Biomimetic Sensing Methods and

Protocols Jessica E. Fitzgerald

Visit to download the full and correct content document:
https://textbookfull.com/product/biomimetic-sensing-methods-and-protocols-jessica-e-
fitzgerald/
More products digital (pdf, epub, mobi) instant
download maybe you interests ...

Targeting Enzymes for Pharmaceutical Development

Methods and Protocols Nikolaos E. Labrou

https://textbookfull.com/product/targeting-enzymes-for-
pharmaceutical-development-methods-and-protocols-nikolaos-e-
labrou/

Handbook of Fathers and Child Development Prenatal to

Preschool Hiram E. Fitzgerald

https://textbookfull.com/product/handbook-of-fathers-and-child-
development-prenatal-to-preschool-hiram-e-fitzgerald/

Handbook of Children and Prejudice Integrating Research

Practice and Policy Hiram E. Fitzgerald

https://textbookfull.com/product/handbook-of-children-and-
prejudice-integrating-research-practice-and-policy-hiram-e-
fitzgerald/

Zebrafish Methods and Protocols Koichi Kawakami

https://textbookfull.com/product/zebrafish-methods-and-protocols-
koichi-kawakami/
SNAREs: Methods and Protocols Rutilio Fratti

https://textbookfull.com/product/snares-methods-and-protocols-
rutilio-fratti/

Epitranscriptomics: Methods and Protocols Narendra

Wajapeyee

https://textbookfull.com/product/epitranscriptomics-methods-and-
protocols-narendra-wajapeyee/

Phytoplasmas: Methods and Protocols Rita Musetti

https://textbookfull.com/product/phytoplasmas-methods-and-
protocols-rita-musetti/

Metalloproteins: Methods and Protocols Yilin Hu

https://textbookfull.com/product/metalloproteins-methods-and-
protocols-yilin-hu/

Nanotoxicity: Methods and Protocols Qunwei Zhang

https://textbookfull.com/product/nanotoxicity-methods-and-
protocols-qunwei-zhang/
Methods in
Molecular Biology 2027

Jessica E. Fitzgerald
Hicham Fenniri Editors

Biomimetic
Sensing
Methods and Protocols
METHODS IN MOLECULAR BIOLOGY

Series Editor
John M. Walker
School of Life and Medical Sciences
University of Hertfordshire
Hatfield, Hertfordshire, UK

For further volumes:

http://www.springer.com/series/7651
For over 35 years, biological scientists have come to rely on the research protocols and
methodologies in the critically acclaimed Methods in Molecular Biology series. The series was
the first to introduce the step-by-step protocols approach that has become the standard in all
biomedical protocol publishing. Each protocol is provided in readily-reproducible step-by-
step fashion, opening with an introductory overview, a list of the materials and reagents
needed to complete the experiment, and followed by a detailed procedure that is supported
with a helpful notes section offering tips and tricks of the trade as well as troubleshooting
advice. These hallmark features were introduced by series editor Dr. John Walker and
constitute the key ingredient in each and every volume of the Methods in Molecular Biology
series. Tested and trusted, comprehensive and reliable, all protocols from the series are
indexed in PubMed.
 Biomimetic Sensing

Methods and Protocols

Edited by

Jessica E. Fitzgerald
Departments of Bioengineering and Chemical Engineering, Northeastern University, Boston, MA, USA

Hicham Fenniri
Departments of Chemical Engineering, Bioengineering, Chemistry and Chemical Biology, Northeastern
University, Boston, MA, USA
Editors
Jessica E. Fitzgerald Hicham Fenniri
Departments of Bioengineering Departments of Chemical Engineering, Bioengineering,
and Chemical Engineering Chemistry and Chemical Biology
Northeastern University Northeastern University
Boston, MA, USA Boston, MA, USA

ISSN 1064-3745 ISSN 1940-6029 (electronic)

Methods in Molecular Biology
ISBN 978-1-4939-9615-5 ISBN 978-1-4939-9616-2 (eBook)
https://doi.org/10.1007/978-1-4939-9616-2
© Springer Science+Business Media, LLC, part of Springer Nature 2019
This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part of the material is
concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction
on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation,
computer software, or by similar or dissimilar methodology now known or hereafter developed.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply,
even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations
and therefore free for general use.
The publisher, the authors, and the editors are safe to assume that the advice and information in this book are believed to
be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty,
express or implied, with respect to the material contained herein or for any errors or omissions that may have been made.
The publisher remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Cover illustration: Barcoded polymer-based cross-reactive sensor array response to a mixture of analytes. The different
colors illustrate different responses from the sensory elements.

This Humana imprint is published by the registered company Springer Science+Business Media, LLC, part of Springer
Nature.
The registered company address is: 233 Spring Street, New York, NY 10013, U.S.A.
Preface

The Gap in Technology: Fast, Facile, and Quantifiable Detection of Analytes in Vapor
and Liquid [1, 2]

Detection of analytes for vapor and liquid deconvolution has been performed in a variety of
fields, including nutrition, toxicology, biomedicine, and chemistry. The gold standard for
vapor sensing is gas chromatography-mass spectrometry (GC-MS), which is a quantitative
sample analysis that provides both the type and amount of analytes present in a sample,
usually in the form of volatile organic compounds (VOCs). While it is advantageous to know
which specific VOCs are present in a sample, GC-MS is not practical for widespread use in
vapor sensing because it requires both specialized, expensive equipment and highly trained
personnel for operation. In addition, some relevant data may be lost due to vapor
pre-concentration and sampling techniques. Similarly, the most common liquid sensing
techniques require sample labeling, or tagging, before performing an assay. This sensing
technique is limited because each analyte requires a specific label (making multiplexed
sensing difficult), the analytes in the sample must be known beforehand so that the correct
label can be selected, and nonspecific binding may take place, affecting the accuracy of the
measured analyte concentration. As sensing needs to continue to expand and develop, there
remains a need for methods that enable fast, facile, and accurate sensing at a low cost. To
meet these needs, many researchers have looked to the mammalian olfactory system as a
model for label-free, multiplexed sensing, developing platforms that are easy to use and
produce and that can be used in a wide variety of applications.

The Mammalian Olfactory System as an Optimal Model for Multiplexed Sensing

There are about 1000 genes that encode olfactory receptors (ORs), and each OR has
multiple sites for odorant binding, enabling the detection of more than one odorant for
each OR, a characteristic called cross-reactivity. Different combinations of activated recep-
tors make up unique signaling codes, or “fingerprints,” for specific odorants, making it
possible to distinguish between thousands. This sensing platform has inspired researchers
over the past several decades to develop sensing devices that are cross-reactive and accurate
and have multiplexing capabilities. These biomimetic devices are called “electronic/artificial
noses” (e-noses) or e-tongues to detect certain analytes present in both vapors and liquids,
respectively. E-devices have proved to be successful in a broad range of scientific and
engineering fields, providing cost-effective, minimally invasive (in the case of clinical use),
and highly accurate vapor and liquid component analysis. In this book, we highlight the
potential of e-device technology to serve as a successful platform for multiplexed sensing,
along with the methods for device fabrication, calibration, and assays in multiple applica-
tions. The subsequent sections describe e-device sensing platforms, explore their use, and
outline existing limitations and future directions in device development.

v
vi Preface

e-Device History and Applications

The first use of the term “electronic nose” was at a conference in 1987, and the first
conference dedicated specifically to artificial olfaction was in 1989. Gardner and Bartlett
originally defined an “electronic nose” as follows: “an instrument, which comprises of an
array of electronic-chemical sensors with partial specificity and an appropriate pattern recog-
nition system, capable of recognising simple or complex odours.” The first devices that utilized
this technology were comprised of sets of distinct active materials, each connected to its own
signal transduction channel. When passing an analyte vapor over the sensor array, activated
sensors would transmit an electrical signal to a processor, which would then alert the user of
the analytes present by cross-referencing the list of activated sensors with a database of
known analyte profiles. The earliest designs employed metal oxide semiconducting field
effect transistors (MOSFETs) as electrical sensors to detect gases such as NO2. This was
based on the principle that the conductivity of semiconductor metals changes upon variance
in the atmospheric gas surrounding the sensor. MOSFETs are usually constructed from a
SiO2 insulating layer, with a semiconductor metal deposited on top as the gate in the circuit.
A voltage is applied to maintain a constant current, and as the gas adsorbs onto the gate, the
conductance of the FET changes, thereby causing the voltage to change. e-devices are able
to differentiate analytes via “fingerprint” outputs; that is, each sample’s analyte profile
(comprised of a unique mixture of analyte types and relative concentrations) produces a
unique response pattern from the sensor array, enabling sample differentiation. New sample
fingerprints are compared to known sample fingerprints via data analysis (in most cases
multivariate data analysis such as principle component analysis), which enables pattern
recognition, clustering, and classification of the unknown analyte sample.
This fingerprint sensing method has been implemented in a number of fields, including
food science, quality control, drug testing, contamination detection, defense efforts (e.g.,
explosive detection), and medical diagnosis. Some of the most pioneering work has been
done in medical diagnostics via exhaled breath analysis. Many metabolites, or metabolic
by-products, have been identified and correlated with specific diseases; some of these can
even be a good indicator of disease progression. Many of these metabolites are volatile
organic compounds (VOCs), which are small molecules that enter exhaled breath through
gas exchange at the alveolar-capillary membrane of the respiratory tract. While the VOCs
produced in each disease are thought to be primarily from oxidative stress, the subsequent
effect of each disease on the body is unique and leads to the production of disease-specific
VOC profiles. A reliable, noninvasive device capable of detecting subtle molecular changes
and differences can be leveraged to implement this personalized medicine approach. The
e-devices included in this work have much potential for implementation in these fields and
others, as their sensitivity and specificity can be tuned toward specific analytes of interest.

e-Device Advances in Technological Development

Within the past several years, researchers have taken advantage of rapid technological
advances to expand and improve e-device sensing platforms and sensor materials, signal
transduction mechanisms, and data processing for pattern recognition and analyte identifi-
cation. Moreover, as more became known about the physiology of the mammalian olfactory
system, sensors further advanced to take advantage of these new findings, from
 Preface vii

incorporating sensors that were able to detect multiple analytes to the development of a
biomimetic flow chamber to enhance analyte detection at low levels. This allowed for a
greater number of analytes to be recognized using a smaller array of multi-selective sensors.
Within the last two decades, electronic tongues for liquid detection have also been devel-
oped, mimicking the olfactory signaling pathways.
The types of sensors employed for both vapor and liquid samples in these devices vary,
including gravimetric, or mass, electrochemical, and optical sensors, allowing characteriza-
tion of analytes based on mass, electrical properties (e.g., conductance impedance), and
electron/photon interactive properties, respectively. Gravimetric sensors are either piezo-
electric (PZ) crystals or microcantilevers, which have a specific resonant frequency. On
binding with an analyte, the resonant frequency of the sensor drops in proportion to the
added mass, due to either viscoelastic or gravimetric effects. Electrochemical e-devices are
comprised of an electronic circuit connected to a network of sensory materials—most
commonly conductive polymers or metal oxides—that provide an electrical response on
binding with a specific known analyte. This response is characterized by monitoring sensor
conductivity, resistivity, or voltage change during vapor exposure. Finally, optical sensors
work by displaying a shift in the emission or absorption of different types of electromagnetic
radiation on binding with a desired analyte. There are two popular means of optical
detection: fluorescent sensors, which fluoresce upon analyte binding, and colorimetric
sensors, which display a visible color change upon analyte binding.
Although gravimetric sensors have proved to be successful, there are many limitations
with this device setup. Inaccuracies due to subtle changes in surface coating, humidity, or
temperature necessitate frequent calibrations, which is unfortunately delicate and time-
consuming. e-devices that employ optical and electrochemical sensors have shown much
promise as they provide an easier and more cost-effective way of identifying analytes in vapor
while maintaining accuracy. Optical sensors offer significant benefits compared with those
mentioned above since they can provide multiple complex data types simultaneously,
including changes in intensity, fluorescence lifetime, wavelength, and spectral shape. This
approach increases the ratio of recognizable analytes to the number of sensors used. This
work mainly focuses on optical (Chapters 1–10) and electrochemical (Chapters 11–13)
methods of sensing, as these are at the forefront of e-device technology, being more stable
and reliable than their gravimetric counterparts. We have also included a chapter on a
cutting-edge mechanochemical sensing method using folded DNA origami structures
(Chapter 14) that have been demonstrated to have a limit of detection down to the single
molecule level. Finally, we highlight here some cutting-edge methods to optimize e-device
data and technology via drift correction and calibration (Chapter 15) and computer model-
ing of sensor output for material optimization (Chapter 16).

Optical Sensor Arrays for e-Devices

Optical sensors in e-device systems have shown much promise to provide a facile, cost-
effective, and accurate way of identifying analytes in vapor and liquid samples. Optical
sensors display a shift in emission or absorption of different types of radiations upon analyte
binding. The two most popular means of detection are spectroscopic and colorimetric
sensing. In this work, we highlight the methods and applications of both means of detec-
tion. The spectroscopic methods employ Raman spectroscopy (Chapter 1), interferometry
(Chapter 2), mass spectrometry (Chapter 3), fluorescence microscopy (Chapter 5), and
viii Preface

surface plasmon resonance (Chapters 8 and 10). The colorimetric sensors herein are pro-
duced in several ways, including microchip fabrication and photolithography (Chapters 4, 6,
and 7). As mentioned previously, optical sensors offer significant benefits compared with
gravimetric sensors; they can provide multiple complex data types simultaneously, including
changes in intensity, fluorescence and colorimetric lifetime, wavelength, and spectral shape.
This increases the ratio of recognizable analytes to the number of sensors used. In addition,
microsphere optical arrays such as those developed by Walt et al. and presented in Chapter 1
provide an advantage over other multisensor systems because billions of beads that produce
an identical response can be made simultaneously, compared with many sensors for which
the fabrication process is tedious. Each type of bead has a distinct, intrinsic response to the
samples presented, which eliminates the need for additional encoding for bead
identification.

Electrochemical Sensor Arrays for e-Devices

Electrochemical sensing e-devices use an electronic circuit connected to a network of

sensory materials that provide an electrical response upon binding with a specific known
analyte. While the most common electrochemical sensing platform is the MOSFET devices,
FET sensors have also been developed that incorporate organic material such as
DNA-decorated graphene FETs (Chapter 13). Another popular, recent sensory material is
conductive polymer (Chapters 11 and 12). Chapter 11 describes a method for layer-by-layer
deposition of conducting polymers in a microfluidic channel as an electronic tongue. The
conducting polymers described in Chapter 12 are electropolymerized in the presence of the
target molecule or template which is then removed after polymerization to create molecu-
larly imprinted polymers. In both cases, the polymer sensors are then placed in an electrical
circuit and act as resistors, reflecting a decrease in their conductance (or an increase in
impedance) upon binding with the analytes in the sample. This decrease in conductance is
most likely due to polymer swelling upon analyte binding—as the polymer swells, gaps
between polymer chains increase, lowering conductivity.

Remaining Challenges and Future Outlook for e-Device Implementation

As e-device implementation continues to grow in breadth, there are certain limiting factors
that must be addressed. First, there remains a lack of standards for sample collection, both
environmental (ambient air, water) and medical (exhaled breath, saliva). When developing a
sampling method, it is important to optimize the collection, preparation, and storage
method to maximize analyte detection without denaturing or altering the chemical profile
of the sample. For example, collecting alveolar breath, the second phase of the breathing
cycle, requires a sampling method that minimizes VOC interference from ambient air while
capturing the alveolar air from a patient who is breathing steadily at a set velocity. The
sample storage material and time of storage also affect analyte recovery. Even after obtaining
an ideal sample, e-device performance accuracy may be limited by extrinsic factors such as
humidity and temperature or intrinsic factors such as sensor drift and instrumentation
errors. Additionally, e-device analyte fingerprint analysis via pattern recognition requires
complex data analysis, which currently limits the widespread implementation of these
devices.
 Preface ix

Finally, though preliminary studies with e-devices have been largely successful, their
reproducibility is limited because methods must be optimized de novo for each specific
application. Standards need to be developed from statistical analysis of device performance
and should include thresholds for success in areas such as response reproducibility, specific-
ity, and sensitivity. In developing these standards, it is also important to consider the
ultimate goal of the device. For example, if the goal is simply to diagnose and classify a
sample, such as one that correlates with a specific disease, selectivity is more important than
sensitivity; however, if the goal is to monitor analyte profile change over time, such as with
disease progression, sensitivity to slight variations in analyte concentrations is of great
importance.
Though these limitations currently serve as a bottleneck for widespread e-device imple-
mentation, researchers continue to work diligently to develop methods that will circumvent
and overcome them. For example, Chapter 15 includes a detailed method for device
calibration that can be applied to a wide variety of sensing mechanisms and e-devices.
Additionally, complementary methods, such as computer modeling of analyte-sensor inter-
action presented in Chapter 16, serve as a way to better predict sensor response to specific
analytes. This can then be leveraged to produce devices that are highly tuned toward a
specific analyte or group of analytes, i.e., much improved sensitivity and selectivity. As the
technology for e-device sensing continues to trend toward portable, accurate, and easy-to-
use platforms, they have great potential to be implemented wherever analyte detection is
required. Indeed, they may be able to reduce the need for highly specialized, expensive
equipment and personnel. For developing countries in particular, e-devices that are simul-
taneously inexpensive may soon be able to take the place of highly specialized equipment in
fields such as defense efforts and explosive detection, water and food contamination, and
personalized medicine.

Boston, MA, USA Jessica E. Fitzgerald

Hicham Fenniri

References

1. Fitzgerald JE, Bui ETH, Simon NM, Fenniri H (2016) Trends Biotechnol. 1
2. Fitzgerald JE, Fenniri H (2016) RSC Adv. 6(84): 80468
Contents

Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . v
Contributors. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xiii

1 Cross-Reactive, Self-Encoded Polymer Film Arrays

for Sensor Applications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
Jessica E. Fitzgerald and Hicham Fenniri
2 Interferometric Reflectance Imaging Sensor (IRIS)
for Molecular Kinetics with a Low-Cost, Disposable Fluidic Cartridge. . . . . . . . . 15
James W. Needham, Nese Lortlar U € nlü, Celalettin Yurdakul,
€
and M. Selim U nlü
3 An Olfactory Sensor Array for Predicting Chemical Odor
Characteristics from Mass Spectra with Deep Learning . . . . . . . . . . . . . . . . . . . . . . 29
Yuji Nozaki and Takamichi Nakamoto
4 A Photochromic Sensor Microchip for High-Performance
Multiplex Metal Ion Detection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49
Meng Qin, Fengyu Li, and Yanlin Song
5 Contact Printing of a Quantum Dot and Polymer
Cross-Reactive Array Sensor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61
Vincent P. Schnee and Collin J. Bright
6 Colorimetric Sensor Array Based on Amino Acid-Modified
Gold Nanoparticles for Toxic Metal Ion Detection in Water. . . . . . . . . . . . . . . . . . 75
Gülsu Şener and Adil Denizli
7 Identification of Several Toxic Metal Ions Using a Colorimetric
Sensor Array. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81
Gülsu Şener and Adil Denizli
8 Real-Time Sensing with Patterned Plasmonic Substrates
and a Compact Imager Chip . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87
Spencer T. Seiler, Isabel S. Rich, and Nathan C. Lindquist
9 Inkjet-Printed Colorimetric Paper-Based Gas Sensor Arrays
for the Discrimination of Volatile Primary Amines
with Amine-Responsive Dye-Encapsulating Polymer Nanoparticles . . . . . . . . . . . 101
Hiroyuki Shibata, Yuma Ikeda, and Daniel Citterio
10 Label-Free Nanoplasmonic Biosensing of Cancer Biomarkers
for Clinical Diagnosis. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 115
Alejandro Portela, Enelia C. Peláez, Olalla Calvo-Lozano,
Mari C. Estévez, and Laura M. Lechuga
11 A Microfluidic E-Tongue System Using Layer-by-Layer
Films Deposited onto Interdigitated Electrodes Inside
a Polydimethylsiloxane Microchannel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 141
Maria L. Braunger, Cristiane M. Daikuzono,
and Antonio Riul Jr

xi
xii Contents

12 Molecularly Imprinted Polymer Thin-Film Electrochemical Sensors. . . . . . . . . . . 151

Vera L. V. Granado, M. Teresa S. R. Gomes, and Alisa Rudnitskaya
13 Scalable Arrays of Chemical Vapor Sensors Based
on DNA-Decorated Graphene. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 163
Jinglei Ping and A. T. Charlie Johnson
14 Single-Molecule Mechanochemical Sensing Using DNA
Origami Nanostructures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 171
Sagun Jonchhe and Hanbin Mao
15 Response Standardization for Drift Correction and Multivariate
Calibration Transfer in “Electronic Tongue” Studies . . . . . . . . . . . . . . . . . . . . . . . . 181
Vitaly Panchuk, Valentin Semenov, Larisa Lvova, Andrey Legin,
and Dmitry Kirsanov
16 Computational Modeling for Biomimetic Sensors. . . . . . . . . . . . . . . . . . . . . . . . . . . 195
Icell M. Sharafeldin, Jessica E. Fitzgerald, Hicham Fenniri,
and Nageh K. Allam

Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 211
Contributors

NAGEH K. ALLAM  Energy Materials Laboratory, School of Sciences and Engineering, The
American University in Cairo, New Cairo, Egypt
MARIA L. BRAUNGER  Department of Applied Physics, “Gleb Wataghin” Institute of Physics
(IFGW), University of Campinas—UNICAMP, Campinas, SP, Brazil
COLLIN J. BRIGHT  U.S Army Combat Capabilities Development Command, C5ISR Center,
Fort Belvoir, VA, USA
OLALLA CALVO-LOZANO  Nanobiosensors and Bioanalytical Applications Group
(NanoB2A), Catalan Institute of Nanoscience and Nanotechnology (ICN2), CSIC and
BIST, Barcelona, Spain; Networking Center on Bioengineering, Biomaterials and
Nanomedicine (CIBER-BBN), Barcelona, Spain
DANIEL CITTERIO  Faculty of Science and Technology, Department of Applied Chemistry,
Keio University, Yokohama, Japan
CRISTIANE M. DAIKUZONO  Centro de Ciências e Tecnologias para Sustentabilidade (CCTS),
Universidade Federal de São Carlos—UFSCar, Sorocaba, SP, Brazil
ADIL DENIZLI  Department of Chemistry, Hacettepe University, Ankara, Turkey
MARI C. ESTÉVEZ  Nanobiosensors and Bioanalytical Applications Group (NanoB2A),
Catalan Institute of Nanoscience and Nanotechnology (ICN2), CSIC and BIST,
Barcelona, Spain; Networking Center on Bioengineering, Biomaterials and Nanomedicine
(CIBER-BBN), Barcelona, Spain
HICHAM FENNIRI  Departments of Chemical Engineering, Bioengineering, Chemistry and
Chemical Biology, Northeastern University, Boston, MA, USA; Department of
Bioengineering, Northeastern University, Boston, MA, USA
JESSICA E. FITZGERALD  Department of Bioengineering and Department on Chemical
Engineering, Northeastern University, Boston, MA, USA
M. TERESA S. R. GOMES  Chemistry Department, University of Aveiro, Aveiro, Portugal;
CESAM, University of Aveiro, Aveiro, Portugal
VERA L. V. GRANADO  Chemistry Department, University of Aveiro, Aveiro, Portugal
YUMA IKEDA  Faculty of Science and Technology, Department of Applied Chemistry, Keio
University, Yokohama, Japan
A. T. CHARLIE JOHNSON  Department of Physics and Astronomy, University of Pennsylvania,
Philadelphia, PA, USA
SAGUN JONCHHE  Department of Chemistry and Biochemistry, Kent State University, Kent,
OH, USA
DMITRY KIRSANOV  Institute of Chemistry, St. Petersburg State University, St. Petersburg,
Russia; Laboratory of Artificial Sensory Systems, ITMO University, St. Petersburg, Russia
LAURA M. LECHUGA  Nanobiosensors and Bioanalytical Applications Group (NanoB2A),
Catalan Institute of Nanoscience and Nanotechnology (ICN2), CSIC and BIST,
Barcelona, Spain; Networking Center on Bioengineering, Biomaterials and Nanomedicine
(CIBER-BBN), Barcelona, Spain
ANDREY LEGIN  Institute of Chemistry, St. Petersburg State University, St. Petersburg,
Russia; Laboratory of Artificial Sensory Systems, ITMO University, St. Petersburg, Russia
FENGYU LI  Key Laboratory of Green Printing, Institute of Chemistry, Chinese Academy of
Sciences (ICCAS), Beijing Engineering Research Center of Nanomaterials for Green

xiii
xiv Contributors

Printing Technology, Beijing National Laboratory for Molecular Sciences (BNLMS),

Beijing, P. R. China
NATHAN C. LINDQUIST  Department of Physics and Engineering, Bethel University, St. Paul,
MN, USA
LARISA LVOVA  Laboratory of Artificial Sensory Systems, ITMO University, St. Petersburg,
Russia; Department of Chemical Science and Technologies, University “Tor Vergata”,
Rome, Italy
HANBIN MAO  Department of Chemistry and Biochemistry, Kent State University, Kent,
OH, USA
TAKAMICHI NAKAMOTO  Institute of Innovative Research, Tokyo Institute of Technology,
Yokohama, Kanagawa, Japan
JAMES W. NEEDHAM  InBios International Inc., Seattle, WA, USA
YUJI NOZAKI  Institute of Innovative Research, Tokyo Institute of Technology, Yokohama,
Kanagawa, Japan
VITALY PANCHUK  Institute of Chemistry, St. Petersburg State University, St. Petersburg,
Russia; Laboratory of Artificial Sensory Systems, ITMO University, St. Petersburg, Russia
ENELIA C. PELÁEZ  Nanobiosensors and Bioanalytical Applications Group (NanoB2A),
Catalan Institute of Nanoscience and Nanotechnology (ICN2), CSIC and BIST,
Barcelona, Spain; Networking Center on Bioengineering, Biomaterials and Nanomedicine
(CIBER-BBN), Barcelona, Spain
JINGLEI PING  Department of Mechanical and Industrial Engineering, University of
Massachusetts Amherst, Amherst, MA, USA
ALEJANDRO PORTELA  Nanobiosensors and Bioanalytical Applications Group (NanoB2A),
Catalan Institute of Nanoscience and Nanotechnology (ICN2), CSIC and BIST,
Barcelona, Spain; Networking Center on Bioengineering, Biomaterials and Nanomedicine
(CIBER-BBN), Barcelona, Spain
MENG QIN  Key Laboratory of Green Printing, Institute of Chemistry, Chinese Academy of
Sciences (ICCAS), Beijing Engineering Research Center of Nanomaterials for Green
Printing Technology, Beijing National Laboratory for Molecular Sciences (BNLMS),
Beijing, P. R. China
ISABEL S. RICH  Department of Physics and Engineering, Bethel University, St. Paul, MN,
USA
ANTONIO RIUL JR  Department of Applied Physics, “Gleb Wataghin” Institute of Physics
(IFGW), University of Campinas—UNICAMP, Campinas, SP, Brazil
ALISA RUDNITSKAYA  Chemistry Department, University of Aveiro, Aveiro, Portugal;
CESAM, University of Aveiro, Aveiro, Portugal
VINCENT P. SCHNEE  U.S Army Combat Capabilities Development Command, C5ISR
Center, Fort Belvoir, VA, USA
SPENCER T. SEILER  Department of Physics and Engineering, Bethel University, St. Paul,
MN, USA
VALENTIN SEMENOV  Institute of Chemistry, St. Petersburg State University, St. Petersburg,
Russia
GÜLSU ŞENER  Department of Chemistry, Hacettepe University, Ankara, Turkey
ICELL M. SHARAFELDIN  Energy Materials Laboratory, School of Sciences and Engineering,
The American University in Cairo, New Cairo, Egypt
HIROYUKI SHIBATA  Faculty of Science and Technology, Department of Applied Chemistry,
Keio University, Yokohama, Japan
 Contributors xv

YANLIN SONG  Key Laboratory of Green Printing, Institute of Chemistry, Chinese Academy
of Sciences (ICCAS), Beijing Engineering Research Center of Nanomaterials for Green
Printing Technology, Beijing National Laboratory for Molecular Sciences (BNLMS),
Beijing, P. R. China
M. SELIM U€ NLÜ  Department of Biomedical Engineering, Boston University, Boston, MA,
USA; Department of Electrical and Computer Engineering, Boston University, Boston,
MA, USA
NESE LORTLAR U € NLÜ  Department of Biomedical Engineering, Boston University, Boston,
MA, USA
CELALETTIN YURDAKUL  Department of Electrical and Computer Engineering, Boston
University, Boston, MA, USA
 Chapter 1

Cross-Reactive, Self-Encoded Polymer Film Arrays

for Sensor Applications
Jessica E. Fitzgerald and Hicham Fenniri

Abstract
The development of chemical sensors continues to be an active area of research, especially the development
of a practical electronic nose. Here, we present a spectroscopic chemical sensor based on an array of 64 self-
encoded polymer films deposited on a microfabricated silicon substrate. The polymer arrays were analyzed
by FTIR and Raman spectroscopy before and after exposure to a series of organic volatiles to monitor
changes in their vibrational fingerprints. We show here that the spectroscopic changes of self-encoded
polymer films can be used to distinguish between volatile organic analytes. Changes induced in the sensor
arrays by the analyte vapor were denoted by a spectroscopic response of the self-encoded polymer sensors
and transformed into a response pattern by multivariate data analysis using partial least squares regression.
The results indicated that the polymer sensors provide a unique and reproducible pattern for each analyte
vapor and can potentially be used in the fabrication of a novel electronic nose device.

Key words Barcoded polymer, Sensor array, Raman spectroscopy, Multivariate data analysis

1 Introduction

Biomimetic engineering is the application of biological principles to

the design of artificial devices or systems. For many years, scientists
and engineers have recognized the power of naturally occurring
systems and their ability to guide the development of certain tech-
nologies. One example of the biomimetic approach is a device
known as the electronic or artificial nose (e-nose) [1]. Gardner
and Bartlett define an e-nose as “an instrument which comprises
an array of electronic chemical sensors with partial specificity and an
appropriate pattern recognition system, capable of recognizing
simple or complex odors” [2]. Applications of e-noses include
food quality control [3], pollution monitoring [4], medical diag-
nosis [5], and landmine detection [6], among others. In this rapidly
growing field, arrays of semi-selective chemical sensors are the main
component of the devices. These arrays are coupled to pattern
recognition software, paralleling the biological olfactory system in

Jessica E. Fitzgerald and Hicham Fenniri (eds.), Biomimetic Sensing: Methods and Protocols, Methods in Molecular Biology,
vol. 2027, https://doi.org/10.1007/978-1-4939-9616-2_1, © Springer Science+Business Media, LLC, part of Springer Nature 2019

1
2 Jessica E. Fitzgerald and Hicham Fenniri

which semi-selective olfactory receptors are combined with higher

order neural processing [7]. It is thought that there are millions of
receptor cells made up of approximately 1000 different receptor
types in the mammalian olfactory epithelium; while none of the
receptors are thought to be particularly sensitive to one specific
analyte, it is likely that their combined signal leads to the highly
sensitive and discriminative sense of smell [8, 9], Similarly, in e-nose
system architecture, no individual detector is highly selective
toward an individual analyte, as would be the case in the traditional
“lock and key” approach to chemical sensing. Instead, each detec-
tor responds to many analytes with varying degrees of intensity,
creating a unique response pattern for each analyte, a principle
known as cross-reactivity. The resulting odor signature from the
array is used to classify, and in some cases quantify, the analyte of
concern.
E-noses are primarily classified by the sensor-transduction
mechanism. E-nose transducers reported to date include semicon-
ducting metal oxide [10], conducting polymer films [11], acoustic
wave devices [12], electrochemical systems [13], carbon-black
loaded polymer film chemoresistors [14], and optical transducers
with immobilized dyes such as Nile Red or various metalloporphyr-
ins [15, 16]. Among these approaches, most array sensors employ-
ing polymers have produced high selectivity, taking advantage of
properties such as polarity, swelling, conductivity, and sorption.
Regardless of the transduction mechanism of the sensor array, a
larger number of unique sensors increases the quantity and accuracy
of the data gathered, thus producing a more complex and specific
pattern for improved analyte identification and classification.
The research and development of sensor arrays often involves
the measurement and analysis of a large number of samples. This
can be very laborious and time consuming because there are many
variables that influence the performance, sensitivity, and precision
of the sensors. Therefore, there is currently a significant interest in
the development of sensor arrays with high selectivity, sensitivity,
reproducibility, and high-throughput capability for applications in
disease diagnosis from patients’ vapor/breath analysis.
Our group has recently reported on a new class of resins
prepared from spectroscopically active styrene monomers, the com-
bination of which produces polymers wherein unique vibrational
fingerprints are associated with each polymer. The spectrum from
each polymer can then be converted into a barcode in which the
position of each bar matches a peak wavenumber in the spectrum.
We have previously demonstrated that each barcoded resin (BCR)
can be selectively recognized and classified by the unique self-
encoded spectra that are produced from the BCRs’ composition.
BCRs have previously been proposed for use in the deconvolution
strategies of resin-supported combinatorial libraries [17].
 Chapter 2

Interferometric Reflectance Imaging Sensor (IRIS)

for Molecular Kinetics with a Low-Cost, Disposable
Fluidic Cartridge
€
James W. Needham, Nese Lortlar Unlü, Celalettin Yurdakul,
€
and M. Selim Unlü

Abstract
The determination of kinetic information and appropriate binding pairs is fundamental to the proper
optimization and selection of ligands used in immunoassays, diagnostics, and therapeutics. However, the
ability to estimate such parameters in a multiplexed and inexpensive format remains difficult and modifica-
tion of the ligand is often necessary. Here, we detail the methods and materials necessary to evaluate
hundreds of unlabeled ligands simultaneously using the interferometric reflectance imaging sensor (IRIS).
The incorporation of a low-cost fluidic cartridge that integrates on the top of the sensor simplifies reagent
handling considerably.

Key words Interferometric reflectance imaging sensor (IRIS), Multiplexed kinetics, Label-free bind-
ing, Dengue NS1

1 Introduction

The measurement and understanding of the kinetic data involved in

ligand-analyte binding events (e.g., the typical antibody-antigen
complex) are foundational to immunoassay development and opti-
mization [1]. A variety of methods exist for the determination of
such kinetic information, including surface plasmon resonance
(SPR) and biolayer interferometry (BLI), among others
[2, 3]. However, most methods are limited in scalability (by cost
or technology) with only one or a handful of ligands that may be
evaluated at a given time, or the necessity for labeling of the ligand
(e.g., biotinylation) in order to perform the assay. Additionally,
evaporation may cause some artifacts with data collection using
BLI [4]. Here, we describe the methods to use the interferometric
reflectance imaging sensor (IRIS) in order to screen hundreds and
potentially thousands of ligands simultaneously using an

Jessica E. Fitzgerald and Hicham Fenniri (eds.), Biomimetic Sensing: Methods and Protocols, Methods in Molecular Biology,
vol. 2027, https://doi.org/10.1007/978-1-4939-9616-2_2, © Springer Science+Business Media, LLC, part of Springer Nature 2019

15
Another random document with
no related content on Scribd:
 NICKEL STEEL.
The addition of a few per cent of nickel to mild steel adds greatly
to its strength—so much so that nickel steel is now world-renowned
as used in armor-plate for navy vessels, and for great guns. Recent
reports from the ordnance bureaus indicate that it will also be of
great use in the barrels of small arms, by means of which they may
be made lighter, and still of sufficient strength. Nickel is so expensive
and it adds so much to the cost of steel that its use for ordinary
structural purposes, bridges, etc., has not been found to be
economical.
Some years ago careful experiments were made with nickel alloy
in a fine grade of high-carbon tool-steel to find out whether such
steel would be improved as much as are the mild steels.
In such case the expense would not count, for if the best steel
can be made better there are many users who would gladly pay a
higher price for a better service.
The results were not encouraging. The high-carbon nickel steel
was not as strong as the same quality of steel without nickel; the
mixture seemed to be imperfect, containing little dark specks,
supposed to be carbon thrown into the graphitic state. The steel did
not refine as well and was not as strong as the carbon steel.
All of this applies to high-carbon tool-steel, hardened and
tempered; no tests were made of the steel unhardened, for they
would have been of no practical use.

ALUMINUM STEEL.
When a heat of steel is boiling violently, is wild, and unfit to be
poured, the addition of a minute quantity of aluminum will have the
effect of quieting it quickly. Half an ounce to an ounce of aluminum to
a ton of steel will be enough usually, and for this purpose aluminum
has become useful to steel-makers. If a little too much aluminum be
added, the ingots will pipe from end to end; therefore the use of
aluminum is restricted to small quantities. Experiments have shown
that a considerable percentage of aluminum adds no good properties
to steel; therefore aluminum steel so called may be treated later
under a different heading.
 IV.
CARBON.

Of all of the abundant elements of nature carbon is presented in

the greatest variety of forms, and admits of the greatest number of
useful applications.
In the form of the diamond it is the hardest of substances, and is
the base used in determining the comparative hardness of all others.
In the form of graphite it is soft and smooth, and is one of the
best and most durable of lubricants.
In the form of soot it is probably the softest of solids.
In the form of coal it is the one great and abundant fuel of the
world, while as graphite again it is one of the best of refractory
materials.
Hard, soft, highly combustible, almost infusible, refractory, it lends
itself to the greatest variety of useful applications. To the iron- and
steel-maker or worker it is simply indispensable; as charcoal or coke
it is the fuel of the smelter; as gas, either carbon monoxide or as a
hydrocarbon, it is the cheapest and most manageable fuel for
melting and for all operations requiring heat.
As graphite, plumbago, mixed with a little fire-clay as a binder, it
is the best material for crucibles in which to melt metals; as soot it
forms the best coating for moulds into which metals are to be cast.
Durable beyond almost any other substance, it would make the
very best paint for metal structures if there were any known way to
make it adhere.

CARBON IN IRON.
 Carbon may be introduced into iron in any quantity from a few
hundredths of one per cent as usually found in wrought iron, and in
what is known as dead-soft steel, up to about four per cent as found
in cast iron. By the addition of manganese as high as six or seven
per cent of carbon has been introduced into iron. Carbon does not
form a true alloy with iron, neither does it form any stable chemical
compound. Its condition in iron seems to be as variable as it is in
nature, and sometimes it has been supposed to be as capricious as
it is variable. It is hoped that the reader of these pages will find that
there is no caprice about it, that its action is governed by as sure
laws as any in nature, and that certain results may be predicated
upon any treatment to which it is subjected.
The theories of its actions are as numerous and variable as are
the actions themselves, and they will be treated in a separate
chapter, this chapter being confined to a statement of known facts.
As stated in Chap. I, carbon may be introduced into iron by
heating carbon and iron in contact when air is excluded; and,
conversely, carbon is burned out of cast iron by the Bessemer and
open-hearth processes to reduce the cast iron to cast steel.
In the crucible any quantity of carbon may be obtained in steel by
melting a mixture of high blister-steel and wrought iron, or cast iron
and wrought iron, or by charging with wrought iron the necessary
quantity of coke or charcoal. When using plumbago crucibles, the
iron takes up some carbon from the crucible; also the spiegel-eisen
or ferro-manganese used adds some carbon; and for these two
sources of carbon the melter allows when he decides upon the
quantity of charcoal needed.
Results from crucible-melting are not strictly uniform; even if
every charge were weighed in a chemical balance accurately the
product would not be uniform, because one crucible gives off more
carbon than another; in one crucible a little more charcoal may be
burned and escape as gas than in another; and most variable of all,
unless the charcoal has been dried thoroughly, is the quantity of
moisture in the charcoal. One charge of charcoal may be dry, and
the next may contain as much as twenty-five per cent of moisture;
obviously equal weights in such a case would not give equal
quantities of carbon to the steel.
In crucible steel this is no disadvantage; a skilful mixer will get
from 75% to 90% of his ingots of the desired temper; the other ingots
will all be in demand for other uses, and as he can separate them all
with absolute certainty by ocular inspection, as described before, he
labors under no fear of bad results.
In the Bessemer process it is usual to burn out all of the carbon
and then to add the required amount in the spiegel; for structural
steels and for rails this method is satisfactory. For high steel—from
fifty to a hundred or more carbon—the spiegel method does not
answer so well, because it increases the quantity of manganese to
too great an amount; higher carbon is then sometimes put in by the
addition of a given quantity of pure pig iron previously melted, or by
putting coke in the ladle, but this is very uncertain on account of the
tendency of the coke to float, and be dissipated as a gas instead of
entering the steel.
The Darby method is to place in the way of the stream of steel as
it is poured from the vessel to the ladle a refractory-lined, funnel-
shaped vessel filled with finely divided, but not powdered, coke. As
the stream rushes through the coke it absorbs carbon with great
rapidity, and it is asserted that the currents and eddies formed in the
ladle by the rush of the stream cause an even distribution of carbon.
That carbon will be taken up in this way is certain; that a required
amount, evenly distributed, can be obtained is not so certain.
In the acid open-hearth as in the Bessemer process for milder
steels it is usual to burn the carbon out almost entirely, and then to
add the desired amount with the spiegel. Higher carbon may be
obtained by the addition of pure pig iron, or by using carbon bricks
pasted together with tar and weighted with iron turnings; these bricks
may be pushed under the surface in different parts of the bath, and
in this way the carbon can be distributed pretty evenly. In good
practice now the melt is stopped at the carbon desired with great
success, thus saving time and expense. In the basic open-hearth the
melter, by the use of a little care and good judgment, stops his melt
at the required carbon, and so avoids any additional operations,
unless his charge is excessively high in phosphorus and his steel is
to be very low in the same; in that case he may have to melt clear
down and re-carbonize.
Steel of 130 carbon with phosphorus <.05 may be made on the
basic hearth from a charge containing 10 to 12 phosphorus without
melting below 130 carbon.
If high-carbon Bessemer steel is not uniform, it is not to be
wondered at, but as a matter of fact it is usually found to be fairly
uniform, sufficiently so to work well.
If open-hearth steel of high carbon is not uniform, it is clearly
because the maker would not take a little trouble to have it so.
Assuming that for convenience cast steel is graded for carbon
content by even tens, and that the different tempers are separated
half-way between the tens, we have:
Carbon.
.10 including from .05 to .15
.20 “ “ .16 “ .25
.30 “ “ .26 “ .35
.40 “ “ .36 “ .45
.50 “ “ .46 “ .55
.60 “ “ .56 “ .65
.70 “ “ .66 “ .75
.80 “ “ .76 “ .85
.90 “ “ .86 “ .95
1.00 “ “ .96 “ 1.05
1.10 “ “ 1.06 “ 1.15
1.20 “ “ 1.16 “ 1.25
1.30 “ “ 1.26 “ 1.35
1.40 “ “ 1.36 “ 1.45
 Carbon.
1.50 “ “ 1.46 “ 1.55
This covers the usual commercial range from what is known as
dead-soft steel up to a high, lathe-temper steel.
Higher steels are used sometimes, even up to 225 carbon, but
they are so exceptional that it is not worth while to continue the list
above 150.
This list allows a variation of .05 carbon above and below the
datum of each temper; some margin must be had of course, and this
is sufficient in the hands of a careful steel-maker; it is found in
practice to be satisfactory to the user. Even in the highest lathe-steel
where the strains from hardening are the greatest, because the
change in volume due to a degree of temperature is the greatest, a
variation of three or four points above and below the mean does not
make enough difference in the results to throw a skilful temperer off
from his desired conditions.
On the other hand, a difference of a full temper will throw the
most skilful worker off from the track, and so that much variation is
not allowable. For instance, if a man be working 130 carbon, and he
should receive a lot of steel of 120 carbon, he would get his work too
soft in following his regular methods; then if he doubted himself, as
he would be apt to do, and raised his heat to correct his supposed
aberration, he would get his work too hard, coarse-grained, and
brittle; if he tried to correct this by drawing to a lower temper color,
his tools would be too soft. Again, if he received a lot of steel of 140
carbon and proceeded in his regular way, he would get a lot of
cracked tools. So that in either case the result would be confusion. It
is probable that in almost any case either 120 or 140 carbon would
make a thoroughly good tool if the temperer knew what he was
working with and adapted his heats to the carbon. But he does not
know of the variation, and even if he did he would say, very rightly,
that he did not propose to make daily changes in his methods to suit
the convenience or the carelessness of the steel-maker.
 It must not be understood, however, that this narrow range for
each temper limits the capacity of the steel; it merely gives the limit
for regular easy working.
To illustrate: A good lathe-tool may be made of 100-carbon steel,
and of 150 carbon; but no worker could use these tempers
indiscriminately, nor even alternately, although he knew which was
which, because he could not change all of his heats say every five
minutes and turn out satisfactory work. A spring of given size, and to
carry a given load, may be made equally good of 60-carbon steel or
of 140 carbon, and such work is done frequently in shops that are
attached to steel-works; but the spring-maker must be told
beforehand what he is to work with, and he must be given enough of
one kind of steel to make say a day’s work, so that he can go along
regularly. The springs will be good, but the one containing 140
carbon will have the highest elasticity and the most life, although
both will have the same modulus of elasticity. The spring-maker who
buys his steel will not submit to any such variations, and he ought
not to be asked to do it, because one temper of steel costs no more
than another, and the selecting out and separating the tempers is
only a matter of a little care.
Is it practicable to keep steel uniform in carbon within such
narrow limits?
In crucible-steel practice it is very easy to do so. All ingots of 60
carbon upwards up to four or four and one half inches square may
be broken completely off at the top, and then the clean fracture will
indicate the quantity of carbon invariably, and after the ingot has
been glanced at and marked properly it is as easy to put it on its
proper pile as to put it on any other. In a good light a competent
inspector will mark thirty or forty ingots per minute and do it correctly;
it is as easy to the trained eye as it is to read a printed page.
This inspection is so important that it should never be neglected.
It is not costly, much less than a dollar a ton.
With larger ingots only a piece can be broken off from the edge,
but if the topper does his work properly, enough can be taken off to
show the temper clearly. Large ingots containing the contents of a
number of crucibles are liable to unevenness of temper from having
uneven mixtures in the pots and from bad teeming into the moulds;
this can be detected usually in the ingot inspection, and if not it can
be found later during another inspection. Such variations are often
called segregations. This question of segregation will be discussed
in a future chapter.
In the Bessemer and the open-hearth practice ocular inspection
of ingots to determine carbon is not used.
Enough examinations have been made to show that the
fractures, although differing from those of crucible-steel, are quite as
characteristic, and ocular inspection could be used. The ingots are
large usually and to handle and top them would be expensive; but
the heats are also large,—from five tons up to thirty tons in one heat,
—and as they are supposed to be homogeneous, one chemical
carbon analysis is enough for each heat.
Below 50 carbon a quick color analysis is accurate enough;
above 50 carbon combustion should be used, for in high carbons the
color test in the best hands is only the wildest guess-work.
The ten-point range of carbon is far more difficult to attain in high-
carbon open-hearth practice than in the crucible. In one case where
the limit fixed in a specification was 90 to 110 carbon, two full
tempers, one of the most skilful and successful concerns in the world
failed to meet the specification in twenty-ton and thirty-ton furnaces.
It was supposed at first that the trouble came from using different
heats, and large lots of billets were sent out with the heat number
stamped on each billet. The same variations were found in every
heat, the carbon ranging from 80 to 120. The specification was met
without any trouble in five-ton furnace.
This illustration should not lead to the conclusion that practically
uniform steel cannot be obtained; there is little doubt that if the 30-
ton heats had been stirred thoroughly in the furnace the required
limits would have been obtained.
Neither is it to be understood that the same variation would occur
in mild steel under 30 carbon. A call for 20 carbon would not result in
steel ranging from below 10 to above 30,—such a result would show
gross carelessness on the part of the melter,—the variation would go
by percentage; thus the variation in the high steel is from 15% below
to 15% above the mean of 100, or even as much as 20%.
If 20-carbon steel be required, a variation of 20% would give a
range from 16 to 24 carbon, or well within the limits of one temper.
This matter will be considered farther under the head of
Segregation.
The appropriate applications of the different tempers of steel
have been stated in a general way, with the advice that for all tool
purposes it is better to leave the selection of the temper to the steel-
maker; also in structural work it may prove to be better to leave the
question of temper, or carbon content, to the steel-maker, who
should know how to meet any specification that is within the capacity
of steel. On the other hand, every engineer should know what is
attainable, and an effort to give this information in more definite form
will be made in later chapters. A general view will now be taken of
what may be called the carbon-line.

Let the horizontal line represent iron, the inclined line iron plus
carbon, and the verticals physical properties.
We do not know the physical properties of pure iron. Assuming
them to be uniform, let the vertical at .05 represent the tensile,
torsional, transverse, or compressional strength of steel of 5 carbon;
then for every increment of carbon up to 90 to 100 there will be an
increase of strength to resist any of these strains, increasing in such
regular amounts as to make the resulting carbon-line practically
straight, as shown in the sketch. Above 100 carbon these
resistances will all decrease, except resistance to compression.
So far as it is known, compressive strength increases slightly with
the carbon, until cast iron is fairly reached; then the presence of
silicon, and the fact that we are dealing with a casting instead of
forged or rolled metal, causes a rapid fall in all resistances until the
strength is below that of 5 carbon steel.
With increase of carbon there is a reduction of ductility, so that
the extension of length and reduction of area decrease as the
strength increases. In every case the engineer must decide how little
ductility he can do with safely in securing the ultimate strength or the
elastic limit he may require.
The highest strength and the greatest ductility cannot be had
together; they are inverse functions one of the other.
If the exact resistances due to carbon were known along the
whole line, it would be of great value to give them here but nearly all
of the thousands of tests published are influenced by the quantities
of silicon, phosphorus, sulphur, manganese, or oxides present, and
an effort to determine the effects of the carbon-line exactly would be
hazardous.
Kirkaldy’s tests of Fagerota steel, published in 1876, furnish a
valuable guide in this direction.
Webster’s experiments on the effects of the different elements,
phosphorus, manganese, etc., are interesting and valuable, but he
has not yet tested a complete carbon-line with no other variables.
It has been stated time and again by experienced steel-makers
that the best steel, the most reliable under all circumstances, is that
which comes nearest to pure iron and carbon.
Some intelligent steel-makers, and engineers cast doubts upon
this statement, and assert that because phosphorus up to a certain
limit, or manganese, or silicon, or in fact it may be said almost any
element, added to dead-soft steel will give an increase of strength,
therefore the presence of one or more of these elements is not only
not harmful, but beneficial.
 As a matter of fact, however, every one of these elements is
harmful, either in producing cold-shortness, or red-shortness, or
brittleness, and not one of them will add any good quality to steel
that may not be obtained better by the use of carbon. Given a
uniform minimum content of these impurities, the carbon-line may be
depended upon to furnish any desirable quality that is obtainable in
steel; and it is certain, always sure, that that steel which is the
nearest to pure carbon and iron will endure the most punishment
with the least harm.
That is to say, that such a steel when overheated a little, or
overworked, or subjected to any of the irregularities that are
inevitable in shop practice, will suffer less permanent harm than a
steel of equal strength where there is less carbon and the additional
strength is given by any other known substance.
It is difficult to show this from testing-machine data, indeed it is
doubtful if any such data exist, but experience in the steel-works, in
the bridge- and machine-shops, and in the field proves it to be true.
For further discussion of this question see Chap. X.
The effects of a small difference in phosphorus or in silicon
contents are shown plainly and unmistakably in high-carbon steel,
and not so plainly in low-carbon steel; but as there is no known hard
and fast line that divides low steel, medium steel, and high steel, so
there is no marked difference in their properties. The same rules
hold all along the line, the same laws govern all of the way through.
There is no set of properties peculiar to low steel and another set
peculiar to high steel; the same laws govern all, and differences are
those of degree and not of law.
Given three samples of steel of the following compositions:
No. 1. No. 2. No. 3.
Silicon .02 .20 .02
Phosphorus .01 .01 .02
Sulphur .005 .005 .005
Manganese .100 .100 .100
 No. 1. No. 2. No. 3.
Carbon 1.100 1.100 1.100
A skilful worker, not knowing the composition of any, will pick
them out invariably by tempering them and testing them with a hand-
hammer and by inspecting the fractures.
He will pronounce No. 1 to be the best and the strongest in every
way; No. 2 to be not quite as strong as No. 1, and more liable to
crack from a little variation in heat; No. 3 to be not so strong as No.
1, and that it will not come quite as fine as either of the others, and,
like No. 2, it will not stand as much variation in heat as No. 1.
Give a ton of each to a skilful axe-maker, from which he will make
one thousand axes of each, and he will be sure to report No. 1 all
right; No. 2 good steel, more loss from cracked axes than in No. 1.
No. 3 good steel, some inclination to crack; it will not refine as
well as No. 1 and is not as strong.
This is no guess-work, nor is it a fancy case; it is simple fact,
borne out by long experience.
Give a skilful die-maker one hundred blocks of each to be made
into dies. He will not break one of No. 1 in hardening them; he will
probably break five to ten of No. 2; and if he breaks none of No. 3—a
doubtful case—he will find in use that No. 1 will do from twice to
twenty times as much work as either of the others. If he is making
expensive dies,—many dies cost hundreds of dollars each for the
engraving,—he will think No. 1 cheap at 25 cents a pound, and
either of the others dear at 15 cents a pound.
In such steel, then, the absence of a few points of silicon, or of a
point or two of phosphorus, is worth easily 10 cents a pound.
Now let the carbon in these three steels be reduced to 10,
making them the mildest structural steel. The differences to be found
in the testing-machine in tensile strength, elastic limit, extension, and
reduction of area will be almost or altogether nothing; in forging,
flanging, punching, etc., under ordinary conditions differences would
not be observable; therefore there would be no practical difference in
value. But let the silicon be raised to 30 or the phosphorus to 10,—
the Bessemer limit,—or let both be raised together, and both the
testing-machine and shop practice would show a marked difference.
This shows that in the absence of carbon the action of these
elements is sluggish as compared to their effects in the presence of
high carbon, or in the low-carbon steels their effects are not so
observable. That their influence is there, there can be no doubt, but if
it be not enough to endanger the material it is not worth while to take
it into account.
Is it safe and wise, then, for steel-users to ignore composition?
Users of tool-steel may do so safely, because the smallest
variations will manifest themselves so unmistakably that they give
immediate warning, and the steel-maker must keep his product up to
a rigid standard of excellence or lose his character and his trade.
Many of the ablest users of structural steel take a similar ground,
and say, We have nothing to do with method or composition if the
material meets our tests.
It is believed that if these men knew how easy it is for a skilful
worker to doctor temporarily an off heat by a little manipulation, and
how dangerous the same may become by a little off practice in the
field, they would be convinced that some limits should be put upon
composition, especially if they could realize that a reasonable
specification would add nothing to cost, as competition would take
care of that.
The reader is referred again to Chap. X on impurities.
 V.
GENERAL PROPERTIES OF STEEL.

Steel is very sensitive to heat. In general it may be stated that, starting with
cold steel, every degree of heat added causes a change in size and in structure,
until the limit is reached where disintegration begins. The changes are not
continuous; there are one or two breaks in the line, notably at the point where we
have what is called recalescence; this is a marked phenomenon and it will be
considered later.
The effects of heat are permanent, so that it is a fact that every variation of
temperature which is marked enough to be visible to the naked eye will leave a
structure, due to that variation, when the steel is cold, which will be observable by
the naked eye, and such structure, when not influenced by external force, such
as by hammering or rolling, is as invariable and certain as is the structure of an
ingot due to the quantity of carbon present.
This property furnishes what may be called the steel-maker’s and the steel-
user’s thermometer. By its means the steel-maker can discover every irregularity
in heating that may have been perpetrated by the operatives; so also the steel-
user can decide whether the steel furnished him has been heated and worked
uniformly and properly, and later he can tell whether those who have shaped this
steel to its final forms have done their work properly. A thorough knowledge of
this property is essential to a steel-maker; until he possesses it he is not fit to
conduct his business. It is of great importance to the steel-user, and every
engineer should try to acquire a knowledge of it in order that he may not be
fooled by the carelessness or rascality of those who have preceded him. The
steel-maker acquires this knowledge by daily contact with the facts; the engineer
does not have it forced upon him in this way, but he should seek opportunities of
observation, which will be abundant in his earlier practice when he is sent upon
inspection duty. Like the structure of ingots, this heat-structure cannot be
illustrated on paper, and an attempt to do so would be misleading; attempts at
description will be made in the hope that by their means the engineer will have a
pretty good idea what to look for, and to know when his suspicions should be
aroused.
In addition to the ocular observations mentioned it has been shown by
specific-gravity determinations, and by delicate electrical tests through small
ranges of temperature, that steel is as truly thermometrical as mercury.
Steel passes through or into four general conditions due to heat. First, in the
cold state, it is a crystalline solid of no uniform structure, for its structure is
influenced by every element that enters into it, and by every irregularity of heat to
which it has been subjected.
Good steel may be described as having a bluish-gray color, uniform grain as
seen by the naked eye, and little lustre. But it should have some lustre and a silky
appearance. When it is right, a steel-worker will say it is “sappy,” and that name,
absurd as it may sound when applied to a metal, really expresses an
appearance, and implies an excellence that it would be hard to find a better word
for. If the structure be dull and sandy-looking, the steel-worker will say it is “dry,”
and that term is as suggestive and appropriate as the word “sappy.”
If the fracture be granular with bright, flashing lustre, the steel-worker will say
it is “fiery,” and again his term is expressive and proper.
It is perfectly safe to say that steel of a “sappy” appearance is good steel; but
in order to know what it is it must be learned by observation, it cannot be
described in exact terms.
It is equally certain that a “dry” fracture indicates a mean steel, a steel
inherently mean,—too much phosphorus, or silicon, or oxides, or all combined,—
and such a steel is incurable.
A “fiery” fracture indicates too much heat. It may be found in the best steel and
in the poorest; it may be corrected by simply heating to a proper temperature. It
shows that some one needs to be reprimanded for careless work.
If now an inquirer will take a piece of good steel of “sappy” fracture, and of
“dry” steel of dull, sandy fracture of the same carbon, and will heat them say first
to dark orange, then to bright orange, dark lemon, and so on, and examine the
fractures after each heating, he will find a “fiery” fracture in the “dry” steel at a
heat much below that which is necessary to make the “sappy” steel “fiery.” This is
one proof that good steel will endure more punishment than poor steel.
Cold steel is not plastic in the common acceptance of the word; strictly
speaking it has some plasticity, as shown in the extension noted in pulling it; this
is its measure of ductility.
Also it may be drawn cold to fine wire of only a few thousandths of an inch in
diameter, and it has been rolled cold to one five thousandth of an inch thick. But
this work must be done with great care; the steel soon becomes brittle, and a little
overdrawing or overrolling will crush the grain and ruin the steel; therefore the
work must be done a little at a time, and be followed by a careful annealing.
To reduce a No. 5 wire rod to .005 inch diameter will require with high steel
suitable for hair-springs about fourteen annealings.
A skilful hammerman will take a piece of mild cold steel, and by means of
light, rapid blows he will heat it up to a bright lemon heat without fracturing it; then
he will have it thoroughly plastic and malleable.
 This has no practical commercial value; it is a beautiful scientific experiment
exhibiting high manual skill, and showing that there is no hard and fast line
between non-plasticity and plasticity.
The first condition, then, is cold steel, not plastic, not malleable.
When steel is heated, it begins to show color at about 700° to 800° F.; the first
color is known as dark cherry red, or, better, orange red: above this color it turns
to a distinct, rather dark, or medium orange color; this is the heat of recalescence,
a good forging-heat, and the best annealing- and quenching-heat. At this heat
and above it good steel is truly plastic and malleable; a roller or hammerman will
say, “It works like wax,” and so it does.
This is the second or plastic condition.
Heated above this plastic condition to a bright lemon in high steel, or to a
creamy, almost scintillating, heat in mild steel, steel will go to pieces under the
hammer or in the rolls; the workman will probably say it is burned, but it is not
burned necessarily; it is simply heated up to the third or granular condition; it is
the beginning of disintegration and the end of plasticity.
This granular condition is important in several ways. It is made use of in
Sweden, and has been demonstrated in the United States, to determine the
quantity of carbon in steel. An intelligent blacksmith is given a set of rods of
predetermined carbon, ranging from 100 carbon to zero, or through any range
that may be necessary; each rod is marked to indicate its carbon. He takes the
rods one by one and heats them until they scintillate, well up into the granular
condition, then lays them on his anvil and hammers them, observing carefully the
color at which each one becomes plastic as it cools slowly. After a little practice
he is given rods that are not marked, and by treating them in the same way he will
give them their proper numbers, rarely missing the carbon by as much as 10
points, or one temper.
It is a beautiful and useful illustration of the effect of carbon. The rule is, the
higher the carbon the lower the granulating-point; or, as is well known, high steel
will melt at a lower temperature than low steel.
This shows that every temper of steel has its disintegration temperature where
it passes from plastic to granular, as fixed as its fusion-point or its point of
recalescence.
Steel passes from the granular condition to the liquid or fourth form.
There is little of interest in the liquid condition of steel to any but the steel-
maker; what there is to be said will be mentioned later.
Steel in cooling from the liquid passes through the granular and the plastic
conditions to the cold state.
 The granular form is of special interest to the steel-maker for the reason that
in this condition the steel has more of adhesion than cohesion; it will stick to
anything it touches, and so cannot be made to flow. This is the cause of “bears,”
“stickers,” and many of the troubles of the melter. Therefore steel must be put into
the moulds while it is still molten, and moulds should be well smoked or lime-
washed to prevent stickers. This condition is of great interest to engineers,
because the failure to roll or shape molten steel by pouring it directly between the
rolls is doubtless due to this adhesive, non-cohesive condition.
To produce sheets, bars, and all sorts of shapes from molten steel direct,
without the expense of making, handling, and re-heating ingots, is an enticing
idea which has occupied the minds and efforts of many able mechanics and
engineers.
If steel passed directly from the liquid to the plastic condition as glass does,
hammers and rolls would soon be replaced by dies at a great saving of cost and
labor. It is no wonder that such a desirable end has led to many persistent and
costly efforts, but until some way can be devised to eliminate this granular form in
cooling it would seem that all such efforts must end in failure.
As steel cools down through the plastic condition the cooling is not
continuous; there are two or three points where it is arrested for a time, and at
one notable point the cooling is not only arrested, but after a few moments of stop
the operation is reversed, the steel becomes visibly hotter, and then the cooling
goes on regularly; there may be other slight pauses, but they are of little
importance compared to this one, which is known as the point of recalescence.
There are many theories of the cause of this recalescence; the ablest scientists
are still working at it; and until some definite conclusion is reached it is not worth
while to write pages of discussion which may be found fully stated and illustrated
over and over again in the various technical journals, and transactions of different
engineering societies.
There are some properties of steel of great interest which seem to cluster
around this recalescence-point; they will be noted as they are reached.
We have seen that there is a marked, definite structure of the grain of ingots
due to every quantity of carbon, and also that there is a fixed limit of malleability
for every quantity of carbon. It is known also that the recalescence-point shifts
slightly with a change of carbon, and that it is much more marked and brighter in
high-carbon steel than in low.
There are no other sure indications of the quantity of carbon present. As soon
as an ingot is heated up to orange color, or the recalescent-point, it loses its
distinctive structure and its fracture no longer furnishes a sure guide.
If three ingots of say, 20, 80, and 120 carbon respectively be heated to orange
and then cooled slowly, their fractures will be so different as to enable an expert
to place them properly in their order of carbon, and to classify them as mild, hard,
and harder; beyond that he could not go; if he attempted to give them their
temper numbers, he would be likely to miss by four or five numbers either way,
and a correct mark would be only a lucky guess.
Hammering and rolling heated steel affect the grain or structure profoundly; a
high steel may be worked so that the grain will look mild, and a mild steel may be
so worked that the grain will look hard. It is common to see a bar of steel with a
fine grain at one end and a coarse grain at the other, and this state of things often
frightens a consumer, who imagines that he has received a very irregular, uneven
article, and he is as often astonished when it is shown to him that at the same
proper heat the two ends will refine and harden equally well, and be exactly alike.
In such a bar one end has been finished a little hotter than the other, and the
grain is due to the heat in each case. This uneven heating may have been
incidental or careless; with skilful workers it is rare.
One end might have been finished so cold as to crush the grain, and the other
end so hot as to cause incipient disintegration, but a competent inspector would
discover either condition at once and reject the bar.
There is, then, a specific structure due to temperature; it is modified by carbon
and by treatment under the hammer or in the rolls. If a bar of steel be heated up
to the highest plastic limit, just so that it will not fall to pieces, and then cooled
slowly without disturbance, and a fracture be taken, it will be found to be coarse
and with an exceedingly brilliant lustre. Now let it be heated again to a bright
lemon color, but still plastic, and cooled as before; it will be found to be coarse,
with bright lustre, but neither so coarse nor so bright as the first piece. Then let it
be treated in this way to lemon color, light orange, medium orange, dark orange,
and orange red; as the heats go down the grain will be finer and the lustre will be
less, until at about medium orange the lustre will be absent.
If any number of bars of even composition be heated in this way, the fractures
will all be alike for each temperature.
If a series of bars of the different full tempers, about seven in all, be treated in
this way, the structures due to a given temperature will all be similar, but there will
be no two exactly alike, because high steel is much more profoundly affected by
heat than low steel.
Seven tempers are mentioned here, because that is the number of full
tempers in common use. Steel is graded out into fifteen tempers ordinarily by the
interpolation of half numbers; this is easy and sure in the ingot inspection. In the
above experiment the differences due to carbon are not quite so delicate, and the
work is hampered in the heating by the personal equation, so that the use of
seven full tempers is refinement enough. There is a difference due to every
separable quantity of carbon, which could be shown if all of the operations of the
experiment were exact.
 If when a bar is broken cold the fracture is uneven, with coarse grain in one
part and fine grain in another, it shows that there has been uneven heating. If one
side has large grain and the other side is fine, the bar has been a great deal
hotter on the side having coarse grain than on the other: the heater has let the
bar lie in the furnace with one side exposed to a hot flame and the other
protected from the flame in some way; he has neglected to turn the bar over and
heat it evenly.
If the outside of the bar is fine and the centre is coarse, the bar has been very
hot all through and has been finished by light blows of the hammer or by light
passes in the rolls; it has been worked superficially and not thoroughly.
If the outside of the bar is coarse and the centre is fine, the steel has been
heated on the surface too hot and too quickly; it has not had time to get hot
through, and it has had too little work in the finishing.
If the grain is dark, with the appearance of a rather heavy india-ink tint, the
steel has been finished too cold, and it will be found to be brittle.
If the grain is very dark, especially about the middle, looking almost black,
then it has been finished altogether too cold: the grain is disintegrated, and the
bar is fit only for the scrap heap.
A bar of this kind containing enough carbon to harden will harden thoroughly,
and often appear to be sound and fine, but it is not sound and will not do good
work; if it be brought up to a proper heat and forged to a point, it will almost
certainly burst, showing that the integrity of the steel has been destroyed.
If a bar, or plate, or beam shows cracks on the surface or at the corners, with
rough, torn surfaces, the steel has either been superficially burned or it is red-
short. In either case it should be rejected, for the cracks, although small, will
provide starting-points for ultimate fractures, whether it be tool-steel that is to be
hardened, or structural steel that is to be strained without hardening. If the steel is
to be machined, so that all of the cracks can be cut out, then in machinery-steel
the removal of these surface defects might leave the finished piece sufficiently
sound and good. If, however the steel is to be hardened, and the defects should
be due to red-shortness, the piece would almost certainly break in the hardening;
and if it were not red-short, then unless the cracks were cut away entirely, if the
least trace of the crack is there, although it may not be visible, that trace will be
sufficient to start a crack when the piece is hardened.

EFFECTS OF COOLING.
Increase of heat causes increase of softness up to the liquid condition.
Decrease of heat—cooling—increases hardness up to the hardness of glass.