic, b. pt. 14lO
760 CaBra, deliquescent needles, b. pt. 810
766 PbSiO a, monoclinic
772 CaCl a, cubic, deliquescent, boils above 1600
775 NaaBi, bluish violet
780 SnTe, gray crystals
784 NdCla, violet prisms
790 KCI, cubic, b. pt. 1500
792 NaaCrO" yellow, orthorhombic
795 MoOa, orthorhombic, sublimes
797 NiS, black trigonal, yellow hexagonal
800 Sr, silvery, b. pt. 1156
800 PdS a , gray, melt decomposes
800 BeF a' vitreous, sublimes above 800
800 V a0 5 , orange, orthorhombic, decomposes at 1750
above 800 VFs, green, orthorhombic, sublimes
800.4 NaCI, cubic, b. pt. 1413
802 Pb~a07' orthorhombic
807 KPO a
810 Ca, silvery, boils at about 1200
812 KaSb, yellowish green
815 PbaP a07' orthorhombic, melt decomposes
820 ThCI" orthorhombic, deliquescent, b. pt. 921
822 CeCla, deliquescent
823 PrCls, green needles
825 AgaS, black, orthorhombic, melt decomposes above 830
826 La, gray, b. pt. 1800
830 CuaAs, hexagonal
837 LisPO" orthorhombic
837 RbaCO s, deliquescent, decomposes at 740
840 Nd, yellowish metal
844 PbCrO" yellow, monoclinic, melt decomposes
845 LiaN, black
849 AgaPO" yellow, cubic
851 NaaCOa, melt decomposes
855 PbF a, orthorhombic, b. pt. 1290
856 MnF 2' red prisms
856 NaaSb, deep blue
 Inorganic Substances that Sublime, Arranged According to Color 457

861 SnSe, steel gray prisms

866 NaaVO" needles
872 ZnF 2' monoclinic, triclinic (1)
873 Sr0l2, cubic
above 875 Na 2Se, deliquescent crystals
880 SnS, brownish black, orthorhombic, b. pt. 1150
880 Ag2Se, gray, cubic
880 KF, cubic, b. pt. 1500
884 Na 2SO" hexagonal
950 Na 2S, amorphous
966 NaBO z, hexagonal prisms, b. pt. 1400
992 NaF, tetragonal, b. pt. 1704

Table 5
Inorganic Substances that Sublime, Arranged According to Color
Approximate
Temperature
°C
56.2 UF 6, black needles, volatilizes
Nil 2 , black, deliquescent, sublimes
446 HgS, black amorphous, sublimes
HgSe, gray cubic plates, sublimes
MOaOs0l5' dark brown, deliquescent, sublimes
MoCI" brown, deliquescent, volatilizes
AsP, red, sublimes and decomposes
HgI2' red tetragonal, transition at 127 to yellow orthorhombic,
m. pt. 259, b. pt. 354, yellow sublimate turning red on cooling
(and scratching)
580 HgS, red, hexagonal, sublimes
FeBra, dark red, deliquescent, sublimes and decomposes
250 Ni-dimethylglyoxime, red needles, sublimes
CoOla, red crystals, sublimes
135 N,S" orange-red, monoclinic, sublimes, melts at 178
135 Sb 2 Ss, golden, decomposition and "sublimate" of sulfur
(NH,)aCS3' yellow, sublimes
615 As" yellow cubic, transition at 358 to black amorphous or
metallic hexagonal, sublimes
HgI2' yellow orthorhombic, see above
140 HgI, yellow tetragonal, melts at 290, decomposes at 310
NbOBra, yellow crystals, sublimes
about 900 NiCla, yellow, deliquescent, sublimes
MoOaBr2' yellow, deliquescent, sublimes
458 Appendix

Mo0 2C1 2, pale yellow, sublimes

(NH4)aU02F5, yellow, tetragonal, volatilizes
250 HgIO a, yellowish white, volatilizes
SeCI 4, yellow or white deliquescent crystals, sublimes and
decomposes at 288
below 100 MoOC1 4, green, deliquescent, sublimes and melts below 100
below 100 MoOCl a, green, sublimes
83 CrCl 3 • 6 H 20, violet or green hexagonal plates, sublimes
high temp. CoCI 2, blue crystals, sublimes
110 HI0 4, colorless crystals, sublimes
450 Te0 2, tetragonal, orthorhombic, sublimes
NH 4HCO a, NH 2 • CO 2 . NH 4, sublimes
520 NH 4CI, cubic, sublimes
542 NH 4Br, cubic, sublimes
551 NH41, cubic, sublimes
120 NH 4HS, orthorhombic, sublimes
230 NH4BF 4' orthorhombic, sublimes
160 PCl5, tetragonal, deliquescent, sublimes
PH 4Cl, cubic, sublimes
250 P 205' amorphous, deliquescent, sublimes
HPO a, sublimes
As 20 a, cubic, monoclinic, sublimes
Na 20, sublimes
178 AICl a, hexagonal, deliquescent, sublimes
400 NbOC1 3 , needles, sublimes
ZrF 4, hexagonal, sublimes
610 ThBr 4 , sublimes
400 Hg 2C1 2, tetragonal, sublimes
345 HgBr, tetragonal, sublimes
440 InCI 3 , plates, sublimes

Table 6
Inorganic Solids which Burst into Flame when Heated in Air
Ignition Temperatures in Centigrades
Metals of the alkali and alkaline earths groups, aluminum foil
The elements sulfur, carbon, and boron
Phosphorus, P 4' yellow hexagonal ignites at 34; black rhombohedral
at 200; red cubic at 725
(P 4H2b yellow at 200
P 2Se5' red needles
BP, maroon powder, at 200
KH 2P0 2, hexagonal, deliquescent
 List of Solids which Explode on Heating 459

ThS 2, yellow, brown or black

Th0 2, yellow crystals
V 202' light gray crystals
W0 2, brown, cubic
WP, gray prisms
U 2Sa, gray to black
K 2Pt 4 S 6 , blue to gray

Table 7
List of Solids which Explode on Heatingl
Detonation temperatures in centigrades
Inorganic Compounds
Peroxides:
Oa0 2 · 8 H 20, pearly, tetragonal, at 275
Zn0 2, yellow, at 212
Ammonium Salts:
NH 4 0IO a, monoclinic, at 100
NH 4I0 4 , tetragonal
NH 4N0 2, needles
NH 4Mn0 4 , orthorhombic, at 60
Nitrides:
Nl a, black
NHaN1a, red, orthorhombic
Se 2N 2, orange or yellow, at 200
HgaN 2' brown powder
Ge 4N 4 , orange or yellow, at 200
Azides:
BaN 6' orthorhombic, above 200
BaN 6 • H 20
AgN a, prisms, at 297
TINa, pale yellow, tetragonal, at 430
HgNa, at 245
PbN 6' needles, at 350
Ohlorites and Ohlorates:
Pb(CI0 2)2' yellow, monoclinic, at 126
HgOIO a, orthorhombic, at 250
Nitrates:
HgNO a · H 20, monoclinic, below 70
1 Like all others, this table is not complete. Thus failure to find a sub-
stance in this table does not prove that it is harmless.
460 Appendix

Various Compounds of Ammonia with Silver, Gold, and the Platinum Metals:
AgNH 2
Au . NH . Cl + Au . NH . NH 2 , yellow
Oxalates:
Ag 2C20 4 , at 140
HgC 20 4 , below 160
Cyanides and Fulminates:
Hg(CN)2 . HgO
Ag 2(NCO)2' needles
Hg(NCO)2' cubic
Acetylides:
Cu 2CZ • HzO, reddish brown
AgzC z
Various Instable Compounds:
(SiOOH)2' white, amorphous
SH z, explodes in air

Organic Compounds
Iodocompounds:
Iodosobenzene, yellow, amorphous, at 210
Iodoxybenzene, needles, at 236 to 237
Peroxides:
Benzoylperoxide, orthorhombic, m. pt. 108, melt explodes
Chlorates, Perchlorates, Nitrates of Organic Bases.
Nitric Esters:
Glycerol dinitrate, oil
Glyceryl trinitrate, oil, at 270
Erythritol tetranitrate, m. pt. 61, melt explodes
Mannitol hexanitrate, needles, m. pt. 112 to 113
Azides and Oximes:
Benzazide, plates, m. pt. 32, melt explodes
Chloramine T, yellowish white powder, at 175 to 180
Benzohydroxamic acid, orthorhombic, m. pt. 132, melt explodes
Phloroglucinol trioxime, powder, at 155
Nitrocompounds:
Nitro uracil, needles
o-Nitrophenylpropiolic acid, needles, at 155
Phenylnitroamine, leafy, m. pt. 46, explodes at 98
Dinitroresorcinol, yellow leaves, m. pt. 148, melt explodes
 Inorganic Solids Moderately Soluble in Water at Room Temperature 461

Nitranilic acid, yellow plates, after loss of 1 H 20 melts at 100, explodes

at 170
3,4-Dinitro-o-xylene, needles, m. pt. 82, explodes at 413
3,5-Dinitro-o-xylene, yellow needles, m. pt. 76, explodes at 438
Trinitromethane, crystals melt at 23 and explode
2, 4, 6-Trinitrotoluene, TNT, crystals melt at 80.1, explode at 280
2,3,4-Trinitrotoluene, crystals melt at 112, melt explodes at 290 to 310
2, 4, 5-Trinitrotoluene, yellow plates melt at 104, melt explodes at 290
2,4,6; 1,3,5-Trinitrotrimethylbenzene, triclinic, m. pt. 232, explodes at 415
Trinitro-p-xylene, monoclinic, m. pt. 140, explodes at 410
2,4,6-Trinitrophenol, yellow, orthorhombic, m. pt. 121.8, explodes
above 300
Trinitro-m-cresol, yellow needles, m. pt. 109.5, explodes at 150
Trinitro orcinol, yellow needles, m. pt. 163, melt explodes
Trinitro-naphthol, yellow needles, m. pt. 190, melt explodes
Trinitro aniline, yellow, monoclinic, m. pt. 188 to 190, melt explodes
Trinitro acetonitrile, waxy, m. pt. 41.5, explodes at 220
Tetranitrophenol, light yellow, m. pt. 140, melt explodes
~-Tetranitronaphthalene, light yellow, m. pt. 259, melt explodes
P-Tetranitronaphthalene, needles, m. pt. 203, melt explodes
Tetranitrodiphenyldisulfide, yellow needles explode above 280
Azocompounds :
Dichloro-azodicarbonamide, yellow needles, at 155
Diazo uracil, red or yellow plates
Diazobenzene chloride, hygroscopic crystals
Diazobenzene nitrate, needles
Diazobenzenesulfonic acid, needles
Diazosalicylic acid, yellow crystals, at 155
Diazo-aminobenzene, yellow leaves, m. pt. 96 to 98, melt explodes

Table 8
Inorganic Solids Moderately Soluble in Water at Room Temperature
LiF, cubic; Na 2 SiF 6 , pink, hexagonal; (Na, Li)· (Mg, Zn)-uranyl acetates,
octahedral, yellow, green fluorescence.
Perchlorates (orthorhombic), permanganates (orthorhombic), fluorosilicates
(cubic, hexagonal), chloroplatinates and chloriridates (cubic), acid
tartrates (orthorhombic), and picrates of K, Rb, Os, NH,; KsTiF 6· H 20,
monoclinic; KBF" cubic, orthorhombic; KIO" tetragonal; KsTeO,·
. 5 H 20, orthorhombic; Os alum, cubic; OsAuOI" yellow, monoclinic;
NH,VO a•
462 Appendix

BeCO a • 4 H 20; Ca(OH)2' hexagonal; Sr(OH)2' 8 H 20, tetragonal; Ba0 2 ·

. 8 H 20, pearly scales; BaF 2' cubic; Ba(BrO a)2' H 20, monoclinic;
Ca(IO a)2, triclinic; CaS0 4 • 2 H 20, monoclinic; BaS 20 a · H 20, ortho-
rhombic; CaW0 4 , tetragonal; SrW0 4 , tetragonal; Ca(BO a)2' 2 H 20,
cubic; BaSiF 6' prismatic; Sr tartrate tetrahydrate, monoclinic.
MoO a, orthorhombic; H 2Mo0 4 • H 20, yellow, monoclinic.
PtBr4, dark brown.
Cu(IO a)2' green, monoclinic; Cu(IO a)2' H 20, blue triclinic; AgN0 2, ortho-
rhombic; AgBrO a, tetragonal; Ag 2 S0 4 , orthorhombic; AgMn04' purple
to black, monoclinic; Ag acetate, needles; Ag tartrate, scales.
HgBr2' orthorhombic; Hg(BrO a)2 . 2 H 2 0, crystals; mercurous formate and
acetate, scales.
AIF a, triclinic; TICI, cubic; T1 2Cl a, yellow, hexagonal; TICNS, tetragonal
needles; Tl aP0 4, needles; T1 4V 20 7 ; TI 4Fe(CN)6' 2 H 20, yellow, triclinic.
Ge0 2 , orthorhombic; GeS, orthorhombic, monoclinic; GeS 2 , white;
PbCI 2, orthorhombic needles; PbBr 2 , orthorhombic.
Sb 20 5 , yellow.
 Literature
General Reference Books
(1) American Chemical Society: Critical Solution Temperatures, Advances in
Chemistry Series No. 31. American Chemical Society Special Issue Sales,
1155 Sixteenth Str., N. W., Washington 6, D. C.
(2) Beilsteins Handbuch der organischen Chemie. Berlin-Gottingen-Heidelberg:
Springer. 1957-1961.
(3) FURMAN, N. H., editor: Scott's Standard Methods of Chemical Analysis,
6th ed. New York: Van Nostrand. 1962.
(4) Gmelin-Institut der Max-Planck-Gesellschaft: Gmelins Handbuch der
anorganischen Chemie, Sth ed. Weinheim: Verlag Chemie. 1930.
(5) Ibid., Magnetic Materials, 2nd supplement to Part D of System No. 59,
Iron. Weinheim: Verlag Chemie. 1959.
(6) HAUSNER, H. H.: Modern Materials, 2 Vols. New York: Academic Press.
1959-1960.
(7) HILLEBRAND, W. F., G. E. F. LUNDELL, H. A. BRIGHT, and J. I. HOFFMAN:
Applied Inorganic Analysis, 2nd ed. New York: John Wiley. 1953.
(S) HODGMAN, C. D., editor: Handbook of Chemistry and Physics. Cleveland,
Ohio: The Chemical Rubber Company.
(9) HORWITZ, W., editor: Official Methods of the Association of Official Agri-
cultural Chemists, 9th ed. Washington: Assoc. of Official Agricultural
Chemists. 1960.
(10) LANGE, N. A., editor: Handbook of Chemistry. Sandusky, Ohio: Hand-
book Publishers.
(11) MELLOR, J. W.: Compre~nsive Treatise on Inorganic and Theoretical
Chemistry. New York: Longmans, Green and Co. 1922-1961.
(12) RYSHKEWITCH, E.: Oxide Ceramics. New York: Academic Press. 1960.

Theory of Chemical Analysis

(13) BENEDETTI-PICHLER, A. A. : Essentials of Quantitative Analysis. New York:
Ronald Press. 1956.
(14) BENEDETTI-PICHLER, A. A.: Theory and Practice of Sampling for Chemical
Analysis, in W. G. BERL: Physical Methods of Chemical Analysis. New
York: Academic Press. 1956.
(15) CHARLOT, G.: Theorie et methode nouvelles d'analyse qualitative, 3rd ed.
Paris: Masson et Cie. 1949.
(16) EMELEUS, H. J., and J. S. ANDERSON: Modern Aspects of Inorganic
Chemistry. New York: Van Nostrand. 1935.
(17) HAGG, G.: Theoretische Grundlagen der analytischen Chemie. Basel: Birk-
hauser. 1950.
(IS) HAMMETT, L. P.: Solutions of Electrolytes, 2nd ed. New York: McGraw-
Hill. 1936.
(19) HILDEBRAND, J. H., and R. L. SCOTT: Regular Solutions. New York:
Prentice-Hall. 1962.
464 IJterature

(20) HOGNESS, T. R, and W. C. JOHNSON: Qualitative Analysis and Chemical

Equilibrium, 4th ed. New York: Holt. 1954.
(21) KOLTHOFF, I. M., and E. B. SANDELL: Textbook of Quantitative Inorganic
Analysis, 3rd ed. New York: Macmillan. 1952.
(22) SMITH, T. B.: Analytical Processes. A Physico-Chemical Interpretation,
2nd ed. London: Edw. Arnold. 1940.
(23) YOE, J. H.: Chemical Principles. New York: John Wiley. 1937.

Reagents
(24) American Chemical Society: Reagent Chemicals, ACS Specifications,
Washington, D. C.: American Chemical Society. 1950.
(25) British Drug Houses: The Reagents for Delicate Analysis Including Spot
Tests, 7th ed. London: British Drug Houses. 1939.
(26) Hopkin & Williams, Ltd.: Organic Reagents for Metals and for Certain
Acid Radicals, 4th ed. London: Hopkin & Williams. Ltd. 1943.
(27) MERCK, E.: Prufung der chemischen Reagenzien auf Reinheit. Darmstadt:
Merck. 1931.
(28) MERCK, E.: Merck Index, 7th ed., Rahway, N. J.: E. Merck. 1960.
(29) NIEUWENBURG, C. J. VAN, W. BOTTGER, F. FEIGL, A. S. KOMAROVSKY, and
N. STRAFFORD: Tables of Reagents for Inorganic Analysis. London: H. K.
Lewis. (Premier rapport de la commission internationale de reactions et
reactifs analytiques nouveau de l'union internationale de chimie.) Leipzig:
Akademische Verlagsgesellschaft. 1938.
(30) ROSIN, J.: Reagent Chemicals and Standards, 2nd ed. New York: Van
Nostrand. 1946.
(31) U. S. Pharmacopoeial Convention: Pharmacopoeia of the United States,
Washington, D. C.
(32) W'ELCHER, F. J.: Organic Analytical Reagents, 4 Vols. New York: Van
Nostrand. 1947.

Standard Tests and Procedures of Qualitative Analysis

(33) ALLEN, P. W., editor: Techniques of Polymer Characterization. New York:
Academic Press. 1959.
(34) BARBER, H. H., and T. 1. TAYLOR: Semimicro Qualitative Analysis.
New York: Harper. 1953.
(35) CAMPBELL, N.: Qualitative Organic Chemistry. New York: Van Nostrand.
1939.
(36) CHARLOT, G., DENISE BEZIER, and R GAUGUIN: Analyse qualitative'rapide
des cations. Paris: Dunod. 1950; translated by R E. OESPER: Rapid
Detection of Cations. New York: Chemical Publishing Co. 1954.
(37) FRESENIUS, W., and G. JANDER, editors: Handbuch der analytischen Ohemie,
2nd part, Vols. Ito IX. Berlin-Gottingen-Heidelberg: Springer. 1944-1956.
(38) KAMM, 0.: Qualitative Organic Analysis, 2nd ed. New York: John Wiley.
1932.
(39) LINSTEAD, R P., and B. C. L. WEEDON: A Guide to Qualitative Organic
Chemical Analysis. New York: Academic Press. 1956.
(40) MCALPINE, R K., and B. A. SOULE: Qualitative Chemical Analysis.
New York: Van Nostrand. 1933.
(41) McELVAIN, S. M.: Characterization of Organic Compounds. New York:
Macmillan. 1953.
 Literature 465

(42) Me GOOKIN, A.: Qualitative Organic Analysis and the Scientific Method.
New York: Reinhold. 1955.
(43) MELDRUM, W. B., and E. W. FLOSDORF: Qualitative Analysis of Inorganic
Materials. New York: American Book Co. 1938.
(44) MIDDLETON, A. R., and J. W. WILLARD: Semimicro Qualitative Analysis.
New York: Prentice-Hall. 1939.
(45) MORRISON, G. H., and H. FREISER: Solvent Extraction in Analytical
Chemistry. New York: John Wiley. 1957.
(46) MULLIKEN, S. F.: The Identification of Pure Organic Compounds, 4 Vols.
New York: John Wiley. 1904--1942.
(47) NIEUWENBURG, C. J. VAN, and J. W. LIGTEN, Qualitative chemische Analyse.
Wien: Springer. 1959.
(48) NOYES, A. A.: Qualitative Chemical Analysis of Inorganic Substances,
9th ed. New York: Macmillan. 1922.
(49) NOYES, A. A., and W. C. BRAY: A System of Qualitative Analysis for the
Rare Elements. New York: Macmillan. 1927.
(50) RADLEY, J. A., and J. GRANT: Fluorescence Analysis in Ultraviolet Light,
2nd ed. New York: Van Nostrand. 1935.
(51) SHRINER, R. L., R. C. FUSON, and D. V. CURTIN: The Systematic Identifica-
tion of Organic Compounds, 4th ed. New York: John Wiley. 1956.
(52) SIGGIA, S.: Quantitative Organic Analysis Via Functional Groups. New
York: John Wiley. 1949.
(53) SWIFT, E. H.: A System of Chemical Analysis (Qualitative and Semi-
Quantitative) for the Common Elements. New York: Prentice-Hall. 1939.
(54) Tables for Identification of Organic Compounds. Cleveland, Ohio: Chemical
Rubber Co.
(55) TREADWELL, F. P.: Kurus Lehrbuch der analytischen Ohemie, I. Qualitative
Analyse, 21st ed. Wien: Deuticke. 1948-1949.
(56) TREADWELL, F. P., and W. T. HALL: Analytical Chemistry, I. Qualitative
Analysis, 9th ed. New York: John Wiley. 1937.
(57) TREADWELL, F. P., and V. MEYER: Tabellen zur qualitativen Analyse,
7th ed. Wien: Deuticke. 1912.
(58) TREYBAL, R. E.: Liquid Extraction. New York: McGraw-Hill. 1951.
(59) VEIBEL, S.: The Identification of Organic Compounds, 2nd English ed.
Copenhagen: G. E. C. Gad. 1960.
(60) VOGEL, A. I.: Text Book of Macro and Semi Micro Qualitative Inorganic
Analysis, 4th ed. London: Longmans, Green and Co. 1954.
(61) VORTMANN, G., and R. LIEBER: Qualitative chemische Analyse nach dem
Schwefelnatriumgang. Wien: Haim. 1933.
(62) WEST, P. W., M. M. VICK, and A. L. LE ROSEN: Qualitative Analysis and
Analytical Separations (without H 2 S). New York: Macmillan. 1953.

Chromatography and Ion Exchange

(63) BLOCK, R. J., E. L. DURRUM, and G. ZWEIG: A Manual of Paper Chromato-
graphy and Paper Electrophoresis, 2nd ed . New York: Academic Press.
1958.
(64) BRIMLEY, R. C., and F. C. BARRETT: Practical Chromatography. New York:
Reinhold. 1953.
(65) CASSIDY, H. G.: Adsorption and Chromatography. New York: Interscience.
1951.
(66) CRAMER, F.: Papierchromatographie. Weinheim: Verlag Chemie. 1952.
Benedetti-Pichler, Identification 30
466 Literature

(67) HESSE, G.: Adsorptionsmethoden im chemischen Laboratorium. Berlin:

de Gruyter. 1943.
(68) LEDERER, E.. and M. LEDERER: Chromatography, 2nd ed. New York:
Van Nostrand. 1957.
(69) LINSKENS, H. F., editor: Papierchromatographie in der Botanik. Berlin-
Gottingen-Heidelberg: Springer. 1955.
(70) NAOHOD, F. C., editor: Ion Exchange, Theory and Application. New York:
Academic Press. 1949.
(71) POLLARD, F. H., and J. F. W. MOOMIE: Chromatographic Methods in
Inorganic Analysis. New York: Academic Press. 1953.
(72) SAMUELSON, 0.: Ion Exchangers in Analytical Chemistry. New York:
John Wiley. 1952.
(73) SMITH, O. C.: Inorganic Chromatography. New York: Van Nostrand. 1953.
(73a) STAHL, E., editor: Dilnnschicht-Chromatographie (a laboratory manual).
Berlin-Gottingen-Heidelberg: Springer. 1962.
(74) STRAIN, H. H.: Chromatographic Adsorption Analysis. New York: Inter-
science. 1942.
(75) WILLIAMS, R. T., and R. L. M. SYNGE: Partition Chromatography. Cam-
bridge: University Press. 1951.
(76) WILLIAMS, T. 1.: An Introduction to Chromatography. New York: Chemical
Publishing Co. 1947.

Instrumental Methods
(77) BERL, W. G., editor: Physical Methods in Chemical Analysis, Vol. 3.
New York: Academic Press. 1956.
(78) BRODE, W. R.: Chemical Spectroscopy, 2nded. New York: John Wiley. 1946.
(79) HAMOS, L. v.: X-Ray Microanalyzer Camera. Goteborg: Elanders
Boktryckeri Actiebolag. 1953.
(80) KLUG, H. P., and L. E. ALEXANDER: X-Ray Diffraction Procedures.
New York: John Wiley. 1954.
(81) LANG, L., editor: Absorption Spectra in the Ultraviolet and Visible Region,
3 Vols. New York: Academic Press. 1961.
(81 a) NAKANISHI, K.: Infrared Absorption Spectroscopy. San Francisco:
Holden-Day. 1962.
(82) NYBURG, S. C.: X-Ray Analyses of Organic Structures. New York:
Interscience. 1961.

Chemical Microscopy
(83) BARKER, T. V.: Systematic Crystallography. London: Murby and Co. 1930.
(84) BLASS, F. D.: An Introduction to the Methods of Optical Crystallography.
New York: Holt, Rinehart and Winston. 1961.
(85) CHAMBERS, R., in C. F. MOCLUNG: Handbook of Microscopical Technique,
1st, ed. New York: Hoeber. 1929.
(86) CHAMBERS, R., and M. J. KOPAO: Handbook of Microscopical Technique,
3rd ed. New York: Hoeber. 1950.
(87) CHAlIWT, E. M.: The Microscopy of Small Arms Primers. Ithaca, N. Y. 1922.
(88) CHAMOT, E. M., and C. W. MASON: Handbook of Chemical Microscopy,
Vol. I, 3rd ed. New York: John Wiley. 1958.
(89) CONN, G. K. T., and F. J. BRADSHAW, editors: Polarized Light in Metallo-
graphy. New York: Academic Press. 1952.
 Literature 467

(90) EL-BADRY, H. M.: Micromanipulators and Micromanipulation. (Mono-

graphien aus dem Gebiete der qualitativen Mikroanalys€, Vol. III.)
Wien: Springer. 1963.
(91) DONNAY, J. D. H., and W. NOVACKI: Crystal Data. Geological Society of
America. 1954.
(92) FONBRUNE,P.DE: Technique de micromanipulation. Paris: Massonet Oie. 1949_
(93) FRY, W. H.: Petrographic Methods for Soil Laboratories. Tech. Bull.
No. 344. Washington, D. 0.: U. S. Department of Agriculture. 1933.
(94) Groth's Encyclopedia of Ohemical and Physical Orystallography, revised ed.,
University Park, Pa., Groth Institute, Pennsylvania State University
(in preparation).
(95) HArTINGER, M.: Die Fluoreszenzanalyse in der Mikrochemie. Wien: Haim.
1937.
(96) HARTSHORNE, N. H., and A. STUART: Crystals and the Polarizing Micro-
scope, 2nd ed. New York: Longmans, Green and 00. 1950.
(97) KERR, P. F.: Optical Mineralogy, 3rd ed. New York: McGraw-Hill. 1959.
(98) KOFLER, L., ADELHEID KOFLER, and A. MAYRHOFER: Mikroskopische
J{ethoden in der Mikrochemie. Wien: Haim. 1936.
(99) LARSEN, E. S., and H. BERMAN: Microscopic Determination of the Non-
Opaque Minerals. U. S. Geol. Survey Bull. 848 (1934).
(100) MARTIN, L. 0., and B. K. JOHNSON: Practical Microscopy. New York:
Ohemical Publishing 00. 1951.
(101) MURDOCH, J.: Microscopical Determination of Opaque Minerals. New York:
John Wiley. 1916.
(102) OLLIVER, O. W.: The Intelligent Use of the Microscope. New York:
Ohemical Publishing 00. 1953.
(103) OTTO, L.: Der Mikromanipulator und seine Hil/sgeriite. Berlin: Verlag
Technik. 1954. See also (1133).
(104) PORTER, MARY W., and R. O. SPILLER: The Barker Index of Orystals,
3rd ed. Oambridge: Heffner and Sons. 1951-1959.
(105) SCHAEFFER, H. F.: Microscopy for Ohemists. New York: Van Nostrand.
1953.
(106) SHILLABER, O. P.: Photomicrography in Theory and Practice. New York:
John Wiley. 1944.
(107) SMITH, H. G.: Minerals and the Microscope, 4th ed. New York: Macmillan.
1956. (An introduction for beginners.)
(108) STOVES, J. L.: Fibre Microscopy. New York: Van Nostrand. 1958.
(109) UYTENBOGAARDT, W.: Tables for Microscopic Identification of Ore
Minerals. Princeton: University Press. 1951.
(110) WAHLSTROM, E. E.: Optical Orystallography, 2nd ed. New York: John
Wiley. 1943.
(111 ) WALLIS, T. E.: Analytical Microscopy, 2nd ed. Boston: Little, Brown
and 00. 1957.
(112) WINCHELL, A. N.: The Optical Properties of Organic Oompounds, 2nd ed.
New York: Academic Press. 1954.
(113) WINCHELL, A. N., and H. WINCHELL: Microscopic Oharacters of Artificial
Inorganic Solid Substances or Artificial Minerals, 3rd ed. New York:
Academic Press. 1961.
(114) WINCHELL, A. N., and H. WINCHELL: Elements of Optical Mineralogy;
I. Principles and Methods, 4th ed.; II. Descriptions of Minerals; III. De-
terminative Tables, 2nd ed. New York: John Wiley. 1951.
30·
468 Literature

Slide Tests and Spot Tests

(115) BEHRENS, H.: A Manual of Microchemical Analysis. London: Macmillan.
1894.
(ll6) BEHRENS, H., and P. D. C. KLEY: Mikrochemi8che AnalY8e, 4th ed.
Leipzig und Hamburg: Voss. 1921.
(ll7) BORICKY, E.: Elemente einer neuen chemi8ch-mikr08kopi8chen Mineral- und
GesteinsanalY8e, reprint from Archiv der naturwi88enschaftlichen Lande8-
durchfor8chung von Bohmen, Prague, 1877. Translated into English by
N. H. WINCHELL, 19th Annual Report of the Geological and Natural
History Survey of Minnesota, Minneapolis. 1892.
(ll8) CHAMOT, E. M., and C. W. MASON: Handbook of Chemical Microscopy,
Vol. II, 2nd ed. New York: John Wiley. 1940.
(ll9) DUVAL, C.: Traite de micro-analY8e minerale, 4 Vols. Paris: Presses
Scientifiques Internationales. 1951.
(120) FEIGL, F.: Qualitative Analy8e mit Hilfe von Tiipfelreaktionen. Leipzig:
Akademische Verlagsgesellschaft. 1931.
(121) FEIGL, F., and R. E. OESPER: Qualitative Analysis by Spot Tests, In-
organic and Organic Applications, 3rd ed. New York: Elsevier. 1948.
(122) FEIGL, F., and R. E. OESPER: Spot Tests, Vol. 2, Organic Applications.
Houston: Elsevier. 1954.
(123) GEILMANN, W.: Bilder zur qualitativen MikroanalY8e anorgani8cher Stoffe,
3rd ed. Weinheim: Verlag Chemie. 1960.
(124) HARTING, P. (translated from the Dutch original by F. W. THEILE):
Theorie und allgemeine Be8chreibung de8 Mikr08kope8. Braunschweig:
Vieweg. 1866.
(125) HAUSHOFER, K.: Anleitung zur Erkennung ver8chiedener Elemente mittel8
mikr08kopi8cher Reaktionen. Braunschweig: Vieweg. 1885.
(126) HELWIG, A.: Das Mikr08kop in der Toxikologie. Mainz. 1865.
(127) HINRICHS, C. G.: First Course in Microchemical Analysis. St. Louis, Mo.:
Carl Gustav Hinrichs; New York and Leipzig: Lemske and Buechner;
London: H. Grevel & Co.; Paris: H. Ie Soudier. 1904.
(128) HUYSSE, A. C.: Atlas zum Gebrauch bei der mikrochemi8chen AnalY8e.
Leiden: Brill. 1932.
(129) KLEMENT, C., and A. RENARD: Reaction8 microchimique8 a cri8taux et
leur application en analY8e qualitative. Brussels: Manceaux. 1886.
(130) KRAMER, G.: Mikroanalyti8che Nachwei8e anorgani8cher Ionen, Ausfiihrung
und Reaktion8bilder. Leipzig: Akademische Verlagsgesellschaft. 1937.
(131) LONGO, R. E.: Mikroanali8i8 inorganica, 2nd ed. Buenos Aires: Ciordia y
Rodriguez. 1951.
(132) MALJAROFF, K. L.: Qualitative anorgani8che MikroanalY8e. Berlin: Verlag
Technik. 1953.
(133) MARTINI, A., and S. SCHAMIS: Nuevo metoda para el reconocimiento micro-
quimico de l08 catione8 mas comune8 en la8 mezcla8 compleja8. Buenos Aires:
Trab. pres. al Segundo Congresa de Quimica. 1924.
(134) PozzI-EscOT, M. E. : AnalY8e microchimique et 8pectr08copic. Paris: Gauthier-
Villars. 1899.
(135) STEPHENSON, C. H., and E. C. PARKER: Some Microchemical Tests for
Alkaloids. Philadelphia: Lippincott. 1921.
(136) WENGER, P. E., and R. DUCKERT: Reactif8 pour l'analy8e qualitative
minerale, 2eme rapport. Basel: Wepf. 1945.
 Literature 469

(137) WENGER, P. E., and YVONNE RUSCONI: Reactifs pour l'analyse qualitative
minerale, 4eme rapport. Paris: Sedes. 1950.
(138) WORMLEY, TH. G.: The Micro-Chemistry of Poisons, 2nd ed. Philadelphia:
Lippincott. 1885.

Micro Analysis and Microtechnique

(139) ALIMARIN,1. P., and M. N. PETRIKOVA: Inorganic Ultramicro Analysis.
Moscow: Academy of Sciences. 1960.
(140) BELCHER, R, and C. L. WILSON: Inorganic Microanalysis, Qualitative
and Quantitative. A short Elementary Course, 2nd ed. London: Long-
mans, Green and Co. 1957.
(141) BENEDETTI-PICHLER, A. A.: Introduction to the Microtechnique of In-
organic Analysis. New York: John Wiley. 1942.
(142) BENEDETTI-PICHLER, A. A.: Waagen und Wiigen. (Handbuch der mikro-
chemischen Methoden, Vol. 1, edited by F. HECHT and M. K. ZACHERL,
Part 2.) Wien: Springer. 1959.
(143) BENEDETTI-PICHLER, A. A., and W. F. SPIKES: Introduction to the Micro-
technique of Inorganic Qualitative Analysis. Douglaston, N. Y.: Micro-
chemical Service. 1935.
(144) BRISCOE, H. V. A., and P. F. HOLT: Inorganic Microanalysis. London:
Longmans, Green and Co. 1950. .
(145) CHERONIS, N. D., editor: Submicrogram Experimentation. New York:
Interscience. 1961.
(146) CHERONIS, N. D., and J. B. ENTRIKIN: Semimicro Qualitative Organic
Analysis. The Systematic Identification of Organic Compounds, 2nd ed.
New York: Interscience. 1957.
(147) CHERONIS, N. D., with A. R RONZIO and T. S. MA: Micro and Semimicro
Methods. (Technique of Organic Chemistry, edited by A. WEISSBERGER,
Vol. VI.) New York: Interscience. 1954.
(148) DONAU, J.: Die Arbeitsmethoden der Mikrochemie. Stuttgart: Franckh'sche
Verlagshandlung. 1913.
(149) EMIeH, F.: Lehrbuch der Mikrochemie. Wiesbaden: Bergmann. 1911.
(150) EMICH, F.: Lehrbuch der Mikrochemie, 2nd ed. Munchen: Bergmann. 1926.
(151) EMICH, F.: Methoden der Mikrochemie, in ABDERHALDEN'S Handbuch der
biologischen Arbeitsmethoden, Abt. 1, Teil 3. Wien und Berlin: Urban und
Schwarzenberg. 1921.
(152) EMICH, F.: Mikrochemisches Praktikum, 1st ed. Munchen: Bergmann. 1924.
(153) EMICH, F.: Umsetzungen sehr kleiner Stoffmengen, in STAHLER-TIEDE-
RICHTER'S Handbuch der Arbeitsmethoden in der anorganischen Chemie,
Vol. 2, Part 2. Berlin: de Gruyter. 1925.
(154) EMICH, F., and F. SCHNEIDER: Microchemical Laboratory Manual.
New York: John Wiley. 1932.
(155) GORBACH, G.: Mikrochemisches Praktikum (Anleitungen fur die chemische
Laboratoriumspraxis, Band 7). Berlin-Gottingen-Heidelberg: Springer.
1956.
(156) HECHT, F., and M. K. ZACHERL, editors: Handbuch der mikrochemischen
Methoden. Wien: Springer. 1954.
(157) KIRK, P. L.: Quantitative Ultramicroanalysis. New York: John Wiley.
1950.
(158) KLEIN, G., and R STREBINGER: Fortschritte der Mikrochemie (1915-1926).
Wien: Deuticke. 1928.
470 Literature

(159) KOFLER, L., and ADELHEID KOFLER: Mikromethoden zur Kennzeichnung

organischer Stoffe und Stoffgemische. Innsbruck: Wagner. 1948.
(160) KOFLER, L., ADELHEID KOFLER, and MARIA BRANDSTATTER: Thermo-
Mikro-Methoden, 3rd ed. Weinheim: Verlag Chemie. 1954.
(161) LEVY, R.: Microanalyse organique elementaire qualitative et quantitative.
Paris: Masson et Cie. 1961.
(162) MALISSA, H., and A. A. BENEDETTI-PICHLER: Anorganische qualitative
M ikroanalyse (M onographien aus dem Gebiete der qualitativen M ikro-
analyse, Vol. I). Wien: Springer. 1958.
(163) MCCRONE, W. C.: Fusion Methods in Chemical Microscopy. New York:
Interscience. 1956.
(164) NIEDERL, J. B., and J. A. SOZZI: Microanalisis elemental organico. Buenos
Aires: (Arcos 2073, Buenos Aires 28) 1958.
(165) REID, D. B.: Rudiments of Chemistry, 3rd ed. Edinburgh. 1848.
(166) SCHNEIDER, F.: Organic Qualitative Analysis. New York: John Wiley.
1946.
(167) SCHNEIDER, F. L.: Practical Organic Qualitative Micro Analysis. Cognition
and Recognition. (Monographien aus dem Gebiete der qualitativen M ikro-
analyse, Vol. II). Wien: Springer. 1964.
(168) SCHOORL, N.: Beitriige zur mikrochemischen Analyse, a collection of papers
published in Z. analyt. Chem., Vols. 46, 47, 48 (1907-1909). Wiesbaden.
1909.
(169) SHORT, M. N.: Microscopic Determination of the Ore Minerals, Geological
Survey Bulletin 825, Washington, D. C.: Superintendent of Documents.
1931.
(170) STRONG, J., H. V. NEHER, A. H. WHITFORD, C. H. CARTWRIGHT, and
R. HAYWARD: Procedures in Experimental Physics. New York: Prentice-
Hall. 1938.
(171) TOEPLER, A.: Vol. 157-158 of OSTWALD'S Klassiker der exakten Wissen-
schaften. Leipzig: Engelmann. 1906.
(172) WEISZ, H.: Microanalysis by the Ring Oven Technique. New York:
Pergamon Press. 1961.
(173) WILSON, C. L.: An Introduction to Microchemical Methods for Senior
Students of Chemistry. New York: Chemical Publishing Co. 1938.

Miscellaneous Applications of Microtechnique

(174) AMELINK, F.: Schema zur mikrochemischen Identifikation von Alkaloiden.
Amsterdam: Centen's. 1934.
(175) CORRINGTON, J. D.: Exploring with Your Microscope. New York: McGraw-
Hill. 1957.
(176) DE WILD, A. M.: The Scientific Examination of Pictures. London:
Bell and Sons. 1929.
(177) ESAU, KATHERINE: Plant Anatomy. New York: John Wiley. 1953.
(178) GARNER, W.: Industrial Microscopy. London: Pitman. 1932.
(179) GRAFF, J. H.: Pulp and Paper Microscopy. Appleton, Wisc.: Inst. Paper
Chemistry. 1942.
(180) HANAUSEK, T. F., A. L. WINTON, and KATEE BARBER WINTON: The
. Microscopy of Technical Products. New York: John Wiley. 1907.
(181) HARRIS, K. L., editor: Microscopic-Analytical Methods in Food and Drug
Control. Washington, D. C.: Food and Drug Technical Bulletin No.1,
Superintendent of Documents. 1960.
 Literature 471

(182) HENRIeI, A. T., C. W. EMMONS, C. E. SKINNER, and H. M. TsucffiYA:

Molds, Yeasts, and Actinomycetes, 2nd ed. New York: John Wiley. 1947.
(183) HERZOG, A.: Mikrophotographischer Atlas der technisch wichtigsten Faser-
stolle. Munchen: Obernetter. 1908.
(184) HEYN, A. N. J.: Fiber Microscopy. New York: Interscience. 1954.
(185) JOHANSEN, D. A.: Plant Microtechnique. New York: McGraw-Hill. 1940.
(186) KIRK, P. L.: Crime Investigation. New York: Interscience. 1953.
(187) KLINGER, P., and W. KOCH: Beitrage zur metallkundlichen Analyse.
Dusseldorf : Verlag Stahleisen. 1949.
(188) KOCH, P. A., in H. SOMMER and F. WINKLER: Die Prillung der Textilien.
Berlin-Gottingen-Heidelberg: Springer. 1960.
(189) LEIFSON, E.: Atlas of Bacterial Flagellation. New York: John Wiley. 1959.
(190) LINDSLEY, L. C.: Industrial Microscopy. Richmond, Va.: Byrd. 1929.
(191) MAUERSBERGER, H. R., editor: Matthews' Textile Fibers, 6th ed. New
York: John Wiley. 1954.
(192) MAYRHOFER, A.: Mikrochemie der Arzneimittel und Gilte, 2 Parts. Berlin:
Urban und Schwarzenberg. 1928.
(193) MAYRHOFER, A.: Qualitative mikrochemische Methoden zur Untersuchung
der Heilmittel, in ABDERHALDEN'S Handbuch der biologischen Arbeits-
methoden, Abt. IV, Teil7 C. Berlin: Urban und Schwarzenberg. 1929.
(194) MOLISCH, H.: Mikrochemie der Pilanze, 3rd ed. Jena: Fischer. 1923.
(195) PARRY, J. W.: Spices, Their Morphology, Histology, and Chemistry.
New York: Chemical Publishing Co. 1962.
(196) PAULSEN: Botanisk Mikrokemi. Copenhagen: 1918.
(197) P6SCHL, V.: Technische Mikroskopie. Stuttgart: Union Deutsche Verlags-
gesellschaft. 1927. (Excellent Illustrations.)
(198) ROSENTHALER, L.: Toxikologische Mikroanalyse. Berlin: Borntraeger. 1935.
(199) SANDELL, E. B.: Colorimetric Determination of Traces of Metals. New
York: Interscience. 1950.
(200) SCHNEIDER, A.: Microanalysis of Powdered Vegetable Drugs. Philadelphia:
Blakiston's Son. 1920.
(201) SCHNEIDER, A.: Microbiology and Microanalysis of Foods. Philadelphia:
Blakiston's Son. 1920.
(202) SCHNEIDER, H., and A. ZIMMERMANN: Botanische Mikrotechnik. Jena:
Fischer. 1922.
(203) STERN,A.C., editor: Air Pollution, 2Vols. New York: Academic Press. 1962.
(204) STOVES, J. L.: Fibre Microscopy. London: Natl. Trade Press. 1957;
New York: Van Nostrand. 1958.
(205) TUNMANN, 0., and L. ROSENTHALER: Pllanzenmikrochemie. Berlin: Born-
traeger. 1931.
(206) VESCE, V. C.: Classification and Microscopic Identification of Organic
Pigments, in MATIELLO: Protective and Decorative Coatings, Vol. 2.
New York: John Wiley. 1942.
(207) WHIPPLE, G. C., revised by G. M. FAIR and M. C. WHIPPLE: The Micro-
scopy of Drinking Water, 4th ed. New York: John Wiley. 1927.

Mineralogy
See also Chemical Microscopy.
(208) BRUSH, G. J., and S. L. PENFIELD: Manual of Determinative Mineralogy,
16thed. New York: John Wiley. 1926.
(209) CAMERON,E.N.: Ore Microscopy. New York: John Wiley. 1961.
472 Literature

(210) DANA, J. D., C. PALACHE, H. BERMAN, and.C. FRONDEL: The System of

Mineralogy, 7th ed. 2 Vols. New York: John Wiley. 1951.
(211) DANA, E. S., and C. HURLBUT, Jr.: Minerals and How to Study Them,
3rd ed. New York: John Wiley. 1949.
(212) LEWIS, J. V., and A. C. HAWKINS: A Manual of Determinative Mineralogy,
4th ed. New York: John Wiley. 1931.

Journals
Acta Chemica Scandinavica
(300) 3, 630 (1949): J. N. OSPENSON.

American Journal of Botany

(310) 30, 477 (1943): B. ESTHER STRUCKMEYER.

American Mineralogi8t
(320) 43, 606 (1958): W. W. VIRGIN, Jr., and C. J. MASSONI.

Analitikeskoi Kimii
(330) 10, 251 (1955): I. P. ALmARIN and M. N. PETRIKOWA.

Analytica Chimica Acta

(340) 3, 15 (1949): F. FEIGL and L. BAUMFELD.
(341) 3, 629 (1949): H. FLASCHKA.
(342) 19, 437 (1958): D. GOLDSTEIN and C. STARK-MEYER.

Analytical Chemi8try
(Analytical Edition of Industrial and Engineering Chemistry, 1929-1946)
(400) 2, 177 (1930): E. R. CALEY.
(401) 2, 309 (1930): A. A. BENEDETTI-PICHLER.
(402) 3, 266 (1931): F. E. BLACET and P. A. LEIGHTON.
(403) 4, 336 (1932): A. A. BENEDETTI-PICHLER.
(404) 5, 272 (1933): F. E. BLACET, G. D. MAC DONALD, and P. A. LEIGHTON.
(405) 6, 334 (1934): F. E. BLACET and G. D. MAC DONALD.
(406) 7, 25 (1935): H. SCHAPIRO.
(407) 7, 218 (1935): B. L. CLARKE and H. W. HERMANCE.
(408) 9, 44 (1937): F. E. BLACET and D. H. VOLMAN.
(409) 9, 149 (1937): A. A. BENEDETTI-PICHLER.
(410) 9, 292 (1937): B. L. CLARKE and H. W. HERMANCE.
(411) 9, 483 (1937): A. A. BENEDETTI-PICHLER.
(412) 9, 496 (1937): A. C. SHEAD.
(413) 9, 589 (1937): A. A. BENEDETTI-PICHLER and J. R. RACHELE.
(414) 10, 47 (1938): H. K. ALBER and C. J. RODDEN.
(415) 10, 107 (1938): A. A. BENEDETTI-PICHLER and J. T. BRYANT.
(416) 10, 224 (1938): C. VAN BRUNT.
(417) 10, 348 (1938): H. K. ALBER.
(418) 10, 591 (1938): B. L. CLARKE and H. W. HERMANCE.
(419) 10, 662 (1938): A. C. SHEAD.
 Literature 473

(420) 11, 117 (1939): A. A. BENEDETTI-PICHLER, W. R. CROWELL, and

C. DONAHUE.
(421) 11, 294 (1939): B. S. ALSTODT and A. A. BENEDETTI-PICHLER.
(422) 11, 403 (1939): P. L. KIRK and C. S. GIBSON.
(423) 11, 409 (1939): J. R. BOWMAN.
(424) 12, 233 (1940): A. A. BENEDETTI-PICHLER and J. R. RACHELE.
(425) 12, 305 (1940): H. K. ALBER and J. T. BRYANT.
(426) 12, 764 (1940): H. K. ALBER.
(427) 12, 777 (1940): A. MARION.
(428) 13, 127 (1941): W. G. BATT and H. K. ALBER.
(429) 13, 494 (1941): A. A. MORTON and J. F. MAHONEY.
(430) 13, 498 (1941): A. A. MORTON and J. F. MAHONEY.
(431) 13, 656 (1941): H. K. ALBER.
(432) 14, 278 (1942): W. W. RAZIM.
(433) 14, 813 (1942): A. A. BENEDETTI-PICHLER and M. CEFOLA.
(434) 15, 135 (1943): H. YAGODA.
(435) 15, 227 (1943): A. A. BENEDETTI-PICHLER and M. CEFOLA.
(436) 15, 648 (1943): C. R. GARCIA.
(437) 17, 187 (1945): ANNE G. LOSCALZO and A. A. BENEDETTI-PICHLER.
(438) 17, 593 (1945): J. D. H. DONNAY and W. A. O'BRIEN.
(439) 18, 81 (1946): D. SMITH and SHffiLEY A. EHRHARDT.
(440) 19, 77 (1947): G. SHEPHERD.
(441) 19, 355 (1947): P. L. KIRK, R. S. ROSENFELD, and J. D. HANAHAN.
(442) 20, 976 (1948): G. C. CROSSMON.
(443) 20, 1122 (1948): P. L. KffiK and M. DANIELSON.
(444) 20, 1241 (1948): H. H. ANDERSON.
(445) 21, 632 (1949): M. J. BABCOCK.
(446) 21, 700 (1949): F. W. CHAPMAN, Jr., G. G. MARVIN, and S. YOUNG
TYREE, Jr.
(447) 22, 600 (1950): A. O. GETTLER, C. J. UMBERGER, and L. GOLDBAUM.
(448) 22, 628 (1950): R. S. TIPSON.
(449) 22, 892 (1950): E. H. GILMORE, MARIE MENAUL, and V. SCHNEIDER.
(450) 23, 196 (1951): R. D. CADLE.
(451) 23, 545 (1951): A. 1. MEDALIA and R. W. STOENNER.
(452) 24, 576 (1952): B. K. SEELY.
(453) 24, 870 (1952): P. W. WEST and L. GRANATELLI.
(454) 26, 1515 (1954): W. PRIMAK and P. DAY.
(455) 26, 1829 (1954): J. P. LODGE.
(456) 27, 93 (1955): B. K. SEELY.
(457) 27, 704 (1955): J. L. MONKMAN.
(458) 27, 865 (1955): A. J. FRANKLIN and S. E. VOLTZ.
(459) 28, 1586 (1956): D. G. GRABAR and RITA HAESSLY.
(460) 29, 167 (1957): S. T. ZENCHELSKY and J. S. SHOWELL.
(461) 29, 167 (1957): R. C. BACKUS.
(462) 29, 169 (1957): E. D. BLACK, J. D. MARGERUM, and G. M. WYMAN.
(463) 29, 45 A (June, 1957): L. T. HALLETT.
(464) 29, 860 (1957): BETTY J. STEINBACH and T. R. P. GIBB, Jr.
(465) 29, 861 (1957): E. C. FIEBIG, E. L. SPENCER, and R. N. McCoy.
(466) 29, 1239 (1957): H. B. BRADLEY.
(467) 30, 593 (1958): I. FANKUCHEN.
(468) 31, 84 A (January, 1959): Norton Company.
474 Literature

(469) 31, 148 (1959): D. E. LASKOWSKI and O. W. ADAMS.

(470) 31, 238 (1959): BARBARA J. TUFTS.
(471) 31, 242 (1959): BARBARA J. TUFTS.
(472) 31, 456 (1959): D. F. H. WALLACH, D. M. SURGENOR, J. SODERBERG, and
E. DELANO.
(473) 31, 947 (1959): P. W. WEST and A. K. MUKHERJI.
(474) 31, 1124 (1959): J. S. MATTHEWS and N. D. COGGESHALL.
(475) 31, 1287 (1959): J. E. STEWART.
(476) 32, 25 A (February, 1960): staff report.
(477) 32, 19 R (1960): S. DAL NOGARE.
(478) 32, 87 R (1960): G. COVEN and R. Cox.
(479) 32, 225 R (1960): R. C. HIRT.
(480) 32, 229 R (1960): B. F. SCRIBNER.
(481) 32, 238 R (1960): R. C. GORE.
(482) 32, 240 R (1960): H. A. LIEBHAFSKY, E. H. WINSLOW, and H. PFEIFFER.
(483) 33, 1559 (1961): L. D. METCALFE.
(484) 34, 143 A (February, 1962): R. H. MtLLER.
(485) 34, 175 (1962): F. DAVIS, C. A. DUBBS, and W. S. ADAMS.
(486) 34, 242 (1962): M. SPARAGANA and W. B. MASON.
(487) 34, 448 (1962): C. SZONYI and J. D. GRASKE.
(488) 34, 880 (1962): C. D. FELTON.
(489) 34, 966 (1962): E. M. DODSON.
(490) 34, 1077 (1962): J. P. FARIS and J. W. WARTON.
(491) 34, 1183 (1962): L. SPIALTER and M. BALLESTER.
(492) 34, 1319 (1962): M. H. SWANN and M. L. ADAMS.
(493) 34, 1346 (1962): R. WASICKY.
(494) 34, 1370 (1962): A. M. LIEBMAN.
(495) 35, 621 (1963): T. O. ZIEBOLD and R. E. OGILVIE.

Annalen der Chemie

(550) 351, 426 (1907): F. EMICR.

Annales de chimie analytique et de chimie appliquee

(560) 8, 130 (1926): R. MEURICE.

Archiv fur das Eisenhuttenwesen

(570) 28, 785 (1957): W. KOCH, H. MALISSA, and D. DITGES.

Archiv fur experimentelle Pathologie und Pharmakologie

(580) 135, 188 (1928): H. EITEL.

Australian Chemical Institute, Journal &: Proceedings

(590) 4, 26 (1937): F. P. DWYER.
(591) 5, 32 (1938): F. P. DWYER.

Berichte der deutschen chemischen Gesellschaft .

(600) 9, 217 (1876): V. WARTHA.
(601) 16, 2234 (1883): E. FISCHER.
 Literature 475

(602) 42,1126 (1909): H. v. WARTENBERG.

(603) 46, 255 (1913): O. BINDER and R. F. WEINLAND.
(604) 51, 1739 (1918): F. PANETH.
(605) 56, 1338 (1923): J. VON BRAUN, W. GMELIN, and A. SCHULTHEISS.
(606) 61, 1654 (1928): G. JAEGER.
(607) 73, 1388 (1940): L. KOFLER and R. WANNENMACHER.

Chemist-Analyst
(620) 32, 4 (1943): F. FEIGL and H. A. SUTER.
(621) 36, 38 (1947): A. F. GILMAN, Jr.
(622) 49, 4 (1960): M. FUJIMOTO.
(623) 49, 20 (1960): R. DELHEZ.
(624) 49, 113 (1960): F. L. HAHN.
(625) 49, 114 (1960): J. T. STOCK and M. A. FILL.
(626) 50, 80 (1961): P. J. HOWE, J. S. HILL, and J. D. SLATER.

Comptes rendus de l'academie des sciences, Paris

(640) 100, 605 (1885): LECOQ DE BOISBAUDRAN.
(641) 176, 1012, 1187 (1923): A. POLICARD.
(642) 180, 538 (1925): L. HERRERA.

Die Chemie - Zeitschrift fur angewandte Chemie

(650) 31, 50 (1918): C. G. SCHWALBE.
(651) 39, 461 (1926): A. STOCK.
(652) 42, 954 (1929): A. A. BENEDETTI-PICHLER.
(653) 55, 244 (1942): R. FISCHER.

Economic Geology and the Bulletin of the Society of Economic Geologists

(660) 26, 415 (1933): M. H. HAYCOK.

Endeavour
(665) 10, 188 (1951): MARY W. PORTER.

Farben-Zeitung
(670) 31, 1456 (1926): C. P. VAN HOEK-HILVERSUM.

Gerberei Collegium
(672) 1916, 16: W. MOELLER.

Halle aux cuirs

(675) 1925, 23, 39, 73: C. GENOT.

Industrial and Engineering Chemistry

(680) 9, 969 (1917): E. M. CHAMOT and H. I. COLE.
(681) 10, 48 (1918): E. M. CHAMOT and H. I. COLE.
(682) 15, 725 (1923): E. B. SPEAR and H. A. ENDRES.
476 Literature

J ahrbuch der M ineralogie

(690) 1876, 1215: W. N. HARTLEY.
(691) 1877, 1251: W. N. HARTLEY.

Journal of the American Chemical Society

(695) 21, 417 (1899): V. LENHER and J. S. C. WELLS.
(696) 27, 104 (1905): T. W. RICHARDS.
(697) 33, 718 (1911): L. J. CURTMANN and P. ROTHBERG.
(698) 39, 1148 (1917): N. F. HALL.
(699) 39, 2186 (1917): O. R. SWEENEY.
(700) 47, 2625 (1925): H. S. BOOTH and N. E. SCHREIBER.
(701) 62, 3165 (1940): C. A. HUTCHISON and H. L. JOHNSTON.

Journal of the Biological Photographers Association

(708) 8, 115 (1940): G. L. ROYER and MARIE E. WISSEMANN.

Journal of Chemical Education

(710) 11, 624 (1934): J. DUNNING and P. PRATT with O. E. LOWMAN.
(711) 34, 381 (1957): A. A. BENEDETTI-PICHLER, F. SCHNEIDER, and O. F.
STEINBACH.
(712) 34, 383 (1957): R. E. FRANK.
(713) 35, 453 (1958): R. T. CONLEY.

Journal of the Chemical Society, London

(720) 93, 1442 (1908): O. BRILL and C. DE B. EVANS.
(721) 125, 1946 (1924): L. REEVE.
(722) 138, 362 (1935): A. LAWSON and E. W. BALSON.
(723) 1940, 1258: CHRISTINA C. MILLER and A. J. LOWE.
(724) 1941, 786: CHRISTINA C. MILLER.

Journal of the Franklin Institute

(730) 182, 19 (1916): G. K. BURGESS.

Journal of the Institution of Petroleum Technologists

(735) 34, 331 (1948): E. GLYNN and L. GRUNBERG.

Journal of the Oil and Colour Chemists' Association

(740) 9, 255 (1926): J. PARRISH.

Journal of the Optical Society of America

(745) 23, 299 (1933): S. B. HENDRICKS and M. E. JEFFERSON.
(746) 45, 740 (1955): W. THORNBURG.

Journal fur praktische Chemie

(750) (1) 74, 341 (1858): J. LOWE.
 Literature 477

Journal ot Research ot the National Bureau ot Standards

(755) 57, 137 (1956): A. R. GLASCOW, Jr., and G. Ross.
(760) 10, 378 (1933): H. G. TAYLOR and J. M. W ALDRAM.
(761) 34, 207 (1957): P. R. ROWLAND and G. W. WHITING.

Laboratory Methods
(765) 55, 151 (1957): G. L. KELLY, H. STEINMETZ, and W. G. MCGONNAGLE.

Klinische W ochenschrift
(766) 32, 988 (1954): G. NOLLER.

Kolloidchemische Beihefte
(767) 23, 309 (1927): J. MIKA.

Metall und Erz

(768) 12, 189 (1929): G. GRANIGG.

Microchemical J oumal
(770) 2, 3 (1958): A. A. BENEDETTI,PICHLER.
(771) 2, 43 (1958): N. D. CHERONIS.
(772) 2, 205 (1958): M. CEFOLA.
(773) 3, 285 (1959): J. KRc, Jr.
(774) 3, 323 (1959): A. A. BENEDETTI-PICHLER.
(775) 3, 433 (1959): N. D. CHERONIS.
(776) 3, 515 (1959): L. FINE and E. A. WYNNE.
(777) 4, 423 (1960): N. D. CHERONIS.
(778) 4, 459 (1960): G. T. CHANG and A. A. BENEDETTI· PICHLER.
(779) 5, 331 (1961): A. A. BENEDETTI-PICHLER.
(780) 5, 525 (1961): N. D. CHERONIS.

Mikrochemie, 1-24 (1923-1938)

Mikrochemie vereinigt mit Mikrochimica Acta, 25-40 (1938-1953)
(850) 1, 4 (1923): F. FEIGL.
(851) 2, 138 (1924): F. STEIDLER.
(852) 2, 197 (1924): A. POLICARD.
(853) Emich-Festschrift 152 (1930): H. HETTERICH.
(854) Emich·Festschrift 243 (1930): P. RAY and P. B. SARKAR.
(855) Emich·Festschrift 275 (1930): A. SOLTYS.
(856) 8, 77 (1930): F. L. HAHN.
(857) 9, 31 (1931): F. L. HAHN.
(858) 9, 385 (1931): H. J. BRENNEIS.
(859) 10, 313 (1931): F. L. HAHN.
(860) 10, 380 (1932): H. HETTERICH.
(861) 11, 167 (1932): A. O. GETTLER, J. B. NIEDERL, and A. A. BENEDETTI·
PICHLER.
(862) 14, 219 (1933-1934): H. K. ALBER.
(863) 15, 271 (1934): A. A. BENEDETTI·PICHLER and W. F. SPIKES.
(864) 17, 165 (1935): F. FEIGL, J. V. SANCHEZ, and R. ZAPPERT.
478 Literature

(865) 17, 279 (1935): W. F. WHITMORE and H. SCHNEIDER.

(866) Molisch-Festschrift 3, 36 (1936): A. A. BENEDETTI-PICHLER and W. F.
SPIKES.
(867) 18, 272 (1935): F. ]'EIGL and O. FREHDEN.
(868) 18, 289 (1935): B. L. CLARKE and H. W. HERMANCE.
(869) 19, 1 (1935): A. A. BENEDETTI-PICHLER and J. R. RACHELE.
(870) 19, 239 (1935): A. A. BENEDETTI-PICHLER and J. R. RACHELE.
(871) 21, 98 (1937): A. K. RUSSANOW.
(872) 21, 133 (1936): J. W. YOUNG.
(873) 21, 268 (1937): A. A. BENEDETTI-PICHLER and W. F. SPIKES.
(874) 21, 300 (1937): A. A. BENEDETTI-PICHLER.
(875) 26,29 (1939): J. I. ADAMS, A. A. BENEDETTI-PICHLER, and J. T. BRYANT.
(876) 26, 143 (1939): D. L. JOHNSON and C. L. SHREWSBURY.
(877) 26, 182 (1939): E. BONTINCK.
(878) 27, 47 (1939): G. BECK.
(879) 27, 249 (1939): W. F. WHITMORE and C. A. WOOD.
(880) 28, 1 (1939): W. F. WHITMORE and C. A. WOOD.
(881) 31, 263 (1943): G. SKALOS.
(882) 31, 309 (1943): G. GORBACH.
(883) 33, 281 (1948): O. KONIG, W. R. CROWELL, and A. A. BENEDETTI-PICHLER.
(884) 33, 300 (1948): O. KONIG and W. R. CROWELL.
(885) 33, 303 (1948): O. KONIG and W. R. CROWELL.
(886) 33, 316 (1948): R. FISCHER and G. KARASEK.
(887) 33, 385 (1948): J. ERDOS.
(888) 34, 39 (1948): A. A. BENEDETTI-PICHLER.
(889) 34, 319 (1949): R. FISCHER and E. NEUPAUER. Tables of data.
(890) 34, 382 (1949): J. LINDNER.
(891) 34, 395 (1949): H. MALISSA.
(892) 35, 34 (1950): H. MALISSA.
(893) 35, 135 (1950): S. S. BURKE.
(894) 35, 213 (1951): H. MALISSA.
(895) 35, 266 (1951): H. MALISSA.
(896) 36/37, 224 (1950): M. C. ALVAREZ QUEROL and C. L. WILSON.
(897) 36/37, 296 (1950): R. FISCHER.
(898) 38, 33 (1951): H. MALISSA.
(899) 38, 50 (1951): J. GILLIS.
(900) 38, 100 (1951): P. W. WEST and W. C. HAMILTON.
(901) 38, 342 (1951): R. FISCHER.
(902) 38, 471 (1951): J. S. WIBERLEY, R. K. SIEGFRIEDT, and A. A. BENEDETTI-
PICHLER.
(903) 40, 141 (1952): H. M. EL-BADRY and C. L. WILSON.
(904) 40, 245 (1952): T. S. MA and R. TEN EYCK SCHENK.

Mikrochimica Acta, 1-3, (1937-1938), 1953-1962

Jlikrochimica et lchnoanalytica Acta, 1963
(910) 1, 266 (1937): H. V. A. BRISCOE and .JANET W. MATTHEWS.
(911) 2, 9, 287 (1937): G. BECK.
(912) 3, 30 (1938): C. DUVAL and P. FAUCONNIER.
(913) 3, 243 (1938): S. AUGUSTI.
(914) 1953, 305: ALICE LACOURT, G. SOMMEREYNS, C. FRANCOTTE, and N. DE-
LANDE.
 Literature 479

(915) 1953, 332: G. SOMMEREYNS.

(916) 1954, 140: H. WEISZ.
(917) 1954, 376: H. WEISZ.
(918) 1954, 460, 785: H. WEISZ.
(919) 1955, 134: H. JURANY.
(920) 1955, 821:' R. M. RUSH and L. B. ROGERS.
(921) 1956, 422: J. MITCHELL, Jr., and ADA L. RYLAND.
(922) 1956, 667: H. WEISZ.
(923) 1956, 1225: H. WEISZ.
(924) 1956, 1565: B. FLASCHENTRAGER and M. S. ZEIN.
(925) 1956, 1705: J. KORBL.
(926) 1956, 1729: CHARLOTTE L. BROWN and P. L. KmK.
(927) 1956, 1783: J. KOLBEK.
(928) 1956, 1856: H. WEISZ and F. SCOTT.
(929) 1957, 341: F. FEIGL, V. GENTIL, and C. STARK-MAYER.
(930) 1957,390: B. FLASCHENTRAGER, SAMIHA M. ABDEL-WAHAB, and G. HABIB
LABIB.
(931) 1957, 417: W. KNODEL and H. WEISZ.
(932) 1957, 427: MARIA BRANDSTATTER-KuHNERT and H. GRIMM.
(933) 1957, 501: J. W. SHELL, C. F. POE, and N. F. WITT.
(934) 1957, 527: E. WIESENBERGER.
(935) 1957, 567: A. A. BENEDETTI-PICHLER and H. E. SCHNEIDER.
(936) 1957,640: P. SENISE. The tolidine is dissolved in 10% acetic acid; private
communication.
(937) 1957,
714: P. L. KmK and CHARLOTTE L. BROWN.
(938) 1957,
720: CHARLOTTE L. BROWN and P. L. KIRK.
(939) 1957,
726: F. FEIGL, J. R. AMARAL, and V. GENTIL.
(940) 1957,
736: P. LUIS and A. CORAZZA.
(941) 1957,
751: H. BALLCZO and H. WEISZ.
(942) 201: R. BELCHER, R. HARRISON, and W. I. STEPHEN.
1958,
(943) 1958,
248, 253: G. SANDRI.
(944) 305: B. M. TURNER.
1958,
(945) 1958,
337: F. FEIGL and J. R. AMARAL.
(946) 1958,
353: L. W. BRADFORD and J. W. BRACKETT.
(947) 1958,
411: CH. R. WILLMS and W. M. HARDING.
(948) 1958,
577: V. M. BHUCHAR.
(949) 1958,
630: O. MANNS and S. PFEIFER.
(950) 1959,
32: H. WEISZ.
(951) 36: H. WEISZ, M. B. OELOP, and V. V. ALMAZAN.
1959,
(952) 1959,
87: W. COLLOM and P. L. KmK.
(953) 1959,
314: H. BALLCZO.
(954) 1959,
357: G. ACKERMANN.
(955) 1959,
406: G. SCHMIDT.
(956) 1959,
432: A. C. SHEAD.
(957) 1959,
541: P. LUIS.
(958) 1959,
657: A. C. SHEAD.
(959) 1959,
801: J. A. JAECKER and F. SCHNEIDER.
(960) 1960,
31: J. W. SHELL, N. F. WITT, and C. F. POE.
(961) 1960,
592: V. ANGER and G. FISCHER.
(962) 1960,
650: F. L. HAHN.
(963) 1960,
830: J. MAHON and A. A. BENEDETTI-PICHLER.
480 Literature

(964) 1960, 946: E. WIESENBERGER.

(965) 1960, 967: ARLEEN PIERCE, R. LOESCH, and F. SCHNEIDER.
(966) 1961, 11: R. WEISS.
(967) 1961, 65: J. CHURACEK.
(968) 1961, 140, 145, 149: H. REIMERS.
(969) 1961, 899: G. PRAZAK.
(970) 1962, 421: A. C. SHEAD.
(971) 1962, 490, 498: A. C. SHEAD.
(972) 1962, 529: J. J. PEIFER.
(973) 1962, 830: H. WEISZ.
(974) 1962, 922: H. WEISZ.
(975) 1962, 1165: P. W. WEST, A. J. LLACER, and C. CIMERMAN.
(976) 1963, 104: J. P. CRISLER, N. F. WITT, and MARJORIE H. CRISLER.
(977) 1963, 316: J. P. CRISLER, N. F. WITT, and MARJORIE H. CRISLER.
(978) 1963, 416: E. R. Du FRESNE.

M onatshefte fur Chemie

(1000) 22, 670 (1901): F. EMICH.
(1001) 23, 76 (1902): F. EMICH.
(1002) 26, 545 (1904): J. DONAU.
(1003) 26, 915 (1904): J. DONAU.
(1004) 28, 825 (1907): F. EMICH and J. DONAU.
(1005) 29, 959 (1908): J. DONAU.
(1006) 34, 949 (1913): J. DONAU.
(1007) 38, 219 (1917): F. EMICH.
(1008) 39, 775 (1918): F. EMICH.
(1009) 42, 411 (1921): H. SCHEUCHER.
(1010) 43, 129 (1922): A. FUCHS.
(1011) 43, 405 (1922): F. LANYAR and L. ZECHUR.
(1012) 46, 261 (1925): F. EMICH.
(1013) 46, 265 (1925): J. VOGEL.
(1014) 60, 745 (1928): F. EMICH.
(1015) 63/04, 312 (1929): F. EMICH et al.
(1016) 63/04, 335 (1929): F. EMICH and H. HAUSLER.
(1017) 68, 399 (1931): E. SCHALLY.
(1018) 64, 385 (1934): E. SCHALLY and F. NAGL.
(1019) 66, 153 (1935): J. HARAND.

Nature
(1050) 134, 809 (1934): Y. D. BERNAL and D. CROWFOOT.
(1051) 136, 305 (1935): Y. D. BERNAL and D. CROWFOOT.
(1052) 172, 809 (1953): J. E. EDSTROM.
(1053) 174, 128 (1954): J. E. EDSTROM and H. HYDEN.
(1054) 179, 628 (1957): S. H. U. BOWIE and K. TAYLOR.
(1055) 180, 50 (1957): H. H. ALLEN.
(1056) 183, 1423 (1959): V. E. COSLETT.

Die N aturwissenschaften
(1060) 38, 287 (1951): P. DECKER.
 Literature 481

Neues Jahrbuch fiir Mineralogie, Geologie und Paliiontologie, Abhandlungen

(1065) 89, 149 (1956): W. LINDENBERG.

Proceedings of the Physical Society, London

(1068) 44, 511 (1932): A. FERGUSON and J. S. KENNEDY.

Recueil des travaux des Pays-Bas et de la Belgique

(1070) II, 16, 369 (1898): M. H. HEMMES.

The Laboratory (Fisher Scientific Co.)

(1072) 31, 48 (1963).

The Review of Scientific Instruments

(1075) 28, 256 (1957): J. S. COURTNEy-PRATT and C. M. HUGGINS.

Science
(1080) 73, 344 (1931): D. Du BOIS.

Sitzungsberichte der Akademie der Wissenschaften in Wien. Abteil'ung I

(1085) 111, 171 (1902): O. RICHTER.
(1086) 130, 383 (1922): H. BRUNSWIK.
(1087) 142, 339 (1933): M. HAITINGER.

Skandinavisches Archiv fiir Physiologie

(1090) 20, 279 (1908): A. KROGH.

Spectrochimica Acta
(1092) 12, 276 (1958): D. A. CLARK and A. P. BOER.

Technical Studies in the Field of Fine Arts

(1092) 1, 1 (1932): R. G. GETTENS.
(1093) 2, 185 (1934): R. G. GETTENS.
(1094) 7, 200 (1938): R. G. GETTENS; data on painting materials.
(1095) 8, 12 (1939): R. G. GETTENS; data on painting materials.

The Analyst
(1100) 57, 2, 107 (1932): J. C. MABY.
(1101) 63, 467 (1938): JANET W. MATTHEWS.

The New Scientist, London

(1105) October 31, 1957: L. W. CODD and W. T. MOORE.

Transactions of the Institution of Mining Engineers, London

(1108) June 1937: H. V. A; BRISCOE, JANET W. MATTHEWS, P. F. HOLT, and
PHYLLIS M. SANDERSON.
Benedetti-Pichler, Identification 31
482 Literature

U. S. National Bureau at Standards Publications

(lll0) Bulletin 3, 345 (1905): G. K. BURGESS.
(lll1) Bulletin 11, 591 (1915): G. K. BURGESS and WILTTENBERG.
(lll2) Research Paper, RS 1809 (1947): B. W. SCRIBNER and W. K. WILSON.

U. S. Department of Agriculture Publications

(1l20) Bur. Chem. Bull. 122, 97 (1909); 137, 189 (19ll): B. J. HOWARD and
C. H. STEPHENSON.

U. S. Geological Survey Publications

(1l25) Bulletin 679 (1921): E. S. LARSEN, The Microscopic Determination of
the Nonopaque Minerals.

V ierteljahresschrift filr gerichtliche M edizin

(1l30) 3. Folge, 59, 233 (1920): G. STRASSMANN.

Zeitschritt des Vereines deutscher I ngenieure

(1l33) 94, 754 (1952): L. OTTO.

Zeitschrift tilr analytische Ghemie

(1l35) 28, 374 (1889): E. LEGER.
(1l36) 36, 195 (1897): L. SUHRE.
(1l37) 46, 658 (1907): N. SCHOORL.
(1l38) 47, 18 (1908): O. LUTZ.
(1l39) 1)4, 493 (1915): F. EMlcH.
(1l40) 1)4, 500 (1915): F. EMlCH.
(1l41) 56, 1 (1917): F. EMlCH.
(1l42) 62, 284 (1922): R. KEMPF.
(1l43) 70, 257 (1927): A. A. BENEDETTI-PICHLER.
(1l44) 73, 54 (1928): H. FISCHER.
(1l45) 74, 191 (1928): A. A. BENEDETTI-PICHLER, review.
(1l46) 75, 395 (1928): A. A. BENEDETTI-PICHLER, review.
(1947) 76, 216 (1929): A. A. BENEDETTI-PICHLER, review.
(1l48) 76, 443 (1929): A. A. BENEDETTI-PICHLER, review .
. (1l49) 77, 130 (1929): A. A. BENEDETTI-PICHLER. review.
(1150) 79, 94 (1929): P. RAY.
(ii51) 80, 247, (1930): H. MEISSNER.
(1l52) 82, 113 (1930): F. L. HAHN.
(1l53) 82, 241 (1930): H. K. ALBER, review.
(1154) 86, 69 (1931): A. A. BENEDETTI-PICHLER and F. SCHNEIDER.
(1l55) 86, 114 (1931): H. ALBER and MARIA v. RENZENBERG.
(1156) 89, 121 (1932): E. EEGRIWE.
(1l57) 90, 87 (1932): H. ALBER.
(1l58) 99, 402 (1934): I. M. KORENMAN.
(1159) 102, 102 (1935): R. W. FELDMANN.

Zeitschrift filr anorganische und allgemeine Ghemie

(1200) 147, 156 (1925): F. HABER and J. JAENICKE.
(1201) 175, 383 (1928): F. PANETH and K. PETERS.
 Literature 483

(1202) 199, 77 (1931): W. GEILMANN and K. BRUNGER.

(1203) 201, 347, 353 (1931): L. WOLF and W. JUNG.

Zeitsckritt tilr Elektrockemie

(12lO) 28, 89 (1922): A. GUNTHER-SCHULZE.

Zeitsckritt tilr pkysikaliscke Okemie

(1215) 76, 491 (1911): J. L. ANDRAE.

Zeitsckritt tilr den pkysikaliscken und ckemiscken Unlerrickt

(1220) 21,17 (1908): V.DvORAK.
(1221) 21, 281 (1908): A. WEINHOLD.

Zeitsckritt tilr wissensckattlicke Mikroskopie

(1225) 38, 1 (1921): K. SPANGENBERG.
(1226) 39, 316 (1923): H. BRUNSWIK.

Zentralblatt tilr M ineralogie, Geologie und Paliiontologie

(1230) A, 1929, 251: H. MORITZ.

Reports
(1240) Galco Technical Bulletin No. 770: Microscopical Techniques for the
Study of Dying: G. L. ROYER, C. MARESH, and ANNA M. HARDING.
(1241) Committee for the Study of New Analytical Reagents, International
Union of Pure and Applied Chemistry, Paris, May 1937.

Theses
(1250) E. FORCHE, Leipzig, 1938.
(1251) R. N. Boos, Master's Thesis, New York University, 1940.
(1252) L. BRANCONI, Master's Thesis, New York University, 1940.
(1253) K. D. FLEISCHER, Master's Thesis, New York University, 1940.
(1254) G. C. T. CHANG, Master's Thesis, Brooklyn College, 1961.

Unpublished Experiments
(1260) Dr. KUNZ ALFONS, Budapest, during a visit to EMICH'S Institute in 1926.
(1261) ANNE G. LOSCALZO at the Washington Square College of N. Y. U., 1940.
(1262) Dr. O. F. STEINBACH at Queens College, 1955.
(1263) SHAO-HsUN Ly, unpublished experiments.

Private Communications
(1270) Suggestion of Dr. GULBRAND LUNDE during a visit to EMlCH'S Institute
in summer 1926.
(1271) C. VAN BRUNT, General Electric Company, Schenectady, October 1943.
(1272) Dr. R. TEN EYCK SCHENK, Washington Square College of N. Y. U., 1950.
(1273) THOMAS P. SCHREIBER, Senior Research Physicist, Chemistry Depart-
ment, General Motors Technical Center, Warren, Michigan.
31·
484 Literature

Demonstrations of Microgram Technique by Dr. M. Cefola

(1280) Meeting of the Metropolitan Microchemical Society of New York.
Washington Square College, January 11, 1940.
(1281) Meeting of the New York Section of the American Chemical Society.
Hotel Pennsylvania, April 5, 1940.
(1282) In Service Training Lecture, Division of Laboratories, Department of
Hospitals. Washington Square College, December 20, 1940.

Meetings
(1283) MARy L. WILLARD, Eastern Analytical Symposium, New York, Nov. 16,
1962.

Addresses
(1290) Alfred Fritsch, Laborgeratebau, HauptstraBe 542, Idar-Oberstein,
West Germany.
(1291) Canal Industrial Corporation, Bethesda, Md.
(1292) Intercontinental Electronics Corporation, Minneola, N. Y.
 Subject Index
Absorption spectrophotometry 296 Barium, tests for 331
Acetate 357 Batch identification by schlieren 45
Air, supply of 97 Bead, dimensions and weight of 205
Akro technique 130 - tests 315
Alloys, analysis of 403 - -, list of 316
-, dissolution of 321, 403 - - , performance of 70, 191
Alkali metals, tests for 325-328 Becke line 42, 43
Alkaline earths, tests for 329-331 Beilstein test 81
Aluminum, tests for 352 Bertrand lens 49, 62
Ammonium, tests for 365 Beryllium, test for 329
Analysis see Identification Biological matter 240
Analyzer 49, 50 Bismuth, tests for 375
Angle, interfacial 247 Blanks 324
- , profile 248 Body color 244
-, - , list of 250 Boiling Point, determination on micro-
Anions, classification with barium gram scale 215
nitrate 424 - -, - - milligram scale 102
-, - with silver nitrate 425 - Range, determination on micro-
- , - with sulfuric acid 301, 309 gram scale 215
-, sensitivity of classification tests 307 Books, classified list 463
-, systematic testing for 422 Boron and borate, tests for 351
Anisotropic matter 47, 52 Bromate, tests for 392
Antimony, tests for 374 Bromide, tests for 391
Apparatus and scale of work 11 Bromine, tests for 391
- - surface forces 15 Bunsen flame, zones of 81
- - volatility 14
-, cleaning of, centigram scale 99 Cadmium, tests for 351
- , - - , general 68 Calcium, tests for 330
-, - -, microgram scale 209 Capillaries, examination of contents of
-, - -, milligram scale 124 162, 164
- , general, on centigram scale 82 -, preparation of 99
- , - , - gram scale 69 -, sealing of 161
-, -, - microgram scale 194-202 -, working in 159
-, - , - milligram scale 124 Capillary cone 198
- , selection of 228 - pipets, calibration of 125
Aperture 24 - -, preparation of 100
- diaphragm 25 - siphon 114
Arsenic, tests for 373 Carbides 356
Artifacts, identification of 242 Carbon monoxide 356
Ashing of tissue 242 -, tests for 299, 355
Autoradiography 254 Carbonate, tests for 356
Axial figures 61 Carius oxidation 166, 200
Azide, tests for 367 Centrifugal pipet 126
486 Subject Index

Centrifuge 84 Critical temperature 236, 238

- , use of 86 Cross-hair eyepiece 49
- tube, angle of taper 13 Cross hairs 32
- - , dimensions and scale of work 13 - - , checking of 52
Cerium, tests for 332 Crystal classes, criteria of 251
Cesium, tests for 328 Crystallographical optical analysis 292,
Chlorate, tests for 390 296
Chloride, tests for 388 Crystals, identification 243
Chlorine, tests for 300, 387 -, list of cubic substances 433
Chromium, from cleaning solution 229 -, - - hexagonal substances 435
- , tests for 336 Cyanate, test for 360
Citrate, tests for 359 Cyanide, tests for 359
Classification tests and general t.ech-
nique 268 Decantation on the centigram scale 86.
Cleaning, general 68, 229 87
- on centigram scale 99 - - - gram scale 72
- - microgram scale 209, 219 - - - microgram scale 211
- - milligram scale 124 - - - milligram scale 113
Cobalt, tests for 341 - - - submilligram scale 150, 151.
Color 54, 244 152, 160, 165
- observation, general 16 Density of liquids, determination of 262
- - under microscope 17, 175 - - solids, determination of 264
- of inorganic substances, table 432 Directions see Working directions
- , orders of 54 Distillation in capillaries 163, 170
Coloriscopic capillary 175 - on centigram scale 90, 101
Columbium see Niobium - - gram scale 74
Compensators 52, 53, 57 - - microgram scale 213
Condenser 33 - - milligram scale 118
- , use of 37 Double refraction 47
- rod, microgram scale 200 - - , sign of 62
Cone see Centrifuge tube Drills, drilling (micro) 232, 233
Confirmatory tests 6, 324-395 Drying in capillary 168, 171
- - by luminescence 193 - on slide 141, 152, 156-157
- - in beads 191 Dust particles, sampling 230
- - on centigram scale 98
- - - exchange resins 182 Electrodes 168
- - - gram scale 75 ' Electrolysis 187
- - - microgram scale 217, 218, - in the centifuge cone 115
221 i-on milligram scale 115
- - - milligram scale 123 - - microgram scale 186
- - - paper 128-136, 177ff. ' - - slide 186
- - - slide 136-149 Electrolytic slide 186
- - - textile fibers 183, 188, 191 Electron diffraction 292
- - - wires 187 I -probe 293
Conflagrating inorganic s~bstances, list ; Emulsions, breaking of 73
of 458 i Europium, tests for 333
Congo fiber 185 Evaporation on centigram scale 90>
Contraction pipet 166 - - fibers 189
Controls 324 - - gram scale 70
Copper, tests for 347 - - microgram scale 212
Critical solution temperature 239 i - - milligram scale 117
 Subject Index 487

Evaporation on paper 178, 179, 184 Fluoride, tests for 385

- - slide 140, 154, 155 Fluorine 385
- , prevention of 14, 15, 194 Forceps 202, 232
Experimentation see Technique Formate, tests for 357
Exploding substances, list of inorganic Fractional distillation on centigram
459 scale 91, 101
- -, - - organic 460 I - - - microgram scale 213
- -, test for 269 . - - - milligram scale 120
Explosives 272 Fractionation by melting in capillary
Extinction angle 56 171
- position 53 - - - on slide 151
-, types of 56 Front lenses 24
Extraction of solids on centigram scale Fusions on the centigram scale 83
88 - - - gram scale 70
- gram scale 73 - - - microgram scale 204
- milligram scale 115 - - milligram scale III
- paper 178, 179, 181 - - submilligram scale 152, 179
- - - submilligram scale 155, (footnote)
157 with potassium pyrosulfate 112, 323
- - liquids on centigram scale 88 - - sodium carbonate Ill, 323
- - - - gram scale 73 - - peroxide Ill, 324
- - - - milligram scale 116 - - sodium-potassium carbonate 323
Eyepiece 31
- micrometer, calibration of 37 Gallium, tests for 353
Gas, expelling of, on the microgram
Fat, test for 242 scale 213
Ferricyanide, tests for 361 -, - - , - - milligram scale 108,
Ferrocyanide, tests for 361 280
Ferromagnetism, test for 254 -, liberation and testing for, on centi-
Fibers 184 gram scale 96, 303
- , identification of 240, 242 -, - - - - , - gram scale 303
-, spot tests on 218 -, - - - - , - milligram scale
-, working on 178, 183, 188-191 123, 280, 308
Field diaphragm 25 -, - - - - , - submilligram
- of vision 36, 38 scale 158, 308
Filter paper, impregnation with - reaction cell 96
reagents 130 -, sensitivity of tests for 307
- -, working on 128 -, test for evolution of 309
Filtration in capillaries 169, 173, 174 Gaseous reagent, treating with, on
- - paper 178 centigram scale 83
of hot solutions in capillaries 169 - -, - -, - gram scale 69
- on centigram scale 86, 87 - -, - -, - microgram scale 210
gram scale 71 - -, - - , - milligram scale 106
- - milligram scale 113 Germanium, tests for 363
- - the slide 150, 151, 152 Gold, tests for 348
Final review of findings 428 Goniometer 248
Flame colorations, list of 314 Grinding 231
- test 313 Gypsum plate 53, 57
- - , performance of 79
Flash figures 62 Halogens, tests for 300, 385-395
Fiuorescence 246 Hard materials, list of 260
488 Subject Index

Hardness tests 258 Ignition on charcoal block 78, 275

Heating see also Ignition Illumination 34
- block 109 - , critical 27
element (flg scale) 201 - in microscopy 35
- in capillaries 162, 165, 166, 167 -, oblique 42, 43
- - cones under pressure 11 0 Immersion method 39, 43
on centigram scale 83 Inclusions, sampling of 232, 233
- gram scale 70 Indium, tests for 354
- microgram scale 211, 213 Interference color 54
- milligram scale 108 - - , determination of order 58
- up to 350 0 C 269 Iodate, tests for 394
Hexagonal crystals, criteria 252 Iodide, tests for 393
- - , list of substances 435 Iodine, tests for 300, 393
History of sample 234 Iridium, tests for 346
Hook of glass 304 Iron, as contamination on platinum 229
- - platinum 128 - , tests for 340
Hydrazine, tests for 366 Isotropic substances 52
Hydrazoic acid, tests for 367
Hydrofluoric acid, treatment with 322, Journal articles 472
323
Laboratory work see Technique
- - , - -, on milligram scale 112
Lanthanides 332
Hydrogen, test for 299, 304 L. C. see Limiting concentration
- peroxide, tests for 376 Lead, tests for 365
Hydroxylamine, tests for 366 L. I. see Limit of identification
Hypochlorite, tests for 389 Lignin, test for 242
Hypophosphate, tests for 370 Limit of identification 4, 5, 324
Hypophosphite, tests for 369 - - - of color tests 17
Limiting concentration 4, 5, 324
Identification, exercises 225-227 - proportions 4, 5, 324
- of artifacts 242 Lithium, tests for 325
- - crystals 243 Litmus fiber 184
- inorganic substances 272, 291, I Loop of glass 305
301 I _ _ platinum 127

- liquids 236 Loss by evaporation 14

organic substances 273, 294, 299 - spreading of drops 16
- organized (biological ) matter 240 - with residual liquid left in apparatus
- solids 244 15
-,record of 235 L. P. see Limiting proportions
- , systematic procedure of 228, 234, Luminescence 193, 246
268, 317, 428
Ignition above 300 0 C 274 Magnesium, tests for 329
below 350 0 C 269 Magnetism, test for 254
in air 77, 274, 283 Magnification 23
chlorine 285 Majors 2
closed tube 75, 274 Manganese, tests for 338
gas stream 276 Manipulators 195, 196, 231
hydrogen 284, 324 Measuring capillary (flg scale) 200
- sulfide 285 (volumes) on centigram scale 82
inert gas 75, 282 - - - gram scale 68
open tube 274 microgram scale 206
oxygen 283 - - - milligram scale 105, 125, 127
 Subject Index 489

Melting point 295 Mixing in capillaries 161, 165

- - , determination in capillary 170, on centigram scale 82
173 gram scale 69
- - of eutectic mixture 295 microgram scale 210
- - - inorganic substances 438 milligram scale 106
- - on slide 269 Moderately soluble inorganic sub-
- - under microscope 269 stances, list of 461
Mercury, tests for 351 Moist chamber 198
Metals, analysis of 403 - - , assembling of 202
- , dissolution of 321, 403 Molybdenum, tests for 337
Metaphosphate 371
Mica plate 59 New substances 1
Micro- see item following the prefix Nickel, tests for 342
micro Nicol prism 48
Microburner, preparation of 100 Nicols 48
Microcone 105, 106 - , crossed 50
- , observation in 124 Niobium, tests for 335
Micrometer eyepiece 31 Nitrate, tests for 368
- - , calibration of 37 Nitrite, tests for 368
- scale 39 Nitrogen, tests for 300, 365
Micromethods, definition 3 Nonmetallic substances, dissolution of
Micropipet 198 321
-, inserting into capillary 205 - - , test for anions 427
- , - - chamber 203
--, mounting 202 Objectives 32
Microprojection 26 Ocular see Eyepiece
Microscope 20 Odor, testing for 256
- , binocular 29 Optical emission spectrochemical
- , cleaning of 28 analysis 294
- , condenser 33, 37 Organic elemental analysis 299
- , field of vision of 36, 38 Organized matter, identification 240
- , focusing 30, 32 Osmium, tests for 345
- for microgram work 194 Oxalate, tests for 358
- , illumination 29, 30, 34, 35, 43 Oxidants, tests for 426
- , inspection of 28, 33, 50 Oxygen, tests for 375
- lamps 28 Ozone, tests for 376
- , monocular 29
- , petrographic 49 Palladium, tests for 344
-.' polarizing 49 Paper, impregnation with reagents 130
- , selection of 26 - , work in 128, 177
- slide see slide Per borate, tests for 302, 377, 382
- , stage 33 Percarbonate, tests for 302, 377, 382
- , - rotating 49 Perchlorate, tests for 390
- , - - , centering of 51 Periodate, tests for 395
- , testing of 28, 33, 50 Permanent mounts 65
- , working distance of 25, 38 Permanganate, test for 339
Microscopical preparations, preserving Peroxide, tests for 302, 376, 382
of 65 Persulfate, tests for 302, 377, 381
- tests, preserving of 65 Petrographic microscope 49, 50
Microsonde 293 pH, adjustment of, on centigram scale
Minors 3 82
490 Subject Index

pH, adjustment of, on gram scale 69 Radiation, test for 253

-, - - , - I?maller scales 19, 20 Rare earths, tests for 332,333,334,335,
Phosphate from cleaning solutions 229 336
- , tests for 371 Reagent container (flg scale) 199
Phosphide, tests for 369 Reagents, adding, on centigram scale 82
Phosphite, tests for 370 - , - , - gram scale 69
Phosphorus 369 - , - , - microgram scale 209
Photomicrography 26 - , - , - milligram scale 105
Pipet holder (flg scale) 197 - , -, - sub milligram scale 152
Plastic ware 229 -, care and use 68, 228
Platinum, tests for 346 i _, - - _ on milligram scale 104
- metals, glow test 343 I - , listing of 432

pL.C.5 Recrystallization, in capillary 161, 168,

Pleochroism 61 171
pL. 1. 5 - on slide 139, 151
Plunger control 197 Reductants, tests for 426
Polarized light 47 Refraction see Refractive index
Polarizer 48, 50 Refractive index, Becke test 42
Polarizing microscope 49 -, determination of 39
- - , testing and adjusting 50 -, - -, in anisotropic matter 57,
- - , data easily determined with 251 60
Polaroid 48 -, double variation method 41
Polars 48 -, oblique illumination 42
-, crossed 51 - - , standard liquids 40
Position of extinction 53 -, - solids 39
Potassium, tests for 326 - -, use in identification 260
Preliminary inspection 235 - - via schlieren observation 45
- treatment 6 Report on analysis 428
Preserving microscopic tests 65 Retardation plates 49, 52, 53, 57
Procedure of analysis, selection of 18 Rhenium, tests for 340
Profile angles, list of 250 Rhodium, tests for 344
- -, measuring 56 Rubidium, tests for 327
Projection 26 Ruthenium, tests for 343
Properties, effect of size upon 9
-, observation of 16 Sample, acid insoluble 322
Proteins, tests for 242 -, definition 3
Pseudomorphs 143, 144, 148 -, dissolution 320
Purification see Distillation, Fractional - , minimum size of 7
distillation, Fractional Melting, Re- Sampling 230
crystallization, Sublimation -, dust 230
Pyrophosphate 371 -, ferromagnetic particles 233
-, inclusions 232, 233
Qualitative analysis, definition 2 Saturation chamber 108
- - , limitations 1 Scale of work, indication of 3
- , method of 6 - - - , theoretical lower limit 8
- - , principal steps 6 Scandium, test for 331
- - , purpose and task 1 Scheme of separation, classical or H 2 S
Quantitative estimations on microgram 396
scale 212, 216 - - - of· NOYES and BRAY 404
- - - milligram, sc~le 14~, Schlieren and observation 44, 45
Quartz wedge 53, 57 Selenate, tests for 383
 Subject Index 491

Selenide, tests for 302, 306, 382 Storing of work on microgram scale
Selenite plate 53, 57 219
- , tests for 383 - - - - milligram scale 124
Selenium 382 Strontium, tests for 330
Sensitivity 4, 19 Study of chemical behavior 224
Separation by mechanical sorting 219 Sublimation from slide to slide 157
- , chemical 395 i - i n capillary 170
- , -, of anions 424 - on centigram scale 92
-, -, - arsenic and copper groups - - gram scale 74
149 - - milligram scale 122
- , - , - arsenic from antimony 222 Subliming inorganic substances, list of
-, - , - cations 396 457
-, -, on fibers 178, 183 Substances not described 1
- , -, - microgram scale 222 Sulfate, tests for 381
-, -, - milligram scale 149 Sulfide fiber 185
-, - , - paper 128, 177. -, tests for 302, 378
-, - , upon slide 153, 156 ! Sulfite, tests for 303, 379

Silhouette angle 56 ! Sulfur, tests for 300, 377

Silicate and silicon, tests 362 Surface color 245

Silver, tests for 347 - force, effects of 15
Slide, chemical work upon 150-161
- tests, examples 139-148
Tantalum, tests for 336
- -, general 136-138
- group, separation 410
- -, preserving of 65
Tartrate, tests for 358
Sodium, tests for 326 Technique and scale of work II
Solubility 285 - - surface forces 15
- , definition, inorganic 287 -, basic rules 67
-, - , organic 290 - of centigram scale 81
- , determination of, inorganic 288 - - gram scale 68
- , - -, organic 290 - - microgram scale 193
- of small particles 10, 287 - - milligram scale 104
-, rules for inorganic substances 288 - - picogram scale 9
Specific gravity see Density - - submilligram scale 128-193
- surface area 9 - - working with microscopic
Specificity 4 drops 9
Spindrying in capillary 171 Tellurium, tests for 302, 306, 383
Spot tests on centigram scale 98 Test solutions 430
- - - microgram scale 218 Testing chemical behavior 224
- - - (sub) milligram scale 128 to -, non destructive 244-268
136 Thallium, tests for 355
- -, sensitivity of 129 Thiocyanate, tests for 361
Stability on heating, test 269 Thionate, tests for 381
Stage micrometer 37 Thiosulfate, tests for 380
- of microscope 33 Thorium, tests for 335
- rotating 49 Tin, tests for 364
- - , centering of 51 Titanium, tests for 333
Standards of refractive index 39, 40 Traces, definition 3
Starch, test for 242 Transfer of liquids and slurries, centi-
Steam bath 109 gram scale 86, 87
Storing of work on centigram scale 99 - - - - -, gram scale 70, 71
- - - - gram scale 75 - - - - -, microgram scale ·208
492 Subject Index

Transfer of liquids and slurries, milli- Washing in capillaries 166, 167

gram scale 113, 114 on centigram scale 87
solids, centigram scale 82, 88 gram scale 71
- - , gram scale 69, 72 microgram scale 211
- - , microgram scale 203, 209, milligram scale 115
219 paper 179
- - , milligram scale 115 sub milligram scale 153
- - , submilligram scale 160 Water, tests for 376
Transition phenomena in polarized - bath (micro scale) 109
light 63 Weighing on centigram scale 82
- points above 300 0 C 274ff., 438 gram scale 68
- - below 350 0 C 269, 438 - - microgram scale 203
Tungsten see Wolfram 337 , - - milligram scale 104
Turmeric linen 188 Wolfram, tests for 337
Working directions, transposition from
scale to scale 8, 9
Ultrasonic jack hammer 233
distance (microscopical) 25
Unknowns, preparation of 431
technique see Technique
Uranium, tests for 338
X-ray diffraction 292
Vanadium, tests for 335 - emission spectrography 292, 296
Vibration directions, determination 55
- - of po lars 50 Ytterbium, tests for 333
- - - specimen 53, 55 Yttrium, tests for 332
Vibrator 73
Volatile substances and size of appa- Zinc, tests for 349
ratus 14 Zirconium, tests for 334

Manzsche Buchdruckerei, Wien IX