Analytical Chemistry Refresher Manual (Kenkel, John)

.
A n a ly tic a l
^ ^ h e m is t r y
REFRESHER MANUAL
-J o h n
K enkel
CRC Press
(cJ* Taylor &.Francis C ro u p
Boca Raton London New York
CRC Press is an imprint of the

Taylor & Francis Group, an inform a business
Published 1992 by CRC Press LLC
Taylor & Francis Group
6000 Broken Sound Parkway NW, Suite 300
Boca Raton, FL 33487-2742
©1992 by Taylor & Francis Group, LLC

CRC Press is an imprint of Taylor & Francis Group, an Informa business
First issued in paperback 2019
No claim to original U.S. Government works
ISBN-13: 978-0-367-45037-3 (pbk)

ISBN-13: 978-0-87371-398-6 (hbk)
This book contains information obtained from authentic and highly regarded sources. Reasonable efforts have been
made to publish reliable data and information, but the author and publisher cannot assume responsibility for the
validity of all materials or the consequences of their use. The authors and publishers have attempted to trace the
copyright holders of all material reproduced in this publication and apologize to copyright holders if permission to
publish in this form has not been obtained. If any copyright material has not been acknowledged please write and let us
know so we may rectify in any future reprint.
Except as permitted under U.S. Copyright Law, no part of this book may be reprinted, reproduced, transmitted, or
utilized in any form by any electronic, mechanical, or other means, now known or hereafter invented, including
photocopying, microfilming, and recording, or in any information storage or retrieval system, without written
permission from the publishers.
For permission to photocopy or use material electronically from this work, please access www.copyright.com
(http://www.copyright.com/) or contact the Copyright Clearance Center, Inc. (CCC), 222 Rosewood Drive, Danvers,
MA 01923, 978-750-8400. CCC is a not-for-profit organization that provides licenses and registration for a variety of
users. For organizations that have been granted a photocopy license by the CCC, a separate system of payment has been
arranged.
Trademark Notice: Product or corporate names may be trademarks or registered trademarks, and are used only for
identification and explanation without intent to infringe.
Visit the Taylor & Francis Web site at

http://www.taylorandfrancis.com
and the CRC Press Web site at

http://www.crcpress.com
Library of Congress Cataloging-in-Publication Data
Kenkel, John.
Analytical chemistry refresher manual/by John Kenkel.
p. cm.
Includes and index.
ISBN 0-87371-398-2
1. Chemistry, Analytic. I. Title.
QD75.2.K447 1992
543— dc20 91-38547
Library of Congress Card Number 91-38547

D ed ica tio n
This book is dedicated to my mother, Carmen Kenkel, who, follow

ing the death of my father when I was quite young, functioned as both
mother and father for me during the years when I was growing up.
In spite of her sudden death in 1974, she continues to this day to be
a major source of inspiration for me and there is no doubt that her
prayers on my behalf from her present heavenly home have resulted
in the divine intervention needed to begin and complete this book.
Thanks, Mom.
A uthor
John Kenkel has been on the chemistry faculty at Southeast Com

munity College in Lincoln, Nebraska, since 1977. He has been chair of
the Environmental Laboratory Technology training program there
since 1979. In this position, he has trained more than 300 students in
the techniques of modern analytical chemistry, including both wet
methods and instrumental methods. Graduates of his program are
acclaimed as outstanding people for technician-level laboratory posi
tions.
Professor Kenkel, born and raised on a small farm in western Iowa,

has a B.A. from Iowa State University (1970) and a M.A. in chemistry
from the University of Texas at Austin (1972). While at Texas, he
worked under Professor Allen Bard in electroanalytical chemistry. He
was employed as a corrosion chemist at the Science Center, Rockwell
International, Thousand Oaks, California, from 1973 to 1977. He has
co-authored more than 15 articles stemming from his research at the
University of Texas and at Rockwell and also 2 papers relating to
chemistry technician training.
His first book, "Analytical Chemistry For Technicians", published

by Lewis Publishers/CRC Press in 1988, has become a primary train
ing manual for analytical chemistry in 2 -year college laboratory tech
nician training programs across the country.
Professor Kenkel is a long-standing member of the 2 -year college
chemistry organization, 2YC3, and of the 2YC 3 Advisory Board,
midwest region. He is also a member of the American Chemical
Society and has been active in the ACS Division of Analytical Chem
istry as a member of their Committee on Education. He is the chair
man of the subcommittee on Associate Degree Education. As a mem
ber of this subcommittee, he has created and is editor of the Newslet
ter for Chemistry Technician Instructors (NCTI), a semiannual news
letter for other instructors in this field.
He has won several awards for teaching excellence. In 1985, he was

awarded the Burlington Northern Faculty Achievement Award at
Southeast Community College. In 1988, he won a regional Catalyst
award for excellence in college chemistry teaching. This award is
sponsored by the Chemical Manufacturers Association. In 1989, he
won the Ohaus/National Science Teachers Association award for
innovation in college science teaching. He has twice been a nominee
for the Wekesser Outstanding Teaching Award at Southeast Commu
nity College.
P r efa ce
This work is intended to be, as the title implies, a refresher in the

techniques and methodology of modern analytical chemistry. It is
directed specifically toward technicians and other workers in the
analytical chemistry laboratory who find that they need a personal
reference manual to help acquaint (or reacquaint) them with tradi
tional and nontraditional chemical analysis techniques that may have
unexpectedly become part of their jobs. These individuals may have
received no formal training in this discipline or may have received
formal training but, given the rapid growth of this science in recent
years, find themselves sinking in the maze of procedures and special
techniques that are encountered today.
It is intended to be a book for individuals who were biology majors,
geology majors, environmental science majors, etc., who find that they
are expected to function as chemical analysis technicians. It is in
tended to be a fundamental, readable reference on analytical chem
istry packed full of tidbits of theory and background highlights of all
popular analytical methods.
The book includes information on sampling and sample prepara
tion, solution preparation, and discussions of wet and instrumental
methods of analysis. The spectrometric techniques of UV, vis, and IR
spectroscopy, including FTIR, are fully covered. NMR and mass
spectrometry, as well as the full complement of atomic spectroscopy
techniques such as flame photometry, flame AA, graphite furnace
A A, and ICP, are covered. A thorough discussion of analytical sepa
rations, including liquid-liquid extraction, liquid-solid extraction,
both instrumental and noninstrumental chromatography, and electro
phoresis, are covered. Gas chromatography and high performance
liquid chromatography are given separate chapters in which the basic
theory and instrument design concepts are covered. Potentiometric
and voltammetric techniques are presented followed by a chapter on
automation. The final chapter in the book discusses the detection and
accounting of laboratory errors.
The author hopes that laboratory workers at all levels in all types
of applications, from environmental analysis to quality assurance,
from private testing laboratories to government analysis laboratories,
from the manufacturing industry to water and wastewater treatment,
and from routine work to academic research, will benefit.
A c k n o w led g m en t s
The Analytical Chemistry Refresher Manual would not have hap

pened were it not for the help and encouragement of a number of
people. I would first like to thank members of the analytical chemistry
community for supporting the concept of analytical chemistry books
for the technician-level laboratory worker, and I hope that this book
will serve their needs.
Second, I would like to thank my editor, Brian Lewis of Lewis
Publishers, Inc., for suggesting this book and for his patience and
encouragement as the project unfolded. It has been a pleasure to work
with him.
Third, I say "thank you" to those chemistry professionals whose
names are unknown who read my initial outline and made sugges
tions. Their help in developing the book in its final form was very
important.
Several people read the first draft of the manuscript and offered
suggestions. These include Brad Godwin of PACE Laboratories, Inc.
of Lenexa, Kansas, and Reza Rafat of SmithKline Beecham Animal
Health in Lincoln, Nebraska, and another who remains anonymous.
Thank you all for your very positive contributions.
I would like to thank my primary artist, Lana Johnson of Lincoln,
Nebraska, who drew most of the illustrations in the book. She has had
a very major impact in the book's quality, and I appreciate it very
much.
I sincerely appreciate the love and understanding of my wife, Lois,
and children, Angie, Jean, and Laura, throughout this period. Without
their smiling faces and their understanding, completing this book
would have been a much greater burden. As it was, it was a very
pleasant experience.
Finally, and most important of all, I want to acknowledge my
heavenly Father for His many blessings on me and my family, espe
cially for the gifts of life and happiness that come from Him. This book
would only have been a passing thought if not for Him.
John Kenkel
Southeast Community College
Lincoln, Nebraska
T a ble o f C o n ten ts
Chapter 1
Introduction to Analytical Chemistry................................................ 1
1 .1 Importance of AnalyticalChem istry.......................................... 1
1 .2 Precision vs A ccuracy...................................................................3
1.3 T erminology.....................................................................................5
1.4 Fundamentals of Measurement.................................................. 6
1.5 Sampling and Planning...............................................................10
1.5.1 Obtaining the Sam ple......................................................10
1.5.2 Handling the Sam ple.......................................................12
1.5.3 Planning the Lab W ork................................................... 12
Chapter 2
Basic Chemical Analysis Tools — Description and U se .............. 17
2.1 Introduction.................................................................................. 17
2 .2 Balances.......................................................................................... 18
2.2.1 The Basic Concept of the Balance................................ 18
2 .2 . 2 The Less Accurate Balances.......................................... 20
2.2.3 The Electronic Analytical Balances.............................. 2 2
2.3 Glassw are......................................................................................24
2.3.1 The Volumetric Flask...................................................... 24
2.3.2 T h e P ip e t............................................................................ 27
2.3.3 Pipetting Devices............................................................. 30
2.3.4 The Buret............................................................................ 31
2.3.5 Cleaning and Storing Procedures................................32
2.4 Reagents and Methods forSample Preparation..................... 36
2.4.1 Water, Concentrated Acids, and B ases...................... 36
2.4.2 Sample Preparation by Extraction...............................39
2.4.3 Fusion..................................................................................42
2.4.4 Reagent Q u ality................................................................42
2.5 Laboratory Safety...........................................................................45
Chapter 3
Solution Preparation...................................................................................47
3.1 Introduction.................................................................................... 47
3.2 Dilution.............................................................................................48
3.3 P ercent..............................................................................................49
3.3.1 Volume/Volume (v/v)...................................................50
3.3.2 Weight/Weight (w /w )...................................................51
3.3.3 Weight/Volume (w /v)...................................................52
3.4 Molarity and Formality................................................................ 54
3.5 N orm ality........................................................................................ 56
3.5.1 Concept of the Equivalent............................................. 57
3.5.2 Solution Preparation.......................................................61
3.6 Parts Per M illion/Billion.............................................................62
3.7 Buffer Solutions..............................................................................65
Chapter 4
Wet Methods and Applications ............................................................. 69
4.1 Introduction.................................................................................... 69
4.2 Gravimetric Analysis.................................................................... 69
4.3 Titrimetric Analysis...................................................................... 70
4.3.1 Standardization................................................................ 72
4.3.2 Titration of Unknow ns...................................................74
4.4 Examples of Real-World Titrimetric Analysis.......................75
4.4.1 The Kjeldahl M ethod......................................................75
4.4.2 Water H ardness............................................................... 78
4.4.3 The Karl Fischer Titration............................................. 80
4.5 Final Com m ents.............................................................................83
Chapter 5
Instrumental Methods — General Discussion ..................................85
5.1 Introduction.................................................................................... 85
5.2 Instrumental Data and Readout................................................ 8 6
5.2.1 Recorders............................................................................ 8 6
5.2.2 Instrument Readout and Concentration.................... 87
5.2.3 Method of Least Squares................................................90
5.2.4 The Correlation Coefficient........................................... 92
5.3 Methods for Quantitative Analysis.........................................93
5.3.1 Series of Standard Solutions
(or Serial Dilution) M ethod........................................... 93
5.3.2 Internal Standard M ethod.............................................94
5.3.3 Method of Standard A dditions...................................94
5.4 Effect of Sample Pretreatment on Calculations...................96
5.5 Use of Computers.........................................................................98
5.5.1 Data Acquisition.............................................................. 98
5.5.2 Data Manipulation and Storage..................................99
5.5.3 Graph Plotting............................................................... 100
5.5.4 Experiment Control....................................................... 100
Chapter 6
Molecular Spectroscopy..........................................................................103
6.1 Introduction.................................................................................103
6.1.1 Nature and Parameters of L ig h t...............................103
6.1.2 Light Absorption and Em ission................................108
6 .2 Names of Techniques and Instruments.............................. I l l
6.3 UV/vis Spectrophotometry.................................................... 112
6.3.1 General Description......................................................112
6.3.2 Instrument Design......................................................... 113
6.3.3 Qualitative and QuantitativeAnalysis..................... 1 2 2
6.4 IR Spectrom etry......................................................................... 129
6.4.1 Introduction.................................................................... 129
6.4.2 Liquid Sam pling............................................................ 131
6.4.3 Solid Sampling............................................................... 133
6.4.4 Instrument Design.........................................................139
6.4.5 Qualitative and QuantitativeAnalysis..................... 144
6.5 Fluorometry........................................................................... 151
6 .6 Nuclear Magnetic Resonance Spectroscopy...................... 155
6.6.1 Introduction.................................................................... 155
6.6.2 The Traditional Instrument........................................158
6.6.3 Chemical Sh ifts..............................................................159
6.6.4 Peak Splitting and Integration.................................. 160
6.7 Mass Spectrometry.................................................................. 162
6.7.1 Introduction.................................................................... 162
6.7.2 Instrument Design..........................................................163
6.7.3 Mass Spectra.................................................................... 164
Chapter 7
Atomic Spectroscopy................................................................................ 167
7.1 Introduction...................................................................................167
7.2 Atom ization...................................................................................168
7.2.1 Fuels and O xidants........................................................168
7.2.2 Burner Designs................................................................ 169
7.3 Excitation........................................................................................ 171
7.4 Flame Photometry.......................................................................172
7.4.1 Introduction..................................................................... 172
7.4.2 Instrumentation...............................................................175
7.4.3 Application...................................................................... 176
7.5 Flame Atomic Absorption......................................................... 178
7.5.1 Introduction..................................................................... 178
7.5.2 Instrumentation...............................................................179
7.5.3 Application...................................................................... 183
7.6 Graphite Furnace and OtherAtom izers................................ 187
7.6.1 Graphite Furnace Atomizer........................................ 188
7.6.2 Vapor Generation M ethods.................. 190
7.6.3 The Delves C u p ..............................................................191
7.7 Inductively Coupled Plasm a....................................................191
7.8 Other Atomic Emission Techniques......................................192
7.8.1 Arc or Spark EmissionSpectrography......................192
7.8.2 Atomic Fluorescence..................................................... 193
7.9 Summary of Atomic Techniques............................................ 193
Chapter 8
Analytical Separations ............................................................................ 195
8.1 Introduction.................................................................................. 195
8.2 Recrystallization...........................................................................196
8.3 Distillation..................................................................................... 196
8.4 Liquid-Liquid Extraction...........................................................198
8.4.1 Introduction.....................................................................198
8.4.2 The Separatory Funnel.................................................199
8.4.3 Theory............................................................................... 200
8.4.4 Countercurrent Distribution....................................... 203
8.4.5 Concentrators................................................................. 205
8.5 Liquid-Solid Extraction............................................................ 205
8 .6 Chromatography........................................................................207
8.6.1 Introduction..................................................................... 207
8.6.2 "Types" of Chromatography......................................208
8.6.3 Chromatography Configurations.............................. 214
8.7 Electrophoresis...........................................................................220
Chapter 9
Gas Chromatography ..............................................................................225
9.1 Introduction................................................................................ 225
9.2 Instrument D esign.....................................................................226
9.3 Sample Injection.........................................................................228
9.4 Columns....................................................................................... 230
9.4.1 Instrument Logistics......................................................230
9.4.2 Packed, Open Tubular, and Preparative
C olum ns............................................................................233
9.4.3 The Nature and Selection of the Stationary
Phase.................................................................................. 234
9.5 Other Variable Parameters......................................................236
9.5.1 Column Tem perature................................................... 236
9.5.2 Carrier Gas Flow Rate.................................................. 238
9.6 The Chromatogram.................................................................. 239
9.7 Detectors...................................................................................... 242
9.7.1 Thermal Conductivity Detector (T C D )....................243
9.7.2 Flame Ionization Detector (FID)................................ 243
9.7.3 Electron Capture Detector (ECD).............................. 245
9.7.4 Nitrogen/Phosphorus Detector (N PD )....................246
9.7.5 Flame Photometric Detector (FPD )........................... 246
9.7.6 Electrolytic Conductivity (Hall)................................. 247
9.7.7 GC-MS and G C -IR .........................................................247
9.7.8 Photoionization Detector (PID)...................................248
9.8 Qualitative Analysis................................................................. 249
9.9 Quantitative A nalysis..............................................................250
9.9.1 Peak Size Measurement............................................... 250
9.9.2 Quantitation Methods....................................................251
9.10 Troubleshooting........................................................................ 257
Chapter 10
High Performance Liquid Chromatography ...................................261
10.1 Introduction........................................................ 261
10.1.1 Basic C oncepts.................................... 261
1 0 .1 . 2 Comparisons with G C ................... / ..............................262
10.1.3 Sample and Mobile Phase Pretreatment..................264
10.2 Solvent Delivery.......................................................................... 266
10.2.1 Pum ps................................................................................ 266
10.2.2 The Gradient Program mer.......................................... 266
10.3 Sample Injection.......................................................................... 268
10.4 Column Selection........................................................................269
10.4.1 Normal Phase Colum ns............................................... 270
10.4.2 Reverse Phase Colum ns............................................... 270
10.4.3 Adsorption C olum ns.....................................................271
10.4.4 Ion-Exchange and Size-Exclusion Colum ns 271
10.4.5 Column Selection............................................................ 272
10.5 The Chromatogram....................................................................274
10.6 Detectors........................................................................................275
10.6.1 UV Absorption................................................................ 275
10.6.2 Diode Array......................................................................276
10.6.3 Fluorescence......................................................................277
10.6.4 Refractive Index.............................................................. 277
10.6.5 Electrochemical............................................................... 278
10.6.6 LC-MS and L C -IR .......................................................... 282
10.7 Qualitative and Quantitative Analysis.................................283
10.8 Troubleshooting.......................................................................... 284
Chapter 11
Electroanalytical M eth o d s......................................................................287
11.1 Introduction................................................................................. 287
11.2 Potentiometry.............................................................................. 288
11.2.1 The Nernst Equation.....................................................289
11.2.2 Reference Electrodes......................................................291
11.2.3 Indicator Electrodes.......................................................296
11.2.4 Potentiometric Titrations.............................................. 302
11.3 Polarography and Voltammetry............................................ 303
11.3.1 Introduction..................................................................... 303
11.3.2 Polarography................................................................... 305
11.3.3 Voltammetry.................................................................... 315
Chapter 12
Automation ........................................................................................................319
12.1 Introduction................................................................................. 319
12.2 Automatic Titrations................................................................. 320
12.3 Segmented Flow M ethods.......................................................320
12.4 Flow Injection M ethods............................................................ 324
12.5 R obotics........................................................................................ 326
Chapter 13
Data Handling and Error Analysis...............................................329
13.1 Modern Data Handling............................................................ 329
13.1.1 Introduction......................................................................329
13.1.2 Data Display and Analysis.......................................... 330
13.1.3 Reporting and Managing R esu lts............................. 331
13.1.4 The State of the A rt.......................................................331
13.2 Introduction to Error Analysis............................................... 332
13.3 Terminology................................................................................. 333
13.4 Distribution of Random Errors.............................................. 335
13.5 Student's t .................................................................................... 336
13.6 Statistics in Quality C ontrol................................................... 338
13.7 Assistance..................................................................................... 338
In d e x ....................................................................................................................341
C h a pter 1
I n t r o d u c t io n to A n a lytic a l
C h em istr y
1.1 IMPORTANCE OF ANALYTICAL CHEMISTRY
We begin this refresher on analytical chemistry with the following

quotation from an editorial which appeared in a recent issue of Chemical
and Engineering News, a publication of the American Chemical Society,
Washington, D.C.
'This should be a golden age for chemists. And in many ways it is. An
unending progression of new instruments and computers continues to
accelerate the pace of chemical research. Determinations that once took years
and were the stuff of doctoral theses can now be done routinely overnight.
Applications of chemistry to interdisciplinary areas are multiplying both the
scope of chemical research and the range of the chemists' contributions to the
public good. The intellectual contributions of chemists to the understanding
of nature remain exciting, rewarding, and boundless. The industries that
chemists have created remain basically sound and progressive. The role of
chemists in handling health and environmental problems is as vital as ever and
growing."’'
While Mr. Heylin's statement precedes an editorial on the critical
Excerpted from Chem. Eng. News, 68(5), 3,1990. With permission. Copyright 1990 Ameri
can Chemical Society.
1
2 Analytical Chemistry Refresher Manual
state of chemistry as a profession in today's world, it could just as

easily precede an editorial on the current and future challenges that
face analytical chemists today. It is truly a golden age. Instrument and
computer manufacturers are producing analytical machines that are
ever-increasing in power and scope. Sophisticated analytical
determinations are indeed becoming routine. Analytical chemistry
now touches more interdisciplinary areas and plays a greater role in
the multiplication of the scope of the chemical science in general than
ever before. The intellectual contributions of analytical chemists to the
understanding of nature, to the chemical industries, and to the sciences
of health and the environment are, to be sure, exciting, rewarding,
sound, progressive, vital, and growing.
The most intriguing adjective Mr. Heylin uses, however, is
"boundless." Such an adjective in effect says that there is no limit to
the scope of the discipline. This word describes modern analytical
chemistry perhaps better than any other. Today, we expect analytical
chemists to be a vital link in an extraordinary number of diverse fields.
Industries that manufacture chemicals, pharmaceuticals, food products,
paints, polymers, plastics, indeed almost any consumable product we
can identify, utilize analytical chemists for quality assurance and for
the research and development of new products. Environmental and
health scientists today rely heavily on analytical chemists for rapid
and accurate analysis of air, water, food, soil, plants, waste, and
biological samples for potentially harmful chemicals. Agricultural
scientists, to a large extent, rely on analytical chemists for the analysis
of feeds, fertilizers, water, soil, plants, and biological samples for
chemical additives and residues. Medical doctors, and thus the general
public, have come to rely on analytical chemistry for an accurate
analysis of blood, urine, skin, and other biological tissue and fluids.
Analytical chemistry has found its way into such fields as veterinary
science, archaeology, farm ing, and space travel. The recent
controversies surrounding the Shroud of Turin (purported burial cloth
of Jesus Christ) and its analysis indicates that even religion is within
its scope. The modern principles and applications of analytical
chemistry in today's world are as important and as relevant to our
existence as any other field that exists. Figure 1.1 presents an illustration
of the dependence of various professional individuals and groups on
the services that an analytical chemist renders.
The purpose of this text is to provide readable and timely descriptions
of fundamental analytical laboratory processes starting with the very
Introduction to Analytical Chemistry 3
FIGURE 1.1 Illustration of numerous interactions that an analytical chemist can

have with other professional individuals and groups in the modern
world. (Courtesy of Steven Zumdahl, University of Illinois,
Champaign-Urbana, IL.)
basic premises upon which all chemical analyses depend, that of the
precision and accuracy of measurement, the fundamentals of sampling,
and the thought processes involved in the planning and execution of
the actual lab work itself.
1.2 PRECISION VS ACCURACY
Analytical chemistry is the science dealing with the determination

of the chemical makeup of real-world material samples, with the
major objective often being the numerical expression of the quantity
of a component present in that sample. The accuracy of this expression
is obviously very important. Analysts should therefore always be
asking themselves if there is any reason to doubt the accuracy of their
results. They must give some thought as to what may be the causes
of inaccurate results. Errors on the part of the analyst himself or of the
equipment or chemicals used are obvious causes. There may be many
potential sources of such error in a given experiment, some of which
can be determined (determinate errors) and some of which cannot
(indeterminate errors).
Determinate errors are errors which are known to have occurred.
They can arise from contamination, wrongly calibrated instruments,
carelessness, etc. They can be avoided by exercising careful laboratory
practices and technique. If it is determined that such an error occurred,
then one obviously cannot trust the answer to be accurate.
Indeterminate errors, however, are impossible to avoid. They are
"random" errors; errors which either were not known to have occurred
or were known to have occurred, but could neither be taken into
account, nor avoided. They are often errors which occur each time a
particular analysis is run. Such errors can determine the predicted
quality of the results, such as the sensitivity and detection limits of
a given experiment. The error in reading a meniscus and the error
associated with an instrument readout are examples of this type of
error. These errors must be dealt with by statistics and the laws of
probability. One way is to repeat a given analysis or procedure a
number of times and use statistics to determine whether the results
are precise. The determination of a "confidence interval" is often the
result of a statistical analysis. (For example, the concentration of nitrate
in a water sample may be expressed as 5.4± 0.2 ppm. The confidence
interval is thus 5.2-5.6 ppm.) We consider precise data as having a
high probability of being accurate.
The two terms "precision" and "accuracy" are often misrepresented.
Accuracy refers to the "correctness" of a given measurement or result.
That is, does the measurement (or measurements) come close to what
the correct answer is or what it is expected to be? A "control" sample
is often run alongside the unknowns as a check on accuracy. Precision
does not necessarily relate to accuracy. Precision refers to how well
a series of identical measurements on the same sample agree with
each other. Figure 1.2 represents a classical illustration of the difference
between accuracy and precision. While precision is not necessarily
synonymous with accuracy, it is often taken to indicate accuracy,
unless a control indicates otherwise or unless there is known to be
a factor which inherently affects accuracy on a continuous and constant
basis. Chapter 13 presents more information on data quality and
methods of handling laboratory data.
Neither precise
nor accurate.
Precise, but
not accurate.
Accurate
and precise.
F IG U R E 1.2 Illustration of precision and accuracy. (Reprinted from Kenkel, J.,

Analytical Chemistry for Technicians, Lewis Publishers, Chelsea, MI,
1988. With permission.)
1.3 TERMINOLOGY
The basic terminology associated with analytical chemistry and

analytical laboratory work is important and may be foreign to persons
who have not been associated with such laboratory work in preparation
for their jobs. We thus present a small glossary of terms in this section.
Other important terms specific to particular analyses are given
elsewhere in this book and can be found in the index.
Chemical Analysis This is the determination of the chemical

composition or chemical makeup of a material
sample.
Qualitative Analysis The determ ination of what substances are
present in a material sample, usually without
the need or desire to determine quantity of

these substances.
Quantitative Analysis The determination of how much of a specified
substance is present in a material sample.
Quantitation This is the determination of quantity, as in the
quantitative analysis above.
Quantification This is another word for quantitation.
Quantitative Transfer A transfer of a chemical or solution from one
container to another, making sure that every
trace of this chemical is in fact transferred.
Analyte This is the substance being analyzed for in an
analytical procedure. This can be an element,
a compound, or an ion.
Assay This is another word for chemical analysis.
1.4 FUNDAMENTALS OF MEASUREMENT
In the analytical chemistry laboratory, many measurements are

made, and the accuracy of these measurements obviously is a very
important consideration. Different measuring devices give us different
degrees of accuracy. A measurement of 0.1427 g is more accurate than
a measurement of 0.14 g simply because it contains more digits. The
former (0.1427 g) was made on an analytical balance, while the latter
was make on an ordinary balance. A measurement recorded in a
notebook should always reflect the accuracy of the measuring device.
It does not make sense to use a very accurate measuring device and
then record a number that is less accurate. For example, suppose a
weight on an analytical balance was found to be 0.14g. It would be
a mistake to record the weight as 0.14 g, even if you know personally
that the weight is 0.14 g. Presumably, there are other people in the
laboratory using the notebook, and your entry will be construed as
to contain only two digits. The following example further illustrates
this point.
Figure 1.3 shows a meter, such as a pH meter, or readout meter on
the face of some other laboratory instrument. The correct reading on
this meter is 56.7. The temptation may be to write down 57. This latter
reading is not correct in the sense that it does not reflect the accuracy
of this measuring device. Measuring devices should always be used
FIGURE 1.3 A measuring device registering a reading of 56.7 and not 57.
(Reprinted from Kenkel, }., Analytical Chemistry for Technicians,
Lewis Publishers, Chelsea, MI, 1988. With permission.)
to their optimum capability, and this means recording all the digits
that are possible from the device. The general rule of thumb for a
device such as the meter in Figure 1.2 is to write down all the digits
you know with certainty and then estimate one more. Obviously, the
meter in Figure 1.3 shows a reading between 56 and 57, or "56 point
something." This "something" is the estimated digit and is estimated
to be a 7. The correct reading is 56.7. For digital readouts, such as with
an analytical balance, this estimation is done for you by the device.
The digits that are actually part of analytical measurements have
come to be known as "significant figures." A knowledge of the subject
of significant figures is important from the standpoint that ( 1 ) one
needs to know the accuracy of a measurement from just seeing it in
a notebook (and not necessarily from actually seeing it displayed on
the measuring device), (2 ) calculations are usually performed using
the measurement, and (3) the correct number of significant figures
must be shown in the result of the calculation.
To cover point number 1 above, the following rules apply.
1. Any nonzero digit is significant. Example, 1.27 (three significant

figures.)
2. Any zero located between nonzero digits is significant regardless
of the position of the decimal point. Example, 1.027 (four
significant figures).
3. Any zero to the left of nonzero digits is not significant, unless it
is covered also by Rule #2. Such zeros are shown only to locate
the decimal point. Example, 0.0127 (three significant figures).
4. Any zero to the right of nonzero digits and to the right of a
decimal point is significant. Example, 1.270.
5. Any zero to the right of nonzero digits and to the left of a decimal
point may or may not be significant. Such zeros may be shown
only to locate the decimal point or they may be part of the
measurement; one does not know unless h e/sh e personally made
the m easu rem en t. Such num bers are actu ally in correctly
recorded. They should be expressed in scientific notation to show
the significance of the zero because then Rule #4 would apply.
Example,
1270. (incorrect)
I.270 x 103 (four significant figures) or 1.27 x 103 (three significant figures)
To cover point number 2, the following rules apply:
1. The correct answer to a multiplication or division calculation

must have the same number of significant figures as in the
number with the least significant figures used in the calculation.
Example, 1.27 x 4.6 = 5.842 = 5.8. Hence, 5.842 is the calculator
answer, and 5.8 is the correctly rounded answer.
2. The correct answer to an addition or subtraction has the same
number of digits to the right of the decimal point as in the number
with the least such digits in the calculation.
Example,
4.271 + 6.96 = 11.231 = 11.23
II.231 is the calculator answer, and 11.23 is the correctly rounded

answer.
3. When several steps are required in a calculation, no rounding
would take place until the final step.
4. When both Rules #1 and #2 apply in the same calculation, the
number of significant figures in the answer is determined by
following both rules in the order in which they are needed, keep
ing in mind that Rule #3 also applies:
Example,
(7 .2 7 - 4 .8 ) x 56.27 = 138.9869 = 1.4 x 102
138.9869 is the calculator answer, and 1.4 x 102 is the correctly

expressed answer.
5. In cases in which a conversion factor that is an exact number is
used in a calculation, the number of significant figures in the
answer depends on all other numbers used in the calculation and
not this conversion factor. To say that a number is exact means
that it has an infinite number of significant figures and as such
would never limit the number of significant figures in the answer.
Example,
1.247 m x 100 cm /m = 124.7 cm (four significant figures)
6. In cases in which the logarithm of a number needs to be

determined, such as in converting [H+] to pH or in the conversion
of percent transmittance to absorbance, the number of digits in
the mantissa of the logarithm (the series of digits to the right of
the decimal point) must equal the number of significant figures
in the original number.
Example,
[ h +] = 4 .9 x 10-6 M
pH = -lo g [H +] = 5.31
This would appear to be an increase in the number of significant

figures compared to the original number (three in 5.31 and only
two in 4.9 x 10"6), but the characteristic of the logarithm, the
digit(s) to the left of the decimal point, represents the exponent
of 10, which serves to only designate the position of the decimal
point in the original number and, as such, is not significant.
1.5 SAMPLING AND PLANNING
1.5.1 Obtaining the Sample
A laboratory analysis is almost always meant to give a result which

is indicative of a concentration in a very large system. A farmer wants
an analysis result to represent the concentration for an entire 40-acre
field. A pharmaceutical manufacturer wants an analysis result to
represent the concentration of an active ingredient in 80 cases of its
product, each case containing three dozen bottles of 1 0 0 tablets each.
A governmental environmental control agency wants a single
laboratory analysis to represent the concentration of a toxic chemical
in every cubic inch of soil within 5 mi of a hazardous waste dump
site.
An ideal analysis is one in which the entire system can be run
through the analysis procedure. Obviously, this is not possible in most
cases. The analyst is then faced with the serious problem of obtaining
a sample from this system which can only be as good as the sample
itself.
Obviously, there are different degrees of difficulty and different
sampling modes involved with obtaining samples for analysis,
depending on the type of sample to be gathered, whether the source
of the sample is homogeneous, the location of and access to the
systems, etc. For example, obtaining a sample of blood from a hospital
patient is completely different from obtaining a sample of coal from
a train car full of coal.
As far as blood is concerned, from what part of the body to sample
a person's blood needs to be considered. Second, the time of day,
along with a knowledge of the patient's recent diet, is important.
Third, perhaps the patient is on some sort of medication which could
affect the analysis.
With the coal sample, it is important to recognize that the coal held
in a train car may not be homogeneous, and a sampling scheme which
includes taking samples from different parts of the whole system
must be implemented.
The key word in any case is "representative." A laboratory analysis
sample must be representative of the whole so that the final result of
the chemical analysis represents the entire system that it is intended
to represent. If there are variations, or at least suspected variations,
in the system, small samples must be taken from all suspect locations.
If results for the entire system are to be reported, these small samples
are then mixed well to give the final sample to be tested. In some cases,
analysis on the individual samples may be more appropriate. Some
examples follow.
Consider the analysis of soil from a farmer's field. The farmer wants
to know whether he needs to apply a nitrogen-containing fertilizer
to his field. It is conceivable that different parts of the field could
provide different types of samples in terms of nitrogen content,
particularly if there is a cattle feed lot nearby, perhaps uphill from
part of the field and downhill from another part of the field. Obviously,
the sample taken should include portions from all parts of the field
which may be different so that it will truly represent the entire field.
Alternatively, two samples could be taken, one from above the feed
lot and one from below the feed lot, so that two analyses are performed
and reported to the farmer. At any rate, one wants the results of the
chemical analysis to be correct for the entire area for which the analysis
is intended.
Consider the analysis of the leaves on a tree for pesticide residue.
The tree grower wants to know if the level of pesticide residue on the
leaves indicates whether the tree needs another pesticide application.
Once again, the analyst must consider all parts of the tree that might
be different. Leaves at the top, in the middle, and at the bottom should
be sampled (one can imagine differences in application rates at the
different heights); leaves on the outside and leaves close to the trunk
should be sampled; and perhaps there would also be differences
between the shady side and the sunny side of the tree. All leaf samples
can then be combined and brought into the laboratory as a single
sample.
Consider the analysis of a blood sample for alcohol content (imagine
that a police officer suspects a motorist to be intoxicated). The problem
here is not sampling different locations within a system, but rather
a time factor. The blood must be sampled within a particular time
frame which would demonstrate intoxication at the time the motorist
was stopped.
The problems associated with sampling are unique to every
individual situation. The analyst simply needs to take all possible
variations into account when obtaining the sample so that the sample
taken to the laboratory truly does represent what it is intended to
represent. A representative sample is one which has all the
characteristics — all the components at their respective average
concentration levels — of the whole from which it is taken.
1.5.2 Handling the Sample
How to get the sample from the sampling site to the laboratory
without contamination or alteration is generally not as challenging,
in most cases, as the problem of how to obtain a representative sample.
There are basically two considerations associated with such sample
preservation: ( 1 ) storage of the sample in a container which will
protect the integrity of the sample, particularly if trace constituents
are to be determined, and (2 ) preservation of the sample from problems
which may be internal, for example, temperature effects or bacterial
effects.
If trace amounts of metals are to be determined, for example, one
would not want to store the sample in a glass container, since glass
can leach small amounts of metals. On the other hand, if trace organics
are to be determined, plastic containers may be deemed inappropriate.
Sometimes, refrigeration may be important to avoid decomposition
from bacterial sources. At any rate, proper sample handling methods
must be employed to ensure sample integrity.
1.5.3 Planning the Lab Work
Once the sampling and sample preservation schemes have been

properly performed, the sample is in the laboratory and ready for the
analysis. The subsequent laboratory work will often involve dissolving
the sample; preparing standards; weighing and adding sample
additives, such as pH adjusters, oxidizing agents and reducing agents,
color developing agents, chelating agents, and the like, before any
analytical measurements are made. What tasks are actually performed
and how they are performed depends on the object of the analysis.
As mentioned previously, analytical work may either be qualitative
or quantitative. With qualitative analysis, we are merely interested in
identifying a constituent of the sample and not determining quantity.
The most important consideration here may be to be certain that the
sample is not contaminated with a chemical that would interfere with
this identification. For example, infrared spectrometry is often used
as a qualitative tool. As will be discussed in Chapter 6 , contaminated
samples used in this technique yield results which may confuse the
analyst. On the other hand, a dilution, or other change in the concen
tration of an analyte in a sample, will not generally affect a qualitative
analysis, as long as the analyte remains above the detection limit of

the technique chosen. Thus, when planning a qualitative analysis, the
use of precise measuring devices, such as analytical balances and
volumetric pipettes, may be excluded from the plan.
A quantitative analytical procedure, however, will certainly involve
measurements and methods which need to be as accurate as possible,
but will also involve measurements and methods which need not be
accurate at all. An efficient analytical technician is one who is able to tell
the difference. What follows is a typical cookbook analytical procedure,
the spectrophotometric determination of manganese in steel, and a
discussion of this procedure, step by step, to illustrate the reasoning
required to discover when accuracy is important and when it is not.
As a general guideline, we can say that accuracy is important when
the lack of accuracy may adversely affect either the quantity of analyte
ultimately measured or the other parameters which are measured and
entered into the calculation of the results.
1.5.3a Exam ple A nalysis — The Spectrophotom etric D eterm ina

tion o f M anganese in Steel
Step 1. Calculate the volumes of 1000 ppm Mn required to prepare

standard solutions of 2 , 5 , 1 0 , 1 5 , and 20 ppm Mn to be placed
ultimately in 100-mL flasks. Measure these volumes into sepa
rate 250-mL beakers. A dd 25 mL of dilute nitric acid (25% by
volume) to each. Use a sixth beaker, containing 25 mL of the
nitric acid, for a blank. Cover with watch glasses.
Step 2. Weigh a sample of the steel (0.25 to 0.30 g) into a seventh
beaker and also add the 25 mL of the dilute nitric acid. Cover
this beaker with a watch glass and simmer on a hot plate in
a fume hood until the steel is dissolved or until only a small
amount of carbon remains.
Step 3. Remove from the hot plate and allow to cool for 10 min. Then
add 0.50 g of ammonium peroxydisulfate to each of the seven
beakers and bring all seven solutions, with watch glasses in
place, to boil in a fume hood and boil gently for 10-15 min.
The purpose of this step is to oxidize the carbon compounds
in the steel.
Step 4. The color developing agent, the oxidizing agent used to
quantitatively oxidize the manganese to permanganate at this
point, is K I0 4. Add, to each beaker, 25 mL of water, 5 mL of
concentrated H3P 0 4, and 0.4 g of K I0 4. Return each to the hot
plate and boil again for 5 min (watch glasses in place). Cool
again and transfer quantitatively to 100-mL volumetric flasks.
Dilute all flasks to the mark and shake.
Step 5. Measure the absorbance of the six solutions (standards and
sample), using the blank for calibration, at 522 nm. Plot
absorbance vs concentration and obtain the concentration of
the sample solution. Calculate the ppm Mn in the steel as
follows:
Ax
C sample, x 0.100
ppm Mn = ----------------------
kg of steel used
1.5.3b Discussion o f Exam ple A nalysis
1. The measurement of the volumes of 1000 ppm Mn placed in the

beakers is ultimately the basis of the standard curve plotted in
Step 5. These volumes will directly affect determination of the
concentration of the m anganese in the sample solution, and
therefore they need to be accurate. The nitric acid, however, is
needed to approximate the matrix of the sample solution, since
nitric acid is needed for the dissolution (Step 2). It will not directly
affect the quantity of manganese in the solutions and thus does
not need to be accurately measured.
2. The weight of the steel needs to be as accurate as possible, since
it enters directly into the calculation of the results in Step 5. Watch
glasses are im portant to prevent losing the manganese via
splashing if and when the solution boils.
3. Since the purpose of the ammonium peroxydisulfate is to oxidize
the carbon compounds in the steel, it does not directly affect the
quantity of Mn measured nor does it enter into the calculation in
Step 5, and it does not need to be measured accurately. An excess
of such chemical additives is generally of little consequence. In
this case, a potential interference is eliminated.
4. The K I0 4 in this step is like the ammonium peroxydisulfate in
Step 3. It is important to get the Mn in the correct oxidation state,
but some excess is unimportant. Its measurement does not have
to be accurate. The same is true of the H3P 0 4. A quantitative
transfer of the solutions to the volumetric flasks is essential, since
the manganese concentration in the standards and sample will be
affected if it is not quantitative. The 100-mL volumes must be

accurate, since again the concentration of the Mn will be affected
and also because this volume enters directly into the calculation
in Step 5 (the 0.100 is liters of solution).
Another important aspect of "planning" is deciding on an approach

to an analysis, including the technique or techniques to use if one is
not specified. The presentation of the theory and application of the
myriad of wet and instrumental techniques and procedures available
to the analyst is the primary objective of this text and cannot be
handled here in a few paragraphs. However, we can present a summary
of the application of these techniques so that you can refer to the
appropriate chapter when requiring more information.
A summary of the applications of the major analytical techniques
for use in this manner is given in Table 1.1. This table is not intended
to be our only answer to the question, however. To know what to do
when presented with an environmental water sample, a soil sample,
a dried paint sample, a rag with something on it, or a chunk of
concrete is often a major problem. Sample preparation schemes as
sociated with such problems — schemes required to prepare a sample
for one of the techniques in Table 1.1 — are also very important.
Chapters 2 and 8 give additional information on this subject, including
methods of complete dissolution of samples and also liquid and solid
extraction.
Finally, because of the value and importance that are usually riding
on the results of a chemical analysis, either qualitative or quantitative,
great care must be exercised in the lab when handling the sample and
all peripheral materials. Contamination and/or loss of sample through
avoidable accidental means cannot be tolerated. The results of a
chemical analysis could affect such ominous decisions as the freedom
or incarceration of a prisoner on trial, whether to proceed with a move
that could mean the loss of one million dollars for an industrial
company, or the life or death of a hospital patient.
Laboratory workers must develop a kind of psychology for operating
in a chemical analysis laboratory. They should always stop and think
before proceeding with a new step in the procedure:. "What might
happen in this step that would cause contamination or loss of the
sample?" In this way, the avoidable (determinate) errors can be
minimized.
Table 1.1 A Summary of the Applications of the Major Analytical Techniques
Technique General Application
Gravimetric analysis Analytes separated by physical means and

easily weighed
Titrimetric analysis Analytes present at high concentrations or for
which there are convenient, well-established
methods
UV-vis spectrophotometry Molecular and ionic analytes capable of
absorbing ultraviolet or visible wavelengths
while in dilute solutions
Infrared spectrophotometry Pure molecular analytes only
Fluorometry Molecular or ionic analytes capable of
fluorescence or fluorescence quenching
while in dilute solution
Atomic absorption and Metals in dilute solution, natural liquids,
emission and extracts and solutions of solids
Gas chromatography Mixtures of volatile organics, organic solvent
extracts, and gases
High performance Complex mixtures (solutions) of analytes,
liquid chromatography (HPLC) including liquids and solids, organic and
inorganic
Electroanalytical chemistry Virtually all forms of analytes, metals, and
nonmetals, molecular and ionic, capable of
oxidation or reduction with small voltages at
an electrode surface or capable of being
detected potentiometricaliy
C h a pter 2
B a s ic C h em ic a l A n alysis
T o o ls — D e sc r ipt io n a n d U se
2.1 INTRODUCTION
Without some introduction to the general basic tools of an

analytical laboratory, the novice in the laboratory may easily become
confused and frustrated. "What is the difference between a top
loading balance and an analytical balance?" "When do I use an
Erlenmeyer flask as opposed to a volumetric flask?" "Why do you
want me to dispense the solution from the buret with my hands in
such an uncomfortable position?" "I thought any acid would dissolve
this piece of copper!" These questions and statements are likely to be
uttered by many a novice. A discussion of basic analytical laboratory
tools is therefore appropriate and necessary.
17
2.2 BALANCES
2.2.1 The Basic Concept of the Balance
The m ost fu ndam ental, and p ossibly the m ost frequ ent,
measurement made in an analysis laboratory is that of weight (or
mass). While we speak of mass and weight often in the same breath,
it is of some importance to recognize that they are not the same. Mass
is the "quantity" or "amount" of a substance being measured. This
quantity is the same no matter where the measurement is made —
on the midwestern plains, on the highest mountain peak, at sea level,
or on the surface of Mars. Weight is a measure of the gravitational
force exerted on a quantity of matter in a certain locale. Weight is one
way to measure mass. In other words, we can measure the quantity
of a substance by measuring the gravitational effect on it. Since 1 0 0 %
of all weight measurements made in any analysis laboratory are made
on the surface of the earth where the gravitational effect is nearly
constant, weight has become the normal method of measuring mass.
Thus, mass and weight have come to be interchangeable quantities
even to the point of the unit involved. Weighing devices are calibrated
in grams, which is defined as the basic unit of mass in the metric
system.
The laboratory instrument built for measuring mass is called the
"balance." A balance is a device in which the weight of an object is
determined by balancing the object usually across a knife-edge
fulcrum with a series of known weights on the other side, as shown
in Figure 2.1. Older balances used in analytical laboratories closely
resembled this basic design, having a pointer in the center to indicate
when a balance between the two pans was achieved. The sum of the
weights on the "known" pan was then calculated, and the answer was
taken to be the weight of the object.
The modern laboratory is of course equipped with balances which
reflect the technological advances that have taken place over the
years. Nearly all modern laboratory balances are of the single pan
variety. The basic principle of the single-pan analytical balance is
shown in Figure 2.2. In this design, there is only one pan. A
permanent, constant counterbalancing weight is in place across a
fulcrum from the pan. The object to be weighed is placed on the pan
and, along with a series of removable weights on the same side, are
Basic Chemical Analysis Tools — Description and Use 19
F IG U R E 2.1 The basic concept of the device known as a "balance." A fulcrum,

or knife-edge, supports a beam across which two "pans" are
balanced; one containing the object being weighed and the other
containing known weights.
Illuminated
display
showing •
weight of
object
Knobs for J T j Counterbalancing

"dialing in" — weight
weights ^ 0
" S ’
F IG U R E 2.2 A schematic diagram of a single-pan balance.
made to balance the constant weight on the other side. When a heavy
object is placed on the pan, some of the removable weights are lifted
via an external control so as to achieve a balance with the
counterbalancing weight. We will discuss these in more detail in the
next subsection (Section 2.2.3).
Different examples of laboratory work require different degrees of
accuracy. Thus, there are a variety of balance designs available which
reflect this need for varying accuracy. Some balances are accurate to
the nearest gram, some are accurate to the nearest tenth or hundredth
of a gram, some are accurate to the nearest milligram, and still others
are accurate to the nearest tenth and hundredth of a milligram. We
now proceed to describe some of the more popular designs, and our
discussions will begin with a shorter treatment of some of the less
accurate balances and end with a detailed treatment of the important
electronic analytical balances, which are very accurate measuring
devices.
2.2.2 The Less Accurate Balances
Figure 2.3 depicts two common balances that can be described as

"less accurate" or "auxiliary." These are used when the weight being
measured does not necessarily need to be accurate. (See Chapter 1 for
a discussion of when to be accurate and when not to be accurate.) The
multiple-beam balance (Figure 2.3a) is accurate to the nearest 0.01 g
and resembles a balance commonly found in a doctor's office to
measure a person's weight. The basic design here is more like the
double pan variety described briefly above in the sense that weights
are added on a counterbalancing arm to balance the object on the pan,
rather than removing weights from the same side as the pan.
Electronic top-loading balances (Figure 2.3b) have variable accura
cies depending on the model purchased. The one illustrated is an
Ohaus Precision Standard balance, which is accurate to the nearest
0.01 g. Other such balances are accurate to the nearest 0.1 and 0.001 g.
The electronic top loaders often have a "tare" feature. A chemical can
conveniently be weighed on a piece of weighing paper, for example,
without having to determine the weight of the paper. "Taring" means
that the balance is simply zeroed with the weighing paper on the pan.
These balances are not based on balance/counterbalance, but rather
utilize a torsion mechanism.
FIGURE 2.3 (a) A multiple-beam balance, (b) An electronic top-loading balance.
(Courtesy of Ohaus Corporation, Florham Park, NJ.)
2.2.3 The Electronic Analytical Balances
The balances that are used to achieve the highest degree of accuracy
in the analytical laboratory are called "analytical" balances and are
accurate to either 0.1 or 0.01 mg. The modern laboratory utilizes
single-pan analytical balances almost exclusively for such accuracy.
Such balances can be either of the balance/counterbalance variety or
the torsion variety.
Figure 2.4 shows a typical modern analytical balance. Notice that
it is a single-pan balance with the pan enclosed. The chamber housing
the pan has transparent walls for easy viewing. Sliding doors on the
right and left sides make the pan accessible for loading and handling
samples. The actual weight measurement is displayed via a digital
readout on the face of the instrument. The counterbalancing
mechanism, if it has one, is hidden in the upper portion of the
instrument and is usually accessible by removing the cover.
The design of the readout and the exact weighing procedure varies
with the age of the balance and the manufacturer. The specific
techniques, such as the use of a "pan arrest" lever, for obtaining the
readout are easy to learn, however, through demonstration and
practice and, therefore, will not be discussed in detail here.
The step-by-step instructions which follow are written to be as
generic as possible. The most important point is that the operator keep
in mind that the analytical balance is extremely sensitive and
extremely accurate and, therefore, must be handled carefully and
correctly. Also, this discussion presumes that, if necessary, the sample
has been dried (such as in an oven) and has been kept dry (by storing
in a desiccator) prior to weighing.
Step 1. Some preliminary considerations are important. First, the

sample to be weighed must be at room temperature. If it is
not at room temperature, it must first be allowed to cool. The
reason for this is that a warm or hot object placed on the pan
in the enclosed chamber can create air currents that can buoy
up the pan and cause an erroneous weight reading. Second,
whenever a weight is being read on the readout, the sliding
doors must be closed. Again, this is to avoid the effect of air
currents on the weight measurement. Third, the pan and the
floor of the chamber must be swept free of chemical debris
from prior spills. A camel-hair brush should be kept near the
FIGURE 2.4 A typical modern analytical balance. (Courtesy Ohaus Corporation,

Florham Park, NJ.)
balance for this purpose. Fourth, care should be taken to avoid

bumping the balance or its support while the measurement
is taken. This would almost always cause the reading to
change or vibrate and could also be damaging to the internal
instrument parts. Fifth, the balance should be level. With most
balances, a leveling method is built in complete with a lev
eling bubble and vertically positionable legs.
Step 2. If the absolute weight of an object on the pan is needed, such
as a crucible weight, the balance must be zeroed. This is a
calibration step in which, when there is nothing on the pan,
the readout is made to read zero. All designs of analytical
balances have a "zeroing control" for this purpose. With
nothing on the pan, the zeroing control is adjusted such that
the readout reads zero even to the fourth or fifth decimal
place. If the absolute weight of an object on the pan is not
needed, such as when weighing by difference or in taring, the
zero step is not absolutely necessary.
Step 3. The object placed on the pan must be kept free of fingerprints,
or other interfering substances, that could add weight and
give an erroneous result. The use of gloves or finger cots is
recommended. When weighing an object to the nearest tenth
of a milligram, seemingly insignificant fingerprints can cause
quite a significant error. Tongs can also be helpful here.
Step 4. The object is placed on the pan and the weight obtained. As
indicated earlier, balances vary in the mechanism for doing
this. Some have a "preweigh" mechanism, and most require
pan arrest while dialing in weights or when adding and re
moving objects from the pan. Most require the operator to
"dial in" the weights so as to achieve a null of some type.
Modern torsion type balances have a digital readout that
automatically displays the weight without "dialing it in."
Step 5. When finished, dial all weight controls to zero if necessary,
clean up any spills, arrest the pan if required, and turn off any
switches.
2.3 GLASSWARE
The accurate measurement of volumes of solutions is a very impor

tant measurement in analytical chemistry. It is important for volumes
of titrants in titrimetric analysis to be measured accurately. Final
solution volumes must be measured accurately when preparing so
lutions of accurate concentration. A variety of reasons exist for accu
rately transferring a volume of a solution from one vessel to another.
It should not be surprising that an analytical chemist needs to be well
versed in the selection and proper use of volumetric glassware.
There are basically three types of volume measuring devices in
common use. These are the volumetric flask, the pipet, and the buret.
Let us study the characteristics of each type individually.
2.3.1 The Volumetric Flask
Let us first deal with the container that is typically used for accurate
solution preparation. We emphasize the word "accurate" here to
distinguish it from solution preparation procedures which do not
need to be accurate. In these latter cases, ordinary beakers, Erlen-

meyer flasks, graduated cylinders, and uncalibrated bottles can be
used. (See Chapter 1 , Section 1.5, for a discussion of when accuracy
is important and when it is not.) The container of choice for accurate
solution preparation is the volumetric flask. A drawing of the
volumetric flask is given in Figure 2.5a.
This flask is characterized by a large base that tapers into a very
narrow neck on which a single calibration line is affixed. This
calibration line is affixed so that the indicated volume is contained
rather than delivered. Accordingly, the legend "TC" is imprinted on
the base of the flask, thus marking the flask as a vessel "to contain"
the volume indicated as opposed to "to deliver" a volume. The reason
this imprint is important is that a contained volume is different from
a delivered volume, since a small volume of solution remains
adhering to the inside wall of the vessel and is not delivered when
the vessel is drained. If a piece of glassware is intended to deliver a
specified volume, the calibration must obviously take this small
volume into account in the sense that it will not be part of the delivered
volume. On the other hand, if a piece of glassware is not intended
to deliver a specified volume, but rather to contain the volume, the
calibration must take this small volume into account in the sense that
it will be part of the contained volume. Other pieces of glassware,
namely, most pipets and all burets, are "TD" vessels, meaning they
are calibrated "to deliver." Figure 2.5b shows a closeup of the base
of a volumetric flask clearly showing the "TC" imprint (just below the
flask's capacity — 500 mL).
Notice the other markings on the base of the flask in Figure 2.5b.
The imprint "A " refers to flasks (and pipets and burets as well) that
have undergone more stringent calibration procedures at the factory
(Class A flasks). Such flasks are more accurate (and also more expen
sive) than flasks that do not have this marking. The imprint "20°C"
indicates that the flask is calibrated to contain the indicated volume
when the temperature is 20°C, which is a standard temperature of
calibration. This marking is needed since the volume of liquids and
liquid solutions changes slightly with temperature. For highest accu
racy, the temperature of the contained fluid should be adjusted to
20°C.
One final marking on the base of the flask is the "19" designation.
This refers to the size of the tapered top on the flask. The stopper that
is used can be either a ground glass stopper, to match the flask
a b
FIGURE 2.5 (a) A drawing of a volumetric flask, (b) A closeup of the base of a
volumetric flask clearly showing the information imprinted there.
opening, or it can simply be a plastic tapered stopper, which can also

be used in a ground glass opening. It has an identical numerical
imprint on it ("19" in this case) and such number designations on
these two items must match to indicate that the stopper is the correct
size. The stopper is not necessarily a tapered stopper as in this
example. It is fairly common for the stopper to be a "snap cap." In
this case, the top of the flask is not tapered nor is it a ground glass
opening. Rather, it has an unusually large lip around the opening,
over which the snap cap is designed to seal. A number designation,
however, is used as with the tapered opening to indicate the size of
the opening and the size of the cap required. This number is found
on both the flask base and the cap as in the other design.
Volumetric flasks are manufactured in a variety of sizes. Since there
is only a single calibration line, the use is limited to rather common
volumes: 5,10, 25, 50,100, 200, 250, and 500 mL, and 1 and 2 L. One
problem with the volumetric flask exists because of its unique shape,
and that is the difficulty in making prepared solutions homogeneous.
When the flask is inverted and shaken, the solution in the neck of the
flask is not agitated. Only when the flask is set upright again is the
solution drained from the neck and mixed. A good practice is to invert
and shake at least a dozen times to ensure homogeneity.
Volumetric flasks should not be used to prepare solutions of
reagents that can etch glass (such as sodium hydroxide and
hydrofluoric acid), since if the glass is etched, its accurate calibration
is lost. Volumetric flasks should not be used for storing solutions.
Their purpose is to prepare solutions accurately. If they are used for
storage, then they are not available for their intended purpose.
Finally, volumetric flasks should not be used to contain solutions
when heating or performing other tasks for which their accurate
calibration serves no useful purpose. There are plenty of other glass
vessels to perform these functions.
2.3.2 The Pipet
As indicated above, most pipets are pieces of glassware that are

designed to deliver (TD) the indicated volume. Pipets come in a
variety of sizes and shapes. The most common is probably the
"volumetric" or "transfer" pipet shown in Figure 2.6. This pipet, like
the volumetric flask, has a single calibration line. It can thus be used
in delivering only rather common volumes — 5 , 1 0 , 1 5 mL, etc. The
correct use of a volumetric pipet is outlined in Table 2.1. The sequence
of events is described with less detail as follows. The bottom tip of
the pipet is placed into the solution to be transferred. A pipet bulb
is evacuated, placed over the top of the pipet, and slowly released.
The solution is drawn up into the pipet as a result of the vacuum.
When the level of the solution has risen above the calibration line, the
bulb is quickly removed from the pipet and replaced with the index
finger. Next, the tip of the pipet is removed from the solution and
wiped with a towel. The pressure exerted with the index finger is then
used to adjust the bottom of the meniscus to coincide with the
calibration line. The tip of the pipet is placed into the receiving vessel,
and the finger is released. The solution is thus drained into the
receiving vessel. When the solution has completely drained from the
pipet, the bottom tip should be placed against the wall of the receiving
vessel so that the last bit of solution will drain out. A small drop of
25 mL TD 35 sec
> KIMAX USA 20°C
FIGURE 2.6 A volumetric pipet and the top of a class "A" volumetric pipet
showing the markings described in the text.
the liquid will remain in the pipet at this point. The pipet should be
given a half-turn (twist) and then removed. Under no circumstances
should this last drop in a volumetric pipet be blown out with the bulb. The
volumetric pipet is not calibrated for "blow-out."
For precise work, volumetric pipets that are labeled as Class A have
a certain time in seconds imprinted near the top (see Figure 2.6) which
is the time that should be allowed to elapse from the time the finger
is released until the pipet is given the half-turn and removed. The
reason for this is that the film of solution adhering to the inner walls
will continue to slowly run down with time, and the length of time
one waits to terminate the delivery thus becomes important. The
intent with Class A pipets, then, is to take this "run-down" time into
account by terminating the delivery in the specified time. After this
specified time has elapsed, the pipet is touched to the wall of the
receiving flask, given the half-turn, and removed.
Several additional styles of pipets other than the volumetric pipet
are in common use. These are shown in Figure 2.7. Pipets that have
graduation lines, much like a buret, are called "measuring" pipets.
They are used whenever odd volumes are needed. There are two
types of measuring pipets: the Mohr pipet and the serological pipet.
The difference is whether or not the calibration lines stop short of the
tip (Mohr Pipet) or go all the way to the tip (serological pipet). The
serological pipet is better in the sense that the meniscus need be read
only once since the solution can be allowed to drain completely out.
In this case, the last drop of solution is blown out with the pipet bulb.
With the Mohr pipet, however, the meniscus must be read twice: once
before the delivery and again after the delivery is complete. The
Table 2.1 The Sequence of Events Involved in Transferring a Volume of

Solution with a Volumetric Pipet (See Text for More Discussion)
Step 1. If the outside of the pipet is wet, dry it with a paper towel first, especially
the tip.
Step 2. Evacuate the pipet bulb (by squeezing).
Step 3. Seat the bulb opening over the top opening of the pipet.
Step 4. If the pipet is wet on the inside with a foreign liquid, immerse the tip of
the pipet into the solution to be delivered while simultaneously releasing
the bulb slightly to immediately draw the solution in to about half the
pipet's capacity. Empty by inverting and draining into a sink through
the top. Repeat this rinsing step at least three times. If the inside of the
pipet is dry, proceed with Step 5.
Step 5. Fill the pipet to well past the calibration line by releasing the squeezing
pressure, as in Step 4. Reevacuate the bulb if necessary.
Step 6. Quickly remove the bulb and seal the top of the pipet with the index
finger.
Step 7. Keeping the index finger in place, remove the tip from the solution and
wipe with a towel. To avoid contamination from the towel, tilt the pipet
to a 45° angle so that a small volume of air is drawn into the tip before
wiping.
Step 8. Slowly release the finger to adjust the meniscus to the calibration line.
Step 9. Touch the tip to the outside of the receiving vessel so as to remove a drop
of the solution that may be suspended there.
Step 10. Place the tip into the mouth of the receiving vessel and completely release
the finger.
Step 11. When the draining is complete, touch the tip to the inside wall of the
vessel and give it a half-twist.
solution flow out of the pipet must be halted at the correct calibration
line, and the error associated with reading a meniscus is thus doubled.
The delivery of 4.62 mL, for example, is done as in Figure 2.8a with
a Mohr pipet, but as in Figure 2.8b, it is with a serological pipet.
It should be stressed that with the serological pipet, every last trace
of solution capable of being blown out must end up inside the
receiving vessel. Some analysts find that this is more difficult and
perhaps introduces more error than reading the meniscus twice with
a Mohr pipet. For this reason, these analysts prefer to use a serological
pipet as if it were a Mohr pipet. It is really a matter of personal
preference. A double or single frosted ring circumscribing the top of
the pipet (above the top graduation line) indicates the pipet is
calibrated for blow-out. "Disposable" pipets are most often of the
serological blow-out type. They are termed disposable because the
calibration lines are not necessarily permanently affixed to the outside
wall of the pipet. The calibration process is thus less expensive, re
sulting in a less expensive product which can be discarded after use.
One of two pipets that is calibrated "TC" is called the lambda pipet
A B C D E F
FIGURE 2.7 Some types of pipets: (A) volumetric, (B) Mohr, (C) serological, (D)
Ostwald-Folin, (E) duopette, and (F) lambda.
(Figure 2.7). Such a pipet is used to transfer unusually viscous solu

tions such as syrups, blood, etc. With such solutions, the thin film
remaining inside would represent a significant nontransferred vol
ume which translates into a significant error by normal "TD" stan
dards. With the lambda pipet, the calibration line is affixed at the
factory so that every trace of solution contained within is transferred
by flushing the solution out with a suitable solvent. Thus, the pipetted
volume is contained within and then quantitatively flushed out. Such
a procedure would actually be acceptable with any "TC" glassware,
including the volumetric flask. Obviously, diluting the solution in the
transfer process must not adversely affect the experiment. Some pi
pets have both a TC line and a TD line and are called "duopettes"
(Figure 2.7). Ostwald-Folin pipets (Figure 2.7) have a single line but
are calibrated for blow-out.
2.3.3 Pipetting Devices
In addition to the several designs of pipets just described, various

pipetting "devices" for measuring solution volumes from 0 to 1 mL
jft
000001000201020200020201
zz =
JL
Bvi
JL
w b
LI j|i
A 1 t
m 1 J4
$
V
F IG U R E 2.8 The delivery of 4.62 mL of solution (a) with a Mohr pipet (the
meniscus is read twice) and (b) with a serological pipet (the menis
cus is read once [at 0.38] and the solution blown out).
have been invented for laboratory use and have become very popular.
These "pipettors" employ a bulb concealed within a plastic fabricated
body, a spring loaded push-button at the top, and a nozzle at the
bottom for accepting a plastic disposable tip. They may be fabricated
for either single or variable volumes. In the latter case, a ratchet-like
device with a digital volume scale is present below the push-button.
The desired volume may be "dialed in" prior to use. An example is
shown in Figure 2.9.
2.3.4 The Buret
The buret has some unique attributes and uses. One could call it
a specialized graduated cylinder, having graduation lines that in
crease from top to bottom, with a usual accuracy of ± 0.01 mL, and
a stopcock at the bottom for dispensing a solution. There are some
FIGURE 2.9 An Eppendorf pipettor with variable volume capacity.
variations in the type of stopcock that warrant some discussion. The

stopcock itself, as well as the barrel into which it fits, can be made
of either glass or Teflon™.’ Some burets have an all glass arrangement,
some have a Teflon stopcock and a glass barrel, and some have the
entire system made of Teflon. The three types are pictured in Figure
2 .10.
When the all-glass system is used, the stopcock needs to be
lubricated so it will turn with ease in the barrel. There are a number
of greases on the market for this purpose. Of course, the grease must
be inert to chemical attack by the solution to be dispensed. Also, the
amount of grease used should be carefully limited so that excess
grease does not pass through the stopcock and plug the tip of the
buret. (Any material stuck in a buret tip can usually be dislodged with
a fine wire inserted from the bottom when the stopcock is open and
the buret full of solution.) The Teflon stopcocks are free of this lubri
cation problem. The only disadvantage is that the Teflon can become
deformed, and this can cause leakage.
The correct way to position one's hands to turn the buret's stopcock
during a titration is shown in Figure 2.11. The natural tendency with
this positioning is to pull the stopcock in as it is turned. This will
prevent the stopcock from being pulled out, causing the titrant to
bypass the stopcock. The other hand is free to swirl the flask as shown.
2.3.5 Cleaning and Storing Procedures
The use of clean glassware is of utmost importance when doing a

chemical analysis. In addition to the obvious need of keeping the
solution free of contaminants, the walls of the vessels, particularly the
transfer vessels (burets and pipets), must be cleaned so the solution
will flow freely and not "bead up" on the wall as the transfer is
Registered Trademark of E.I. DuPont Nemours and Company, Inc., Wilmington, DE.
FIGURE 2.10 The three types of stopcocks: (left) Teflon stopcock, Teflon barrel,
(middle) Teflon stopcock, glass barrel, and (right) glass stopcock,
glass barrel. (From Kenkel, J., Analytical Chemistry for Technicians,
Lewis Publishers, Inc., Chelsea, MI, 1988. With permission.)
performed. If the solution beads up, it is obvious that the pipet or

buret is not delivering the volume of solution that one intends to
deliver. It also means that there is a film of grease on the wall which
could introduce contaminants. The analyst should examine, clean,
and reexamine his/her glassware in advance so that the free flow of
solution down the inside of the glassware is observed. For the
volumetric flask, at least the neck must be cleaned in this manner so
as to ensure a well-formed meniscus.
Of course, the next question is "What cleaning procedures are
used?" For most cleaning requirements, ordinary soap and water
used with a brush where possible is sufficient. A commercial labo
ratory soap called Alconox™ is one product often used. Other excel
lent phosphate-free detergents are available.* With burets, a cylindri-
* Registered trademark of Alconox, Inc., New York, NY.

FIGURE 2.11 The correct way to position one's hands for a titration. (From
Kenkel, J., Analytical Chemistry for Technicians, Lewis Publishers,
Inc., Chelsea, MI, 1988. With permission.)
cal brush with a long handle (buret brush) is used to scrub the inner
wall. With flasks, a bottle or test tube brush is used to clean the neck.
Also, there are special bent brushes available to contact and scrub the
inside of the base of the flask.
Pipets pose a special problem. Brushes cannot be used because of
the shape of some pipets and the narrowness of the openings. In this
case, if soap is to be used, one must resort to soaking with a warm
soapy water solution for a period of time proportional to the severity
of the particular cleaning problem. Commercial soaking and washing
units are available for this latter technique. Soap tablets are manufac
tured for such units and are easy to use.
For pipets and for difficult cleaning problems for other pieces of
glassware, special cleaning solutions, which chemically break down
Table 2.2 Recipes for Preparing Glassware Cleaning Solutions
Chromic acid, Method 1 Dissolve 90-100 g of sodium or potassium

dichromate in 450 mL of water and add 80 mL of
concentrated sulfuric acid. The solution will become
a semisolid red mass. Add just enough sulfuric acid
to dissolve this mass.
Chromic acid, Method 2 Add the contents of a 25-mL bottle of Monostat
"Chromerge" (available from chemical products
suppliers) to a standard 9-lb bottle of concentrated
sulfuric acid. Add approximately 5 mL at a time
and shake well after each addition.
Alcoholic KOH Dissolve 105 g of KOH in 120 mL of water. Add 1 L
of 95% ethanol (or propanol) and shake well. This
solution will etch glass to a small extent and
therefore should not be left in contact with
volumetric glassware or ground glass joints for more
than about 15 min at a time.
Reprinted from Kenkel, ]., Analytical Chemistry for Technicians, Lewis Publishers, Inc.,
Chelsea, MI, 1988. With permission.
greasy films through soaking for a period of time, are used. One
commonly-used such cleaning solution is chromic acid. This is a
solution of concentrated sulfuric acid and potassium dichromate.
Another is a solution of potassium hydroxide in ethanol or propanol.
Recipes for preparing these solutions are given in Table 2 .2 . These are
very tough on serious cleaning problems.
Safety should be stressed in the use of these cleaning solutions. The
highly corrosive nature of the ingredients makes it imperative to
prevent spills and splashes on one's person. In addition to normal
safety gear (safety glasses, eye wash, shower, first aid kit), one should
also have solutions of weak acids (e.g., acetic acid, citric) and weak
bases (e.g., sodium carbonate) handy to neutralize spills quickly. It
is highly recommended that use of these cleaning solutions be re
stricted to a good fume hood. Spent cleaning solutions should not be
poured down the drain, but disposed of like any other hazardous
waste.
Once the glassware has been cleaned (by whatever method), one
should also take steps to keep it clean. One technique is to rinse the
items thoroughly with distilled water and then dry them in an oven.
Following this, they are cooled and stored in a drawer. For shorter
time periods, it may be convenient to store them in a soaker under
distilled water. This prevents their possible recontamination during
dry storage. When attempting to use a soaker-stored pipet or buret,
however, it must be remembered that the thin film of water is present

on inner walls, and this must be removed by rinsing with the solution
to be transferred, being careful not to contaminate the solution in the
process.
2.4 REAGENTS AND M ETHODS FOR

SAMPLE PREPARATION
For the vast majority of chemical analysis schemes, the samples to

be analyzed must be dissolved in a liquid solution. Many samples,
when they first arrive in the laboratory, are either undissolved solids
or liquids from which the analyte must be removed and dissolved in
another solvent before proceeding. The first operation often
performed on a sample is therefore either total dissolution or an
extraction. There is a broad array of substances available, and a few
special techniques used for these operations and their use depends
on the nature of the sample and whether complete dissolution or an
extraction is needed. We now present a discussion of some of the more
common laboratory reagents and extraction techniques and examples
of their use.
2.4.1 Water, Concentrated Acids, and Bases
Water — It should come as no surprise that ordinary water can be

an excellent solvent for many samples. Due to its extremely polar
nature, water will dissolve most substances of likewise polar or ionic
nature. Obviously, then, when samples are composed solely of ionic
salts, water would be an excellent choice. An example might be the
analysis of a commercial table salt for sodium iodide content. A list
of "solubility rules" for ionic compounds in water can be found in
Table 2.3.
Hydrochloric Acid — Strong acids are used frequently for the
purpose of sample dissolution when water will not do the job. One
of these is hydrochloric acid, HC1. Concentrated HC1 is actually a
saturated solution of hydrogen chloride gas, and the fumes are very
pungent. Such a solution is 38.0% HC1 and about 12 M. Hydrochloric
Table 2.3 Solubility Rules for Some Common Inorganic Compounds in Water
Compound Class Soluble? Exceptions
Nitrates Yes None

Acetates Yes Silver acetate sparingly
soluble
Chlorides Yes Chlorides of Ag, Pb,
and Hg are insoluble
Sulfates Yes Sulfates of Ba, Pb are
insoluble; sulfates of
Ag, Hg, and Ca are
slightly soluble
Carbonates No Carbonates of Na, K,
and NH4 are soluble
Phosphates No Phosphates of Na, K,
and NH4 are soluble
Chromates No Chromates of Na, K,
NH4, and Mg are
soluble
Hydroxides No Hydroxides of Na, K,
NH4 are soluble;
hydroxides of Ba, Ca,
and Sr are slightly
soluble
Sulfides No Sulfides of Na, K,
NH4, Ca, Mg, and Ba
are soluble
Sodium salts Yes Some rare exceptions
Potassium salts Yes Some rare exceptions
Ammonium salts Yes Some rare exceptions
Silver salts No Silver nitrate,
perchlorate; silver
acetate and sulfate are
sparingly soluble
acid solutions are used especially for dissolving metals, metal oxides,
and carbonates not ordinarily dissolved by water. Examples are iron
and zinc metals, iron oxide ore, and the metal carbonates of which
the scales in boilers and humidifiers are composed. Being a strong
acid, it is very toxic and must be handled with care.
Sulfuric Acid — An acid that is considered a stronger acid than
HC1 in many respects is sulfuric acid, H2 S 0 4. When sulfuric acid
contacts clothing, paper, etc., one can see an almost instantaneous
reaction — paper towels turn black and disintegrate, and clothing
fibers become weak and holes readily form. Concentrated sulfuric
acid is about 96% H2 SO 4 (the remainder being water) or about 18 M
and is a clear, colorless, syrupy, dense liquid. It reacts violently with
water, evolving much heat, and so water solutions of sulfuric acid

must be prepared cautiously, often to include a means of cooling the
container. Its sample dissolution application is limited mostly to
organic material, such as vegetable plants. It is the solvent of choice
for the Kjeldahl analysis (Chapter 4) for such materials as grains and
products of grain processing. It is also used to dissolve aluminum and
titanium oxides on airplane parts.
Nitric Acid — Another acid that has significant application is nitric
acid, H N 03. This acid is also very dangerous and corrosive and is
aptly referred to as an oxidizing acid. This means that a reaction other
than hydrogen gas displacement (as with HC1) occurs when it con
tacts metals. Frequently, oxides of nitrogen form in such a reaction,
and noxious brown, white, and colorless gases are evolved. Concen
trated HNO3 is 70% HNO 3 (16 M) and is used for applications where
a strong acid with additional oxidizing power is needed. These in
clude metals such as silver and copper, as well as organic materials
such as in a wastewater sample. Nitric acid will turn skin yellow after
only a few seconds of contact.
Hydrofluoric Acid — An acid that has some very useful and
specific applications, but is also very dangerous, is hydrofluoric acid,
HF. This acid reacts with skin in a way that is not noticeable at first,
but becomes quite serious if left in contact for a period of time. It has
been known to be especially serious if trapped against the skin and
after diffusing under fingernails. Treatment of this is difficult and
painful. Concentrated HF is about 50% HF (26 M). It is an excellent
solvent for silica (SiC^l-based materials such as sand, rocks, and glass.
It can also be used for stainless steel alloys. Since it dissolves glass,
it must be stored in plastic containers. This is also true for low pH
solutions of fluoride salts.
Perchloric Acid — Another important acid for sample preparation
and dissolution is perchloric acid, HCIO4 . It is an oxidizing acid like
HNO 3 , but is considered to be even more powerful in that regard
when hot and concentrated. It is most useful for more difficult organic
samples, such as leathers and rubbers. It can also be used for stainless
steels and other more stable alloys. Commercial HCIO4 is 72% HCIO4
(12 M). It is a very dangerous acid and should only be used in a fume
hood designed for the collection of its vapors. Contact with alcohols,
including polymeric alcohols, such as cellulose, and other oxidizable
materials, should be avoided due to the potential for explosions.
"Aqua Regia" — An acid mixture that is prepared by mixing one

part concentrated HNO 3 with three parts concentrated HC1 is called
"aqua regia." This mixture is among the most powerful dissolving
agents known. It will dissolve the very noble metals (gold and plati
num), as well as the most stable of alloys.
Sodium Hydroxide and Ammonium Hydroxide — Bases such as
sodium hydroxide and ammonium hydroxide, while quite commonly
used in the analytical laboratory, are used sparingly in sample prepa
ration. The high pH of such solutions often does not help with sample
dissolution and in fact may hinder it. A case in point is metals. While
many metals are quite soluble at low pH values, many form highly
insoluble hydroxides at basic pH values. Thus, bases cannot be used
to dissolve samples containing metals (unless it is an extraction in
which the metal is not the analyte — see following subsection). How
ever, a hydroxide precipitate may be a desired product, such as in a
scheme for the separation of a metal from an analyte by its precipi
tation. When high pH values are desirable, both sodium hydroxide,
a strong base, and ammonium hydroxide, a weak base, have a use
fulness.
2.4.2 Sample Preparation By Extraction
In cases in which it is not practical or necessary to dissolve an entire

sample or in which it is easy to affect a "separation" of the analyte
by selective dissolution, an extraction procedure is used. This
technique is described in some detail in Chapter 8 .
Briefly, the sample can be either a solution (typically a water
solution) of the analyte or a solid material, e.g., soil. When the sample
is a water solution, solvent extraction refers to an experiment in which
an organic solvent, immiscible with the sample, is brought into inti
mate contact with the sample (in a "separatory funnel" — see Chapter
8 ) in such a way that the substance of interest is at least partially
removed (extracted) from the water and dissolved in the solvent.

When the sample is a solid, the extracting solvent is often an
aqueous solution of an inorganic compound, such an acid, although
it can also be an organic solvent. Here again, the sample is brought
into intimate contact with the extracting solvent by shaking in the
same container (usually after grinding to a fine powder) or by
repeated contact with fresh solvent in an apparatus called a Soxhlet

extractor (Chapter 8 ). With both of these methods, the analyte is at
least partially separated from the sample and dissolved in this
solvent, while most other materials remain undissolved. This method
is acceptable since the undissolved materials play no role at all in the
analysis and, in fact, may contain interfering substances which are
best left undissolved. The analysis can then proceed on the sample
"extract" as with any other sample solution. Please refer to Chapter
8 for more details of the extraction technique.
Following is a discussion of the properties of common organic

solvents used in the laboratory for extractions and other applications.
n-Hexane — n-Hexane is nonpolar and an alkane, C 6 H14. Its

solubility in water is virtually nil. It is less dense than water (density
= 0.66 g/mL); thus it would be the top layer in a separatory funnel
with a water solution. It is obviously a poor solvent for polar
compounds, but is very good for extracting traces of nonpolar solutes
in water samples. It is used as a solvent for UV spectrophotometry.
It is highly flammable and has a low toxicity level. It can be used as
the mobile phase in normal phase chromatography.
Acetonitrile — Acetonitrile is a polar solvent, CH^C^N. It is com
pletely miscible with water. It is used for the extraction of polar
components of solid samples, such as the extractions of chlorinated
pesticide residues from vegetables, and also as a mobile phase for
reverse phase chromatography. It can be used as a solvent for UV and
NMR work. It is flammable and toxic and has an unpleasant odor.
Breathing acetonitrile vapors should be avoided.
Methylene Chloride — This solvent is a somewhat polar solvent
also known as dichloromethane, CH 2 CI2 . Its solubility in water is
about 2%. It is denser than water (density = 1.33 g/mL); thus it would
be the bottom layer when used with a water solution in a separatory
funnel. It may form an emulsion when shaken in a separatory funnel
with water solutions. It is used for extractions, as well as for infrared
(IR) and NMR solvents. It is not flammable and is considered to have
a low toxicity level.
Acetone — Acetone (CH3 COCH 3 ) is a volatile, highly flammable
liquid that has intermediate polarity and is completely miscible with
water. It is used as a solvent for both polar and nonpolar substances,
and is therefore used often as a cleaning solvent for glassware and
as a degreaser. Its high volatility makes it useful for rapid drying of
glassware. It can be used as a solvent for NMR work. It is occasionally

used to extract components from solid samples. It is considered
nontoxic, but prolonged breathing of vapors should be avoided.
Benzene — Benzene (C6 H6) is a nonpolar aromatic liquid insoluble
in water. It can be used as a solvent in NMR work. For extraction
work, it has been used for extraction of nonpolar components of water
samples, but is prone to emulsion formation. Its density is 0.87 g/mL.
It is highly flammable and toxic. It is especially noted as a cancer-
causing carcinogen. Breathing of vapors and skin contact should be
avoided. Due to its toxicity, it has been replaced in most laboratory
applications with toluene.
Toluene — Toluene (C6 H5 -C H 3) is also a nonpolar aromatic liquid
insoluble in water. It is flammable, but less toxic than benzene. Its
density is 0.87 g/mL and is slightly soluble in water (0.47 g/L). It is
sometimes used as a mobile phase in normal phase liquid chroma
tography and as an extraction solvent for nonpolar sample compo
nents.
M ethanol — Methanol (CH 3 -O H ) is a polar organic solvent
completely miscible with water and most other solvents. It is a
flammable and poisonous liquid. It is useful as a solvent for UV and
NMR work, as well as a mobile phase for reverse phase liquid
chromatography. Like acetone, it is sometimes used to quick-dry
laboratory glassware, since it is volatile. It is poisonous when
ingested, but inhalation should also be avoided.
D iethyl Ether — Diethyl ether (CH 3 CH 2 -O -C H 2 CH 3 ), also
frequently referred to as "ethyl ether" and as "ether," is a mostly
nonpolar organic liquid that is highly volatile and extremely
flammable, a dangerous combination. Also, explosive peroxides form
with time. Precautions regarding storage to discourage peroxide
formation include storage in metal containers in explosion-proof
refrigerators. It should be disposed of after about nine months of
storage.
Ether is only slightly soluble (75 g/L) in water. Given its slight
polarity, it is useful to extract nonpolar components from water. Its
density is 0.71 g/mL. Since it is volatile, it is easily evaporated from
extraction fractions.
C hloroform — Chloroform (trichlorom ethane, CHCI3 ) is a
nonflammable organic liquid with low miscibility with water ( 1 0 g/
L). It is useful as an extracting liquid for water samples; as a solvent
for UV, IR, and NMR work; and as a mobile phase for HPLC. It is
very toxic and should be avoided when another solvent would do as

well. Vapors should not be inhaled and contact with the skin should
be avoided.
2.4.3 Fusion
For extremely difficult samples, a method called "fusion" may be

employed. Fusion is the dissolving of a sample using a molten
inorganic salt generally called a "flux." This flux dissolves the sample
and, upon cooling, results in a solid mass that is then soluble in a
liquid reagent. The dissolving action is mostly due to the extremely
high temperatures (usually 300 to 1000°C) required to render most
inorganic salts molten.
Additional problems arise within fusion methods, however. One
is the fact that the flux must be present in a fairly large quantity in
order to be successful. The measurement of the analyte must not be
affected by this large quantity. Also, while a flux may be an excellent
solvent for difficult samples, it will also dissolve the container to some
extent, creating contamination problems. Platinum crucibles are
com m only used, but nickel, gold, and porcelain have been
successfully used for some applications.
Probably the most common flux material is sodium carbonate. It
may be used by itself (melting point 851 °C) or in combination with
other compounds, such as oxidizing agents (nitrates, chlorates, and
peroxides). These may be used to dissolve silicates and silica-based
samples, as well as samples containing alumina (AI2 O3 ).
For dissolving particularly difficult metal oxides, the acidic flux
potassium pyrosulfate (K2 S2 O 7 ) may be used. The required tempera
ture for this flux is about 400°C.
2.4.4 Reagent Quality
It is of utmost and obvious importance in an analytical laboratory

to be able to trust all reagents used for both the sample preparation
schemes, which have been outlined in this section, as well as for other
analytical preparations. A stringent quality assurance program for
such reagents is necessary. This means that the quality of all inven
toried, and newly purchased chemicals as well, must be assured. It

also means that the integrity of the various solutions prepared and
samples gathered for analytical purposes in the laboratory must be
maintained.
First, chemicals purchased and/or stored will have a designation
of their purity on the label. This designation will likely be one of the
following.*
Primary Standard This grade designates specially manufactured

analytical reagent of exceptional purity for
standardizing solutions and preparing refer
ence standards.
Reagent (ACS) Maximum limits of purity for most commonly
used reagents (mostly inorganic) have been
established by the Committee on Analytical
Reagents of the American Chemical Society.
Whenever a specification exists and a product
is produced in conformity with it, the bottle is
so labeled.
Reagent When the American Chemical Society has not
developed specifications for a specific reagent,
the m anufacturer establishes its own stan
dards, and the maximum limits of allowable
impurities are shown on the labels of these
reagents.
C.P. "Chemically Pure" grades of chemicals are
offered by m anufacturers. They m eet or
exceed U.S.P. or N.F. requirements, but are
lower grade than "R eagent" or "Reagent,
ACS" chemicals.
U.S.P. Chemicals labeled U.S.P. meet the require
ments of the U.S. Pharmacopeia. Generally of
interest to the pharm aceutical profession,
these specifications may not be adequate for
reagent use. This designation and the N.F.
designation are now the same.
N.F. Chemicals labeled N.F. meet the requirements
of the National Formulary. Chemicals ordered
with an N.F. label may not be useful for re-
These designations have been reprinted, in part, from the Modern Chemical Technology
Guidebook, Revised Edition, 1972, 157. With permission. Copyright by the American
Chemical Society, Washington, D.C.
agents. It will be necessary to check the N a

tional Formulary in each case. This designa
tion and the U.S.P. designation are now the
same.
Practical This grade designates chemicals of sufficiently
high quality to be suitable for use in some
syntheses. O rganic chem icals of practical
g ra d e m ay co n tain sm all am o u n ts of
intermediates, isomers, or homologs.
Purified This is a grade of chemical for which care has
been exercised by the manufacturer to offer a
product that is physically clean and of good
quality, but not meeting Reagent ACS, Re
agent, U.S.P., or C.P. standards.
Technical This is a grade of chemical generally suitable
for industrial use. Purity is not specified and
is generally determined by on-site analysis.
Spectro Grade This is a designation for organic solvents
which have been prepared for use in UV or IR
spectroscopy without further purification.
Many of these also conform to Reagent or
Reagent ACS standards.
HPLC Grade This is a designation for organic solvents
which have been specifically prepared for use
as HPLC mobile phases.
Second, the integrity of inventoried chemicals, or chemicals that

have been on the shelf for a period of time, can be determined. This
determination of integrity, or shelf-life, can involve proper labeling
at the manufacturing plant, various reference sources, such as the
Merck Index, or various laboratory tests to determine purity.
Third, samples and solutions gathered or prepared by laboratory
personnel must also be properly labeled at the time of sampling or
preparation. This means that the label must include the name of the
sample or the chemical(s) dissolved and the date the sample was
gathered or solution prepared. In addition, the laboratory should
maintain a notebook of all samplings and preparations giving details
of each sampling or preparation and the references which called for
the particular chemicals or techniques used. This would allow
tracking the reagent or sample in the event an error is suspected.
We made the point in Chapter 1 that "an analysis is only as good
as the sample" used. We can now extend this statement and say that
an analysis is only as good as the sample and the reagents that are used.
2.5 LABORATORY SAFETY
The analytical chemistry laboratory is a very safe place to work. The

reason for this is that the dangers associated with contact with haz
ardous chemicals, flames, etc. are very well documented. As a result,
laboratories are constructed and procedures are carried with these
dangers in mind. Hazardous chemical fumes, for example, are vented
into the outdoor atmosphere by fume hoods. Safety showers for
diluting spills of concentrated acids on clothing are now common
place. Eye wash stations are strategically located for the immediate
washing of one's eyes in the event of accidental contact of a hazardous
chemical with the eyes. Fire blankets, extinguishers, and sprinkler
systems are also located in and around analytical laboratories for
immediately extinguishing flames and fires. A variety of safety gear,
such as safety glasses, aprons, and shields, is available. There is never
a good excuse for personal injury in a well-equipped laboratory where
well-informed analysts are working.
While the pieces of equipment mentioned above are now common
place, it remains for the analysts to be well informed of potential
dangers and of appropriate safety measures. Chemistry analysts
should be well versed in the procedures used in the event of an
accident at his/her worksite. These include (1) total familiarity with
the safety features of the site, meaning location of the eye wash
stations, safety showers, fire extinguishers; (2 ) total familiarity with
the employer's laboratory safety manuals, and if one does not exist,
it should be the analyst's responsibility to urge management to write
one; and (3) knowledge of various reference books devoted to or at
least emphasizing safety.*
An important development with regard to safety is the requirement
of chemical manufacturers to provide Material Safety Data Sheets
(MSDS) with all chemicals sold. The MSDS is a piece of paper that
provides essential safety information concerning the chemical, such,
as reactivity properties, fire hazards, identification information, what
to do in case of a spill, what special equipment is required for han
dling, and much more. MSDSs should be stored in an easily accessible
location known to all laboratory workers.
See, for example, Shugar, G.J. and Ballinger, J.T., Chemical Technicians' Ready Reference
Handbook, 3rd ed., McGraw-Hill, New York, 1990, and references contained therein.
Laboratory work can be very important work indeed, but nothing

is so important that a person's safety is jeopardized.
C hapter 3
S olution P reparation
3.1 INTRODUCTION
One activity that is common to all analytical laboratories is solution

preparation. Solutions must be prepared no matter what analytical method
is employed or sample analyzed. This includes all wet chemical methods,
as well as instrumental methods. Standard solutions must be prepared
in order to generate a standard curve in instrumental analysis. Solutions
must be prepared and standardized in order to perform titrations and
to calculate the results of titrations. Buffer solutions must be prepared in
order to provide a constant pH in solutions that require it. Solutions of
reagents required for the chemistry of a given system are often needed
for addition to the system at an appropriate time. Solutions are even often
prepared for safety measures in the event of the need to neutralize a
chemical spill. The list seems to be endless.
The accuracy with which the chemist prepares a solution depends, of
course, on the application; some require the highest degree of accuracy,
while others require only minimal accuracy (Chapter 1 ). If it is required
that a solution concentration be of high accuracy, such as for the titrant
in a titrimetric method or for a solution used to help generate a standard
curve in an instrumental method, and if a standardization procedure
47
(Chapter 4) is not planned, then its concentration must be known accu

rately directly through its preparation. This means that an analytical
balance must be used to weigh the chemical (if the solute is a pure solid),
an accurate pipet must be used to measure the volume of a solution of
the chemical (if the chemical is already in solution), and a volumetric flask
must be used to measure the total solution volume in order to know that
just the correct amount of solvent has been added. (See Chapter 2 for a
description of the volumetric flask and its use.) However, if the solution
concentration need not be known accurately, then any balance, or any
type of liquid transfer glassware if the solute is already in solution, would
work, and a volumetric flask need not be used.
With these thoughts in mind, we proceed with a series of discussions
giving specific instructions as to calculations and methods used in a
variety of situations involving various concentration units.
3.2 DILUTION
One method of preparing solutions, as alluded to in the last section,

is by dilution, i.e., the solute is already in solution, but a lower concentration
of it is called for. Regardless of the units with which the concentration
(or volume) is expressed, the general calculation and method of preparation
is the same. The calculation uses the following formula:
C b * V b = C a x Va (3.1)
in which "C" refers to "concentration," "V" is "volume," "B" refers to

"before dilution," and "A" to "after dilution." The concentration of the
solution before dilution times its volume is equal to the concentration of
the solution after dilution times its volume. The concentrations before and
after dilution must be known, and the volume after dilution must be
known. The volume before dilution is being calculated in order to know
how much of the more concentrated solution to measure out and dilute.
The units of concentration must be the same on both sides of the equation
(e.g., percent, molarity, normality, etc.). The units of volume are also the
same on both sides.
Example 1
How would you prepare 500 mL of a 2.0% NaCl solution from one
that is 1 0 %?
Solution Preparation 49
Solution 1
1 0 x V B = 2 .0 x 5 0 0
2 .0 x 5 0 0
= 100 m L
100 mL of the 10% solution are measured into a 500 mL container, and
water is added to the 500 mL line.
A common example of a dilution problem is encountered when wanting

to dilute commercially available concentrated acids, such as concentrated
HC1 or concentrated H2 S 0 4. When a particular molarity of the acid is
called for, it is necessary to know the molarity of the concentrated reagent
in order to use the dilution formula. Such molarities are rarely displayed
on the labels of these acids. They are listed in Table 3.1.
There can be considerable confusion when a dilution procedure in the
form of a "ratio" is called for. Apparently, many clinical and biological
applications call for a 1:10 dilution or a 2:5 dilution, etc. There is a
difference in using the word "dilution" as opposed to the word "ratio."
A 1:10 dilution is meant to mean 1 part solute added to make 10 parts
total solution. A 1:10 ratio, or other synonym, is meant to mean 1 part
solute added to 1 0 parts of solvent.
Additional dilution problems, those in which the concentrations before
and after dilution have different units, and others are variously cited in
the next four sections.
3.3 PERCENT
The concept of percent is well known. For solution preparation in an

analytical laboratory, however, this concept can be confusing. The kind
of units used for the numbers can be variable. Not only is it common for
a solution concentration to be expressed as a volume/volume percent
(the numerator and denominator both have volume units) or a weight/
weight percent (the numerator and denominator both have weight units),
but it is also common that the numerator has weight units, and the
Table 3.1 Molarities of Some Common Commercially Available Concentrated

Acid and Base Solutions
Acid or Base Molarity
Acetic acid (HC2H30 2) 17

Ammonium hydroxide (NH4OH) 15
Hydrobromic acid (HBr) 9
Hydrochloric acid (HC1) 12
Hydrofluoric acid (HF) 26
Nitric acid (H N 03) 16
Perchloric acid (HC104) 12
Phosphoric acid (H3P 0 4) 15
Sulfuric acid (H2S 0 4) 18
denominator has volume units (so-called weight/volume percent).
, , , , x volume of solute .
volume/volume (v/v) percent = ----------------------------x 100 (3.2)
volume of solution
•, , •. . , , weight of solute
weight/weight (w/w) percent = x 100 (3 3)
weight of solution '
•. , . , , s weight o f solute ..
weight/volume (w /v) percent = ---------------------------x 100 (3.4)
volume of solution
Note that the denominator in each case is the quantity of solution and
not solvent. Also, the units in Equations 3.2 and 3.3 can be any unit of
weight or volume as long as they are the same in both numerator and
denominator. The units in Equation 3.4 must be grams and milliliters,
kilograms and liters, or milligrams and microliters, etc.
3.3.1 Volume/Volume (v/v)
A volume/volume percent is common when the solute is a liquid. The

reason is simply that volumes of liquids are easier to measure than
weights. They can be pipetted. The following formula is useful.
volume to measure = (% desired/100) x volume desired (3.5)
The solute is measured (pipetted, if accuracy is important) into the

container (a volumetric flask, if accuracy is important), and the solvent

is added such that the total volume is the volume of solution desired.
Example 2
How would you prepare 500 mL of a 15% (v/v) ethanol solution in
water?
Solution 2
volume to measure = (15/100) x 500
= 75 mL
75 mL of ethanol are measured into the vessel to contain the solution.

Water is then added to the 500 mL level.
3.3.2 Weight/Weight (w/w)
A weight/weight percent is common when the solute is a solid, since

a solid is easily weighed, but may also be encountered when the solute
is a pure liquid. If the solute is a pure liquid, its volume is more easily
measured, and it would be beneficial to calculate the volume corresponding
to the weight (if the density is known). In addition, the weight of the total
solution desired must be known. Thus, in the final stage of preparation,
the solvent must be added such that the total weight is measured and
matched to the weight of solution desired. Alternatively, if the density
of the final solution is known, its volume may be measured instead. The
following formula is useful.
weight to measure = (% desired/100) x solution weight desired (3.6)
Example 3
How would you prepare 500 g of a 12% (w/w) solution of NaCl?
Solution 3
weight to measure = (12/100) x 500
= 60 g
60 g of NaCl are weighed, and placed in a container capable of holding

500 g of solution. Solvent is added until the solution weighs 500 g.
If the desired volume of the solution is given rather than the weight,
the weight of the solution can be determined using the known density.
weight of solution = volume of solution x density (3.7)
Example 4
How would you prepare 500 mL of a 1 2 % (w/w) solution of NaCl if
the density of the solution is 1.05 g/mL?
Solution 4
weight of solution = 500 x 1.05 = 525 g
weight to measure = (12/100) x 525 = 63 g
63 g of NaCl are weighed and placed in the vessel to contain the solution.
The solvent is added up to the 500 mL level.
If the density is not known, it may be acceptable to assume that the

density of dilute water solutions is equal to 1 (the density of pure water
is equal to 1 at 4°C) such that the solution weight required (in grams)
is numerically equal to the volume required (in milliliters).
3.3.3 Weight/Volume (w/v)
The concept alluded to in the last paragraph of assuming that the

density of a dilute water solution is equal to 1 likely gave rise to the
weight/volume expression for solution concentration. It is a convenient
concentration expression because the amount of solute to be weighed is
a percentage of the total solution volume rather than total solution weight.
weight to measure = (% desired/100) x solution volume desired (3.8)
Thus, the solute can be weighed, placed in the container, and water added
up to the solution volume desired without regard for the total weight.
This means that the solution does not have to be weighed, nor do we
need to make the possibly invalid assumption that the density is equal
to 1 .
Example 5
How would you prepare 500 mL of a 5.0% (w/v) solution of NaCl?
Solution 5
weight to measure = (5.0/100) x 500

= 25 g
25 g of NaCl are weighed and placed in the container. Water is then added
such the total solution volume is 500 mL.
One source of confusion is in the preparation of weight/volume

solutions from commercial concentrated acids, such as HC1, HNOs, etc.,
which have known weight/weight percent concentrations. The density
(or specific gravity) of the concentrated acid is used in the calculation and
can often be found on the label of the concentrated acid bottle along with
the weight/weight percent. This may be considered a dilution problem
(see section 3.2), but the two concentrations, before dilution and after
dilution, must have the same unit in order to correctly apply the dilution
formula. We must convert weight/weight percent to weight/volume
percent.
w/v% = w/w % x specific gravity (3.9)
We thus will have a weight/volume percent that we can plug into the
dilution formula.
Example 6
How would you prepare 500 mL of a 10.0% (w/v) solution of HC1 from
concentrated HC1, which is 37.0% (w/w)? The specific gravity of the HC1,
from the label on the bottle, is 1.18.
Solution 6
Solution 6
C B(w/v% ) = 3 7 .0 (w /w % )x 1.18
= 43.7% (w/v)
43. 7 % (w /v )x V b = 1 0 . 0 x 500
1 0 .0 x 5 0 0
VB = 114 mL
114 mL of concentrated HC1 are diluted to 500 mL with water.
The percent composition (w/w) and the densities of the common

concentrated acids and bases are given in Table 3.2.
Frequently, it is unclear whether a certain percent concentration called
for in a given method is weight/weight or weight/volume. The analyst
may need to make an educated guess in these instances. Since, for dilute
water solutions, the density is very nearly equal to 1 , the easiest method,
usually the weight/volume, is chosen if accuracy is not important.
3.4 MOLARITY AND FORMALITY
The number of moles of solute dissolved per liter of solution (molarity)

is another common method of expressing concentration. It is thus very
common for an analyst to need to prepare solutions of a particular
molarity, both by dilution and by weighing a pure chemical. In this case,
solutions are referred to as being, for example, 2.0 molar or 2.0 M. The
"M " refers to "molar," and we say that the solution has a molarity of 2.0,
i.e., 2.0 moles dissolved per liter of solution. It is important to recognize
that it is the number of moles dissolved per liter of solution and not per
liter of solvent.
In addition, the terms "formality," "F," and "formal" are often encoun
tered. These terms are meant to precisely describe solutions of ionic salts
since, on an atomic scale, such salts technically do not exist as molecules,
but rather as ions in a three-dimensional array. This is also the reason
Table 3.2 The Densities and Percent Compositions of Some Common

Commercially Available Concentrated Acids and Bases
Acid or Base Density % Composition
Acetic acid (HC2H30 2) 1.05 99.5

Ammonium hydroxide (NH4OH) 0.90 58
Hydrobromic acid (HBr) 1.52 48
Hydrochloric acid (HC1) 1.18 37
Hydrofluoric acid (HF) 1.14 45
Nitric acid (H N 03) 1.42 72
Perchloric acid (H CL04) 1.67 70
Phosphoric acid (H3P 0 4) 1.69 85
Sulfuric acid (H2S 0 4) 1.84 96
that the term "formula weight" is used as opposed to "molecular weight."

Thus, the formula weight is the weight represented by the formula of the
compound, which is also what molecular weight represents for nonionic
compounds. The two are often used interchangeably. In the following
discussions, the terms "molecular weight," "molar," and "M" will be
used exclusively for ionic and nonionic compounds alike.
To prepare a solution of a certain molarity when a pure chemical is
to be measured, the following formula is useful.
grams to measure = L D x MD x MWSQL (3.10)
in which LD refers to liters desired, MD to molarity desired, and MWgoL

to molecular weight of solute. The grams of chemical thus calculated is
weighed out, placed in the container, and water added to dissolve and
dilute to volume.
Example 7
How would you prepare 500 mL of a 0.20 M solution of NaOH from
pure, solid NaOH?
Solution 7
grams to measure = 0.500 x 0.20 x 40.00
= 4 .0 g
The analyst would weigh 4.0 g of NaOH, place it in a 500 mL container,

add water to dissolve the solid, and then dilute to volume.
If the solute is a liquid (somewhat rare), the grams calculated from

Equation 3.10 can be converted to milliliters using the density of the
liquid. In this way, the volume of the liquid can then be measured, rather
than its weight, and it can then be pipetted into the container.
Sometimes it may be necessary to prepare a solution of a certain
molarity by diluting another solution, the concentration of which is known
in, for example, weight/volume percent. As noted earlier (Section 3.2),
however, the dilution formula can only be applied when the two
concentrations have the same units. In this case, the weight/volume
percent can be converted to molarity and the volume to be diluted
calculated in the usual way.
Example 8
How would you prepare 500 mL of a 0.20 M solution of NaOH from
an NaOH solution that is 10% (w/v)?
Solution 8
10% (w /v) = 10 g / 0.1 0 L
= 1 0 0 g /L
w 1 100
Molarity = ----------= 2 .5 M
4 0 .0 0
2 .5 x V B = 0 . 2 0 x 5 0 0
Vn = 40 mL
40 mL of the 10% NaOH would be measured into a 500 mL container

and diluted to volume.
3.5 NORMALITY
Another method of expressing solution concentration is normality.

Normality is the number of equivalents per liter. It is the typical
concentration unit when working within the system in which quantities
of chemicals are expressed as equivalents and equivalent weights. This

system is completely analogous to the system of moles and molecular
weights. That means that in all calculations performed in which the mole
and the molecular weight are used (including solution preparation), the
equivalent and equivalent weight may also be used. Normality is the
number of equivalents of solute dissolved per liter of solution, while
molarity is the number of moles of solute dissolved per liter of solution.
3.5.1 Concept of the Equivalent
To understand normality, we obviously must understand the concept

of the equivalent. An equivalent is almost always either the same as the
mole or it is some fraction of a mole, such as half a mole, a third of a
mole, etc. What that fraction is depends on the reaction in which the
substance takes part. The equivalent of a substance is that part of a mole
which reacts on a one-to-one basis with another substance. If one mole
of one substance reacts with one mole of another substance, it may be
more convenient to work in terms of moles and molarity rather than
equivalents and normality. If, however, one mole of one substance reacts
with, say, two moles of another substance, then, if you want to work with
normality, solution preparation schemes and associated calculations must
involve the equivalent, rather than the mole, and the equivalent weight
rather than the molecular weight.
How does one determine what an equivalent is or what an equivalent
weight is? The answer depends on the kind of reaction in which the
substance of interest takes part. The method for determining the equivalent
weights for add-base reactions, for example, is different from the method
for oxidation-reduction reactions. Let us consider the acid-base case first.
3.5.1a Acid/Base
The equivalent weight of an acid is the molecular weight divided by

the number of hydrogens donated per molecule in the reaction. The
equivalent weight of a base is the molecular weight divided by the
number of hydrogens accepted per molecule in the reaction. (By the
Bronsted-Lowry Theory, an add is a substance that donates hydrogens
in a reaction, while a base is a substance that accepts hydrogens.) Let us
look at a few examples.
HC1 + NaOH -> NaCl + H 20 (3.11)
H2S 0 4 + 2NaOH Na2S 0 4 + 2H 20 (3.12)
2HC1 + Ba(O H )2 -> BaCl2 + 2H 20 (3.13)
H3P 0 4 + 2NaOH Na2H P 0 4 + 2H zO (3.14)
The equivalent weights of both HC1 and NaOH in Equation 3.11 are
equal to their respective molecular weights. One hydrogen is donated in
the case of HC1 (it has only one per molecule), and one hydrogen is
accepted in the case of NaOH (it has only one hydroxide per molecule
to accept hydrogens). The same can be said for the HC1 in Equation 3.13
and the NaOH in Equations 3.12 and 3.14. Remember, the definitions state
that it is the hydrogens donated or accepted per molecule. The HjSO^
however, has two hydrogens to be donated per molecule, and indeed
they both get donated, since two water molecules were formed and there
is no longer any hydrogens remaining with the sulfate on the right side.
The equivalent weight of sulfuric acid in this case is the molecular weight
divided by 2 . Similarly, the Ba(OH) 2 in Equation 3.13 has two hydroxides
to accept hydrogens, and indeed they both do accept hydrogens, since
two water molecules were formed and there are no hydroxides remaining
with the barium. The equivalent weight is the molecular weight divided
by 2 .
The H3 P 0 4 in Equation 3.14 has three hydrogens to be donated.
However, there is still one hydrogen remaining with the phosphate on
the right side and therefore only two hydrogens were donated. Thus the
equivalent weight of the H3 P 0 4 is the molecular weight divided by 2. This
does not preclude the possibility of H3 P 0 4 donating either one or three
hydrogens in some other reaction. The equivalent weight depends on the
reaction involved.
Some general statements can be made concerning the above discus
sion. Sulfuric acid almost never donates just one of its hydrogens. Thus,
its equivalent weight is almost always the molecular weight divided by
2 . Hydroxides, such as the Ba(OH)2, almost always donate all the OH
groups that they have. Thus, the equivalent weight of a hydroxide is

always the molecular weight divided by the number of hydroxides that
it has per molecule.
3.5.1b Redox
With oxidizing agents or reducing agents, the equivalent weight is the

molecular weight divided by the number of electrons involved in the
balanced half-reaction per molecule. Thus again, we must know the reaction
involved and also be able to isolate and balance the half-reactions both
for the elements involved and for charge, since the charge balance gives
the number of electrons. Let us again look at a few examples. First,
Fe+2 + M n04“ -» Mn+2 + Fe+3 (3.15)
The balanced half-reactions for this reaction are
Fe+2 -» Fe+3 + le~ (3-16)
and
5e" + 8H+ + M n04~ -4 Mn+2 + 4H20 (3.17)
The equivalent weight of the iron compound used as the source of the
Fe, such as FeCl2, Fe(N03)2, etc., is the molecular weight divided by 1
(one electron given up per molecule in Equation 3.16). The equivalent
weight of the permanganate (KMnO^ NaMn04, etc.) is the molecular
weight divided by 5 (five electrons taken on per molecule in Equation
3.17).
Another example is the iodide in the iodide/iodine chemistry.
I ' + Cr20 7~2 -» I2 + Cr+3 (3.18)
The half reactions involved are
2F - » I2 + 2e~ (3.19)
and
6e~ + 14H+ + Cr20 7“2 -> 2C r+3 + 7H 20 (3.20)

The equivalent weight of potassium iodide, or whatever the source of the

iodide, is the molecular weight divided by 1 (remember — "per mol
ecule"), while the equivalent weight of potassium dichromate, for ex
ample, is the molecular weight divided by 6 .
3.5.1c Precipitation/Complexation
For precipitation and complexation reactions, the equivalent weight of

the substance furnishing the cation is the molecular weight divided by
the total positive charge present per molecule due to the cation. For the
substance furnishing the anion or ligand, it is that weight that reacts with
one equivalent of the substance furnishing the cation. Here are some
examples.
A g N 03 + NaCl -> AgCl + N aN 03 (3.21)
BaCl2 + Na2S 0 4 -> B a S 0 4 + 2NaCl (3.22)
3C u(N0 3 ) + 2 K 3P 0 4 -> Cu3 ( P 0 4 )2 + 6KNOa (3.23)
Ca+2 + EDTA-4 -4 Ca(EDTA)-2 (3.24)
Cu+2 + 4N H 3 - » Cu(NH3)4+2 (3.25)
The equivalent weights of both reactants in Equation 3.21 are the molecular
weights divided by 1. The equivalent weights of both reactants in Equation
3.22 are the molecular weights divided by 2. In Equation 3.23, the equivalent
weight of Cu(N0 3 ) 2 is the molecular weight divided by 2 . The equivalent
weight of the K3 P 0 4 is the molecular weight divided by 3. In Equation
3.24, the equivalent weights of the calcium compound and the EDTA are
the molecular weights divided by 2. In Equation 3.25, the equivalent
weight of the copper compound is the molecular weight divided by 2 .
The equivalent weight of the ammonia is the molecular weight multiplied
by 2 .
When there is no equation given, the equivalent weight is often de

termined ambiguously according to the above definition. If an analyst is
asked to prepare 500 mL of a 0.10 N Na2 S 0 4 solution, for example, with
no equation to guide them, they would likely decide ambiguously that
the equivalent weight is the molecular weight divided by 2 , since the total
charge due to the sodium is +2 .
The most important complexation reaction is the reaction of metal
cations with ethylenediaminetetraacetic acid (EDTA) (see Chapter 4) as
in Equation 3.24 above. Since this is always a one-to-one reaction in terms
of moles, the analyst almost always works with moles and molarity (or
titer — see Chapter 4) in this application.
3.5.2 Solution Preparation
To prepare a solution of a certain normality (N) when a pure chemical

is to be measured, the following formula is useful.
grams to measure = L D x ND x EWSOL (3.26)
in which LD is the liters desired, NDis the normality desired, and EWgoL
is the equivalent weight of the solute.
To prepare a solution of a certain normality when a solution of a certain
greater normality is to be diluted, the usual dilution equation is useful
(Equation 3.1). However, if only the molarity of the solution to be diluted
is known, it is first necessary to convert to normality.
N b = M b x the number of equivalents per mole (3.27)
in which NBis the normality before dilution, and MBis the molarity before
dilution. The number of equivalents per mole is, of course, the number
by which the molecular weight is divided to obtain the equivalent weight.
Example 9
How would you prepare 500 mL of a 0.20 N solution of KH2 P 0 4 (a
pure, solid chemical) if it is to be used as in the following equation?
KH2P 0 4 + 2KOH K3P 0 4 + 2H2Q

Solution 9
grams to measure = 0.500 x 0.20 x
= 6.8 g
6 .8 g of KH2 P 0 4 are measured into a 500 mL container. Water is added

to dissolve and dilute to volume.
Example 10
How would you prepare 500 mL of a 0.20 N solution of H2 S 0 4 from
a solution of concentrated H2 S 0 4 which is 18 M?
Solution 10
N b
= 18x2
= 36
C b x Vb = C A1 x V A
1
3 6 x Vb = 0 .2 0 x 5 0 0
VB
= 2.8 mL
2.8 mL are measured into a 500 mL container and diluted to volume.
3.6 PARTS PER MILLION/BILLION
Solution concentration is often expressed in parts per million (ppm)

or parts per billion (ppb). For such solutions, parts per million is most
often assumed to mean milligrams of the indicated solute per liter of
solution, and parts per billion to mean micrograms of such solute per liter
of solution. Technically, this should be milligrams per kilogram, or
micrograms per kilogram, but, as stated previously in the percent dis-
micrograms per kilogram, but, as stated previously in the percent dis

cussion, since the density of water is nearly 1 at room temperature, the
liter and the kilogram for such dilute solutions represent the same quan
tity. The grams to be measured (when the solute is a pure solid or liquid)
are calculated as follows.
L n x ppmn
grams to measure = — ---------- — (3.28)
6 1000 v '
and
L n x ppbn
grams to measure = 6— — (3.29)
in which LD is the liters desired, ppmD and ppbD represent the

concentrations desired, and 1 0 0 0 and 1 0 6 are the conversion factors which
convert milligrams and micrograms, respectively, to grams.
It may be more convenient or even required at times to weigh a
quantity of a chemical that "contains" the solute rather than the pure
solute. For example, it is more convenient to weigh the appropriate
quantity of sodium chloride, rather than pure sodium, when preparing
a solution of a certain parts per million sodium. Likewise, when preparing
a solution of a certain parts per million nitrate, it would be necessary to
weigh a quantity of a nitrate salt, such as potassium nitrate. In these cases,
a "gravimetric factor," which would convert the weight of the solute to
the weight of the substance actually weighed, would be required as part
of the calculation.
L n x ppm x gravimetric factor

grams to measure = —- --------------------------------------- (3.30)
1000 v '
and
L x ppb x gravimetric factor

grams to measure = — --------- -— ^ -------------------- (3.31)
With both parts per million and parts per billion, the quantity weighed
is often an extremely small quantity— frequently too small to be weighed
accurately. If this is the case, a more concentrated solution should be
prepared and then diluted. This would obviously involve a dilution
Also, for many solutions of this type, such as solutions of metals to be

prepared for atomic absorption analysis (Chapter 7), the more concen
trated solutions may be available commercially. With such solutions,
then, only a dilution is required, but it may be necessary to convert from
parts per billion to parts per million in order to have both concentration
terms in the dilution equation of the same unit. In order to do this, the
following equation is useful.
ppb = ppm x 1000 (3.32)
Some examples follow.
Example 11
How would you prepare 500 mL of a 25.0 ppm copper solution using
pure copper metal as the solute?
Solution 11
2 5 x 0 .5 0 0
grams to measure = ---------------- = 0.0125 g
1000
0.0125 g of copper metal is weighed into a 500 mL flask, a nitric acid

solution is added to dissolve the metal, and water is added to the 500
mL mark.
Example 12
How would you prepare 500 mL of a 100 ppb sodium solution using
sodium chloride as the solute?
Solution 12
The gravimetric factor for converting the weight of sodium to sodium
chloride is the molecular weight of sodium chloride divided by the atomic
weight of sodium, which is 2.542.
1 0 0 x 0 .5 0 0 x 2 .5 4 2
grams to measure = ------------^ -------------= 0.0001271 g
Given such a small quantity of grams calculated, it would be expedient

to prepare a much more concentrated solution (say 1 0 0 ppm) and dilute.
So 100 ppm is 1000 times more concentrated, and thus would require 1000
times more grams, or 0.1271 g. Equations 3.1 and 3.32 would then prove
useful.
ppb = 100x 1000 = 100,000
100,000 xV B =100x500
Vn
15
= 0.500 mL
0.500 mL of the 100 ppm solution would be measured into a 500 mL flask
and diluted to volume.
3.7 BUFFER SOLUTIONS
Buffer solutions, or solutions which resist changes in pH even when

a strong acid or base is added, are almost always composed of a weak
acid or weak base and the salt of this weak acid or base. The reason for
the resistance to pH change is that the weak acid or weak base ionization
equilibrium shifts in these solutions such that the H+ or OH” added are
consumed, thus resulting in no net pH change.
Although commercially prepared buffer solutions are available, these
are most often utilized solely for pH meter calibration and not for adjusting
or maintaining a chemical reaction system at a given pH. It is not surprising,
therefore, that the analyst often needs to prepare h is /h e r ow n solutions
for this purpose. It then becomes a question of what proportions of the
acid, or base, and its salt should be mixed to give the desired pH.
The answer is in the expression for the ionization constant, Ka or Kb,
where the ratio of the salt concentration to the acid concentration is found.
In the case of a weak acid,
H A ^=^ H+A- (3.33)
[ H 1 1[ a - ]
k. = W (M 4)
and in the case of a weak base,
b + h 2o ^=^ BH+ + OH~ (3.35)

[ b h +][ o h ~]
Kb = i ----- 1 (3-36)
~ [B]
Knowing the value of Ka or Kj, for a given weak acid or base and
knowing the desired pH value, one can calculate the ratio of salt concen
tration to acid (or base) concentration that will produce the given pH.
Rearranging Equation 3.34, for example, would show the method for
calculating this ratio in the case of a weak add and its salt.
K. M
[JF fiiix j (M 7)
This is one form of the Henderson-Hasselbalch equation for dealing with

buffer solutions. It says that one would simply divide the Ka by the [H+]
to obtain the required ratio. It should be stressed that since the Kg, or Kb,
enters into the calculation, how weak the acid or base is dictates what
is a workable pH range for that acid or base. Table 3.3 gives commonly
used examples of acid/salt and base/salt combinations and for what pH
range each is useful.
It should also be stressed that the pH value of an actual buffer solution,
prepared by mixing quantities of the weak acid or base and its salt based
on the calculated ratio, will likely be different from what is expected. The
reason for this is the use of approximations in the calculations, such as
the use of molar concentration rather than activity, and the fact that the
values for the un-ionized acid and base concentrations in the denomina
tors in Equations 3.34 and 3.36 are approximations. A better method of
preparing a buffer solution would be to prepare a solution of the salt (the
concentration is not important) and then add a solution of a strong acid
(or base, if the salt is a salt of a weak base) until the pH, as measured
by a pH meter, is the desired pH. The strong acid (or base), in combination
with the salt, creates the equilibrium required for the buffering action.
For example, to prepare a pH = 9 buffer solution, one would prepare a
solution of ammonium chloride (refer to Table 3.3) and then add a
solution of sodium hydroxide while stirring and monitoring the pH with
a pH meter. The preparation is complete when the pH reaches 9. The
equilibrium created would be
Table 3.3 Commonly Used Examples of Acid/Salt and Base/Salt Combinations

and Corresponding pH Ranges
Combination pH Range
Trichloroacetic acid +
sodium trichloroacetate 1.8-3.8
Acetic acid +
sodium acetate 3.T-5.7
Sodium dihydrogen phosphate +
sodium monohydrogen phosphate 6.1-8.1
Ammonium hydroxide +
ammonium chloride 8.3-10.3
Table 3.4 Recipes for Some of the More Popular Buffer Solutions
pH = 4.0 phthalate buffer Dissolve 10.12 g of dried potassium hydrogen

phthalate (KHP) in 1 L of solution
pH = 6.9 phosphate buffer Dissolve 3.39 g of dried potassium dihydrogen
phosphate and 3.53 g of dried sodium
monohydrogen phosphate in 1 L of solution
pH = 10.0 ammonia buffer Dissolve 70.0 g of dried ammonium chloride and
570 mL of concentrated ammonium hydroxide in
1L of solution
and the solution would be a solution containing a weak base (NH4 OH)
and its salt (NH4 C1).
Recipes for standard buffer solutions can be useful, however. Table 3.4
gives specific directions for preparing some popular buffer solutions.
C hapter 4
W et M ethods and A pplications
4.1 INTRODUCTION
There are two classical methods of chemical analysis that are referred
to as "wet" methods. These are "gravimetric" analysis and "titrimetric"
(or "volumetric") analysis. They are called wet methods because they
rarely make use of any electronic instrumentation beyond the analytical
balance. They employ physical separation schemes and/or chemical
reactions and classical reaction stoichiometry as the sole basis for arriving
at the results. Laboratories referred to as "wet laboratories" are those in
which analyses and preparations involving titrating, weighing, extracting,
etc. are performed. Such procedures can be sample pretreatment
procedures, such as extractions, that are performed in advance of
instrumental analyses, as well as the self-contained gravimetric and
titrimetric procedures.
4.2 GRAVIMETRIC ANALYSIS
Gravimetric analysis is the classical wet method characterized by the

fact that only one kind of measurement, that of weight, is made on the
69
sample, its constituents, and reaction products. Only measurements of

weight are thus used in the calculation.
Gravimetric procedures always involve the separation of the analyte
constituent from the sample so that it can be isolated and weighed. This
separation can be a physical separation, such as through solubility or
volatilization, or it can be a chemical separation, i.e., by chemical reaction,
so as to form a precipitate that can be weighed. Physical separations for
gravimetric analysis are often conceptually simple. An example would
be the analysis of a wastewater sample for suspended solids and total
solids. The procedures for such analytes are purely gravimetric and
involve simple physical separation schemes. In the case of total solids,
a particular volume of a wastewater sample is added to a preweighed
evaporating dish, the water is evaporated in an oven, and the dish is
weighed again. The difference in the two weights is the total solids in
the sample. For suspended solids, a preweighed filtering crucible is used
to filter the undissolved solids from a given volume. After drying, this
crucible is weighed again, and the difference in the two weights is the
suspended solid content of that sample. Similarly, moisture or any volatile
component in various types of samples can be determined by evaporation
and weighing the container before and after.
Classical gravimetric analysis, which is now mostly obsolete, involved
the chemical separation scheme referred to above. An example would be
the analysis of a powdered sample for sulfate content. In this classical
example, the sulfate is separated by reacting it with a selective reagent,
such as a solution of a barium salt, such that an insoluble precipitate,
barium sulfate in this case, is formed. The precipitate is then filtered,
dried, often "ignited," and weighed, with the analyte calculated through
classical reaction stoichiometry. Since this "chemical" separation scheme
is now mostly obsolete, this method will not be considered further in this
text.
4.3 TITRIMETRIC ANALYSIS
The other classical wet chemical method, titrimetric analysis, unlike

gravimetric analysis, involves the accurate measurement of volume, as
well as weight, and is not obsolete. A technique known as a "titration"
is at the heart of the method. In a titration, a buret (Chapter 2) is used
Wet Methods and Applications 71
INDICATOR
FIGURE 4.1 The equipment and terminology of a titration experiment. (From

Kenkel, }., Analytical Chemistry for Technicians, Lewis Publishers, Inc.,
Chelsea, MI, 1988. With permission.)
to deliver a particular volume of a solution to a reaction flask. The

substance held in the buret is called the "titrant." The substance originally
held in the reaction flask is called the "titrand" or "substance titrated."
In addition, a substance known as an "indicator" is often added to the
reaction flask (see Figure 4.1). Typically, the titrant is added from the
buret for the purpose of quantitatively reacting with the substance titrated.
When the last bit of substance titrated is reacted with the added titrant
(the so-called "equivalence point"), a change, such as a color change, is
observed in the reaction flask. This change is usually caused by the
chemistry of the indicator occurring at the equivalence point and is used
to indicate when the addition of the titrant should be stopped. The analyst
must strive to add a minute amount of titrant when near the equivalence
point (so as to precisely arrive at the equivalence point) so that the precise
amount of titrant required is then what is added and measured.*
While indicators are commonly used to detect equivalence points, there are other tech
niques which are described in later chapters, namely potentiometry and amperometry. The
Karl Fischer method for water is a method that utilizes potentiometry. It will be discussed
later in this chapter.
While the equivalence point is the exact point at which the last bit of
substance titrated is reacted, the indicator usually does not respond
precisely at that moment. Although this is not usually considered a
problem, there is a distinction made between the point at which the
reaction is complete and the point at which the reaction is complete as
indicated by the indicator. The former is the equivalence point and the latter
is the "end point."
A titration experiment usually takes place for two reasons: (1) for the
standardization of a titrant and (2 ) for the quantitative determination of
a constituent (analyte) in a sample dissolved and placed in the reaction
flask. In either case, the volume added from the buret must be accurately
read and recorded, and the quantity of substance titrated originally placed
in the reaction flask must be accurately known. Let us now discuss each
of these types of experiments.
4.3.1 Standardization
A "standard solution" is a solution which has a known concentration,

meaning that it is known to the accuracy required for a given experiment.
"Standardization" refers to an experiment in which the concentration of
a solution is determined to the desired accuracy. To "standardize" a
solution means to determine the concentration of a solution via a
standardization experiment. A "primary standard" is the substance to
which a solution is compared which allows standardization to take place.
This last definition warrants some additional comment. Obviously, the
quality of the primary standard substance is ultimately the basis for a
successful standardization. This means that it must meet some special
requirements with respect to purity, etc., and these are enumerated in the
following:
1. It must be 100% pure, or at least its purity must be known.

2. If it is impure, the impurity must be inert.
3. It should be stable at drying oven temperatures.
4. It should not be hygroscopic; it should not absorb water when
exposed to laboratory air.
5. The reaction in which it takes part must be quantitative and pref
erably fast.
6. A high molecular weight is desirable.
Most substances used as a primary standard can be purchased as

primary standard grade (Chapter 2), and this is appropriate and sufficient
for a standardization experiment.
There are two general approaches to a standardization. One is to weigh
the primary standard directly into the reaction flask. This approach could
be called a standardization with a primary standard. The other approach
is to measure a volume of a solution into the reaction flask; a solution
that already has a known concentration (standard solution) or at least a
concentration that will become known in subsequent experiments. This
approach can be called standardization with a standard solution. In this
latter approach, the concentration of the standard solution used can be
known directly through its preparation, if the solute is a primary standard,
or it can be determined through an independent standardization
experiment. In any case, the calculation involved is based on the fact that
the equivalents of titrant added at the end point are equal to the equivalents
of substance titrated originally in the flask (see Chapter 3 for the definition
of the "equivalent").
equivalents of titrant (T) = equivalents of substance titrated (ST) (4.1)
or
equiv.T = equiv.ST (4.2)
If a primary standard was weighed into the reaction flask, then Equa
tion 4.2 becomes
grams ST
C x N t = — ---------------------------------------------------------(43)
equiv. wt. ST
in which Lj- is the liters of titrant (the buret reading converted to liters);
Nt is the normality of the titrant (the concentration — see Chapter 3);
gramsSTis the weight of the substance titrated, the primary standard; and
equiv. wt.ST is the equivalent weight of this substance titrated. In the
actual experiment, NT is the only unknown in Equation 4.3, and thus it
can be calculated.
If a standard solution was pipetted into the reaction flask, then Equation
4.2 becomes:
(4.4)
in which LgT is the volume pipetted, and NST is the normality of the
solution pipetted. Here again, NT is the only unknown, and thus it can
be calculated. If a primary standard was used to prepare the solution to
be pipetted into the reaction flask, NST can be calculated as follows:
gramsps/equiv. wt.ps
liters prepared
(4.5)
in which "PS" refers to "primary standard." A volumetric flask (Chapter

2 ) is required for this, since the "liters prepared" needs to be known
accurately.
One further comment concerning Equation 4.4 and this experiment in
general is that the solution pipetted into the reaction flask, rather than
the titrant, may be the solution to be standardized. In that case, NST in
Equation 4.4 is the unknown concentration to be calculated and NT is
known.
It should be mentioned that a standardization experiment can be
avoided completely if the titrant solute is a primary standard material,
and the concentration can thus be known through its preparation. This
would involve a procedure identical to that described above for the
standard solution used in "standardization with a solution" — weighing
the solute accurately into a volumetric flask and calculating the normality
as in Equation 4.5.
Finally, a concentration can be represented as "titer." The titer of a
solution is the weight of a substance that reacts with 1 mL of the solution,
such as a titrant. The calculation here involves dividing the grams of
primary standard present in the reaction flask by the milliliters of titrant
needed to titrate it. This would be the titer of the titrant and a measure
of its strength or concentration. See the determination of water hardness
and the Karl Fischer method for water in organic solvents in Section 4.4
for example applications of this concept.
4.3.2 Titration of Unknowns
The quantitative determination of an analyte in a solution is the other

use for a titration experiment. In such an experiment, in contrast to the
standardization experiment, the titrant concentration is known, either

directly through its preparation or because it was standardized, while the
quantity of substance titrated, the analyte, is unknown. In such an
experiment, the sample is measured, pipetted or weighed, into the reaction
flask and titrated with the titrant to a suitable end point.
The calculation used depends on what result is desired. It is common
to calculate the percent of an analyte in a sample that was weighed into
the reaction flask, in which case the following derivation is useful.
„ , analyte wt. ir>n

% analyte = ------------- x 100 (a f.)
sample wt.
The analyte weight can be calculated from the titrant's normality and
volume and the equivalent weight of the analyte:
analyte wt.= L T x NT x equiv. wt.analyte (4.7)
Combining Equations 4.6 and 4.7, we obtain the equation for the percent
of an analyte:
L T x NT x equiv. wt.anal te
% analyte = -----------------------------— x 100 (4 «)
sample wt.
Sometimes parts per million of analyte may be calculated, particularly

if the sample is a liquid. The water hardness titration discussed in Section
4.4 is an example of such a titration.
4.4 EXAMPLES OF REAL-WORLD TITRIMETRIC ANALYSIS
4.4.1 The Kjeldahl Method
A titrimetric method that has been used for many years for the deter
mination of nitrogen and/or protein in a sample is the Kjeldahl method.
Examples of samples include grain, protein supplements for animals
feed, fertilizers, and food products. It is a method that may use a special
technique called a "back titration." We will now describe this technique

in detail.
The method consists of three parts: (1) the digestion, (2) the distillation,
and (3) the titration. The digestion step is in essence the dissolving step.
The sample is weighed and placed in a Kjeldahl flask, which is a round
bottom flask with a long neck similar in appearance to a volumetric flask,
except for the round bottom and the lack of a calibration line. A fairly
small volume of concentrated sulfuric acid along with a quantity of K2 S 0 4
(to raise the boiling of the sulfuric acid) and a catalyst (typically an
amount of CuS04, selenium, or a selenium compound) is added, and the
flask is placed in a heating mantle and heated. The sulfuric acid boils and
sample digests for a period of time until it is evident that the sample is
dissolved and a clear solution is contained in the flask. The digestion must
be carried out in a fume hood, since thick S 0 3 fumes evolve from the flask
until the sample is dissolved. At this point, the contents of the flask is
diluted with water, and an amount of fairly concentrated sodium hydroxide
is added to neutralize the acid. Immediately upon neutralization, the
nitrogen originally present in the sample is converted to ammonia. At
this point the distillation step is begun.
A laboratory that runs Kjeldahl analyses routinely would likely have
a special apparatus set up for the distillation. The essence of this apparatus
is shown in Figure 4.2. A baffle is placed on the top of the Kjeldahl flask
and subsequently connected to a condenser which in turn guides the
distillate into a receiving flask as shown. The ammonia is thus distilled
into the receiving flask. The receiving flask contains an acid for reaction
with the ammonia.
The acid in the receiving flask can either be a dilute (perhaps 0.10
normal) standardized solution of a strong acid, such as sulfuric acid, or
a solution of boric acid. If it is the former, it is an example of a '"back
titration." If it is the latter, it is an example of an indirect titration.
In the back titration method, an excess, but carefully measured, amount
of the standardized acid is contained in the flask such that, after the
ammonia bubbles through it and is consumed, an excess remains. The
flask is then removed from the apparatus, and the excess acid is titrated
with a standardized NaOH solution. The analyte in this procedure is the
nitrogen (in the form of ammonia as it enters the flask), and thus the
amount of acid consumed is the important measurement. The amount
of acid consumed is the difference between the total amount present and
the amount that was in excess. It is called a back titration because the
amount of acid in the flask is in excess and in essence goes beyond the
FIGURE 4.2 A typical apparatus for the Kjeldahl distillation step.
end point for the reaction with the ammonia. Thus, the analyst must come
back to the end point with the sodium hydroxide. The calculation is.
(l t x n_ - x n b t ) x 14.00
l bt
N= —-------S li-------------x 100 (4.9)
sample wt.
in which the titrant is the dilute sulfuric acid, and "BT" refers to "back
titrant," the NaOH. The equivalent weight shown is the atomic weight
of nitrogen. The percent protein may also be calculated, in which case
the equivalent weight of the protein is substituted for the 14.00.
In the indirect method using boric acid, the ammonia reacts with the
boric acid producing a partially neutralized salt of boric acid, H2 BO3 '1,
which can then be titrated with a standardized acid. The amount of
standardized acid needed is proportional to the amount of ammonia that
bubbled through. It is called an indirect method because the ammonia
is determined by titration of the HjBC^'1. In a direct titration, the analyte
would be reacted directly with the titrant. Equation 4.8 is used for this
as in the direct method.
I
N d Q -C CH.
\ n c h 2c h 2 n ;
N d “0 - C CH. < CH, C-O H

I
FIGURE 4.3 Disodium dihydrogen EDTA.
4.4.2 Water Hardness
The analysis of water samples for hardness is a common analysis in

water and environmental laboratories. The usual procedure for this is a
titrimetric procedure in which a solution of disodium dihydrogen
ethylenediaminetetracetic acid dihydrate (Na2 H2 EDTA2H 2 0 ) is the titrant
(see Figure 4.3). This compound furnishes a hexadentate ligand which
reacts with the calcium, magnesium, and iron ions in hard water forming
complex ions. The calcium, magnesium, and iron ions are the major
contributors to water hardness. The EDTA is in the buret, and the water
sample in the reaction flask.
A complex ion is a polyatomic charged aggregate consisting of a
positively charged metal ion combined with a ligand. A ligand is an
uncharged or negatively charged chemical species that reacts with a metal
ion to form a complex ion. Ligands are classified according to the number
of bonding sites in their structure available for bonding to the metal ion.
A ligand with one site is "monodentate," one with two sites is "bidentate,"
etc. The EDTA species, in somewhat basic solution (pH = 10), is
"hexadentate"; there are six bonding sites available for bonding to metal
ions. The reaction with metal ions, however, is one-to-one. The calcium-
EDTA complex ion is shown in Figure 4.4.
The usual indicator for this titration is eriochrome black T, an organic
dye. Eriochrome black T is actually a ligand that also reacts with the metal
ions, like EDTA. In the free uncombined form and at the basic pH, it
imparts a sky blue color to the solution, but if it is a part of a complex
ion with either calcium or magnesium ions, it is a wine red color. Thus,
before adding any EDTA from a buret, the hard water sample, containing
a pH = 10 buffer (such as an ammonia buffer— see Chapter 3) and several
drops of the indicator solution, will be wine red. As the EDTA solution
is added, the EDTA ligand reacts with the free metal ions and then
FIGURE 4.4 The calcium-EDTA complex ion.
actually reacts with the metal/indicator complex ion, displacing and

complexing the metal ions and resulting in the free indicator ligand,
which, as mentioned above, gives the solution a sky blue color. The color
change, then, is the total conversion of the wine red color to the sky blue
color, with every trace of red disappearing at the end point.
It is known that this color change is quite sharp when magnesium ions
are present. In cases in which magnesium ions are not present in the water
samples, the end point will not be sharp. Because of this, a small amount
of magnesium chloride is added to the EDTA as it is prepared, and thus
a sharp end point is assured. Also, other indicators, similar to eriochrome
black T, are sometimes used; these are murexide and calmagite. Calmagite
shows approximately the same end point and has an advantage in that
its solutions are more stable with time. Eriochrome black T solutions need
to be prepared fresh about every two weeks.
Water hardness is usually reported either as parts per million CaC0 3
or grains per gallon CaC03. The parts per million unit, as discussed in
Chapter 3, represents the number of milligrams present per liter. Since
it is easier to measure the volumes of liquids, as opposed to weight, a
volume of water is measured (pipetted) into the reaction flask, rather than
a weight, and thus it is logical and easy to calculate the milligram per
liter, or parts per million.
L edta x M edta x F.W. CaCP3 x 1000

ppm C aC 03 = (4.10)
liters of water
The numerator is multiplied by 1000 to convert grams to milligrams. Also,

multiplying by the formula weight of CaC0 3 in the numerator has the
effect of reporting all metals that react with the EDTA as CaCO^ a
common practice in water laboratories.
Molarity and formula weight (F.W.) are used here since the reaction
is one-to-one in terms of moles, and the moles of EDTA (L x M) are thus
converted directly to grams of C aC 0 3 by multiplying by the formula
weight of CaC03. Standardization experiments for the EDTA have as
their goal, then, the molarity, rather than the normality of the solution.
For this reason, Equations 4.3,4.4, and 4.5 would involve molarity instead
of normality, and formula weight instead of equivalent weight in this
case. The primary standard for this experiment is typically a pure grade
of C aC 03.
The concentration of the EDTA for this titration is typically 0.01 M.
Because of this, and since the F.W. of CaC0 3 is 100.09, if 100 mL of water
is used, the results can be calculated by simply multiplying the buret
reading (in milliliters) by 1 0 .
Finally, the "titer" (see Section 4.3) of EDTA, or the number of grams
of C aC 0 3 that reacts with 1 mL of EDTA, can be used here.
C aC 03 titer of EDTA x mL of EDTA x 1000

ppm C aC 03 = (4.11)
liters of water
The parts per million unit can, of course, be converted to grains per
gallon or vice versa. To convert parts per million to grains per gallon, we
multiply the given parts per million by 0.05833. To convert grains per
gallon to parts per million, we multiply the given grains per gallon by
17.14.
4.4.3 The Karl Fischer Titration
The Karl Fischer titration is a titration for the determination of small

quantities of water in organic solvents or crystalline solids. It is an elec-
troanalytical method (see Chapter 11) utilizing a pair of electrodes dipped
into the reaction vessel. More precisely, it is a "bipotentiometric" tech-
nique in which a power source is connected across the electrodes so as

to maintain a constant current between them. The power source may be
as simple as a pH meter which has a pair of input terminals on it labeled
"K-F" for Karl Fischer. In the modem laboratory, however, the Karl
Fischer apparatus is typically a self-contained unit complete with titration
cell, power source, automatic titrator (Chapter 12), reagent reservoir, and
a computer, which includes digital and hardcopy readout and sometimes
a keyboard with monitor. Since it is the water content that is being
determined, moisture contamination must be scrupulously avoided. This
means that the sample and titrant must be protected. Figure 4.5 shows
a photograph of a modem Karl Fischer unit including all components
referred to above (except the monitor). Drying tubes are shown protecting
the titrant in the reservoir and the sample in the cell from atmospheric
moisture.
The "titrant" for this method is the so-called Karl Fischer reagent. We
have placed the word titrant in quotes here because the sample is actually
added from the buret, not the reagent, although it is the reagent that must
be standardized. The reagent is a mixture of chemicals, including iodine,
pyridine, and S 0 2, in a 1:10:3 ratio and usually dissolved in methanol
solvent. Addition of the sample to this reagent begins a sequence of
reactions which includes the consumption of water and iodine and the
formation of iodide ion. In a solution which includes both iodine and
iodide, the potential required to maintain the small constant current is
small. However, at the moment that the last trace of iodine is consumed
by the water from the added sample, the required potential is quite large,
and there is a sudden shift to higher potential values, signaling the end
point.
The typical standardization procedure involves the use of a sample of
methanol with a known amount of dissolved water (a standard solution
of water) prepared by adding a weight of water to a particular volume
of water-free methanol solvent (or a volume of methanol solvent in which
the water content is known) such that the milligrams per milliliter of
water in the methanol is known. A carefully measured (pipetted) amount
of the Karl Fischer reagent is placed in the cell, and the standard water
sample is added to the end point. In this way the titer of the reagent, the
milligram of water per milliliter of reagent, is determined.
mg water/mL methanol solution x mL methanol solution needed

reagent titer = -----------------------------------------------------------------------------------------
mL of reagent used
(4.12)
FIGURE 4.5 A photograph of a modern Karl Fischer titrator. (Photograph courtesy

of Brinkman Instruments, Inc., Westbury, NY.)
The water content in an unknown sample can then be determined by

multiplying the reagent titer by the milliliters of reagent used and divid
ing by the milliliters of sample needed.
, T , reagent titer x mL reagent used

mg water/mL sample = ------------------------------------------------- (A 13)
ml eam nlp nppHpH ' ' '
4.5 FINAL COMMENTS
In general, wet methods of analysis are used in the modern laboratory

in situations in which (1 ) they are more convenient and/or accurate than
instrumental methods and (2 ) analyte concentrations are so high that
instrumental methods are not appropriate. The examples cited above are
mostly a matter of the convenience and accuracy of the established method,
although both the Kjeldahl and Karl Fischer methods, in their modern
form, may be considered at least partially instrumental. Another example
method based on convenience is "total alkalinity." Alkalinity is the capacity
of dissolved solutes to neutralize an added strong acid. The strong acid
is conveniently added from a buret, and the neutralization point (end
point) is conveniently signaled with the use of an indicator or pH meter.
This procedure is in common use in laboratories in which total alkalinity
of water-based samples needs to be determined.
An example of an analyte concentration being too high for instrumental
methods is in the analysis of plating solutions. Some, but not all, of the
ingredients in such solutions are present at such high concentrations that
the instrumental methods cannot be used without considerable dilution.
It is not unusual to encounter titrimetric procedures in use in plating
industry laboratories for such analytes as chloride, cyanide, sulfates, etc.
The instrumental methods are usually most useful for low concentrations
(parts per million level).
C hapter 5
I nstrumental M ethods —
G eneral D iscussion
5.1 INTRODUCTION
The remainder of this text is devoted to techniques and methods of

chemical analysis involving electronic instrumentation. In Chapter 4, we
made a distinction between wet methods and instrumental methods,
saying that the wet methods known as gravimetric and titrimetric analysis
are more classical as compared to instrumental techniques. They are
classical techniques in the sense that they have adequately served the
chemical analysis laboratory for many years, but have now largely been
replaced by the streamlined and computerized instrumental methods,
especially where there is a need to detect trace amounts of analytes and
to analyze more complicated samples. There are certain aspects of
instrumental analysis that are common to virtually all instrumental
techniques. These aspects will be dealt with in detail in this chapter.
Characteristics of the individual instruments and associated procedures
will be covered in the chapters to follow.
85
5.2 INSTRUMENTAL DATA AND READOUT
5.2.1 Recorders
Most laboratory instruments measure some physical or chemical

parameter of a solution of a sample and display the measurement as a
proportional voltage on some type of readout device. This device can be
a simple meter (like the display on a pH meter), digital or otherwise; a
recorder, which may either be a strip-chart recorder or x-y recorder; a
computer monitoring screen and/or a printer-plotter; or any combination
of these. Let us first discuss the details of the two types of recorders and
how they are used.
A recorder has one or more writing pens driven by a servomotor and
records information on paper. The position of the pen on the paper
represents the voltage or voltages fed into the recorder from the instrument,
just as the position of the pointer on a meter represents such a voltage.
The advantage of a recorder is that the voltage can be permanently
recorded over a period of time, a capability which is of significant
importance when the voltage output of an instrument changes with time,
such as with gas chromatography (Chapter 9) or high performance liquid
chromatography (Chapter 10), or when two parameters change with time,
one as a function of the other, such as when recording a molecular
absorption spectrum with a spectrophotometer (Chapter 6 ). This
characteristic of voltage signals changing with time is quite common, and
thus recorders are very commonly used in instrumental laboratories as
readout devices.
The differences between the strip-chart recorder and the x-y recorder
are noteworthy. The strip-chart recorder records the voltage output of an
instrument strictly as a function of time. In other words, the pen records
the voltage on the paper as the paper moves at a certain rate through the
recorder. Thus, one axis of the resulting graph is time, and the other
is the voltage level. Multiple pens in a strip-chart recorder can be used
to record multiple voltage levels and how they change with time. The
x-y recorder is a recording device which accepts two voltages from the
instrument for recording with only one pen. One of the voltage levels
determines the x-axis position of the pen, while the other determines the
y-axis position. The paper does not move, but rather the pen moves over
the paper as the two voltage levels change. Figure 5.1 shows representations
Instrumental Methods — General Discussion 87
FIGURE 5.1 Drawings of (a) a strip-chart recorder (note only one set of input
terminals for only one incoming signal), and (b) an x-y recorder (note
two sets of input terminals for two incoming signals: one for the x-
axis and one for the y-axis).
of (a) a strip-chart recorder and (b) an x-y recorder, while Figure 5.2 shows
some sample recordings.
5.2.2 Instrument Readout and Concentration
Of course the goal of instrumental methods of analysis most of the time

is the concentration of a constituent in an unknown sample. Thus, the
_U 1
!' 1
- i....
i !
FIGURE 5.2 (a) The absorbances of a series of standard solutions recorded on a

strip-chart recorder, (b) A molecular absorption spectrum recorded
with an x-y recorder. (From Kenkel, J., Analytical Chemistry for Tech
nicians, Lewis Publishers, Inc., Chelsea, MI, 1988. With permission.)
concentration of the desired constituent in the solution measured is to

be determined from the readout resulting from that solution. As we have
said, the readout (let us call it "R"), whatever it represents, is proportional
to this concentration.
R = KC (5.1)
If the proportionality constant, K, is known, then the concentration can

be calculated.
C = R/K (5.2)
Often, however, the proportionality constant is not known, and the

experiment is not as simple as implied above. One alternate method
would be to prepare a solution of the constituent so that the concentration
is known (a standard solution), measure R, and then calculate K.
K = R/C (5.3)
In this equation, Rg is the readout for the standard, and Cs is the

concentration of the standard. The value of K is then used to calculate
C for the unknown, assuming, of course, that the parameters which
contribute to the value of K for both solutions are identical at both
concentrations.
C U= R U/ K (5.4)
Here, R„ is thereadoutfor the unknown, and Cu is the concentration

of the unknown. Actually, the value of K need not be calculated at all,
as is obvious from the following:
and
In other words, the value of Cu is related to that of Cs by the ratio of

their instrument readings. Or, Rg is to Ry as Cs is to Cu.
This treatment requires that the parameters that contribute to the value
of K not change between the two concentration levels, as indicated above.
If the concentrations are very nearly the same, or if the relationship
between the readout and concentration is of the same proportion at the
different concentration levels, then this treatment is entirely valid. How
ever, this may not be the case, and it is therefore useful to determine the
linearity of R as a function of C over the concentration range in question.
Such a linearity would verify the constancy of K. The procedure involves
preparing a series of standard solutions and making a graph of R vs C
to determine this linearity (see Figure 5.3a). If the results show linearity
over the concentration range in question, the unknown concentration
may be determined from the graph as shown in Figure 5.3b. Because of
the uncertainty in knowing whether the proportionality constant is in
deed constant over the concentration range studied, this series of stan
dard solution method is commonplace in an instrumental analysis labo
ratory for virtually all quantitative instrumental procedures. Examples of
this abound in this text for many spectrophotometxic, chromatographic,
and other techniques.
5.2.3 Method of Least Squares
The linearity of instrument readout vs concentration data must be

established for best results. Random indeterminate errors during the
solution preparation and during the measurement of R may cause the
results to appear to deviate from linearity to some extent. In that case,
a method must be adopted which will fit a straight line to the data as
well as possible. It may happen that some (or even all) of the points may
not fall exactly on the line because of these random errors, but a straight
line must still be drawn, since the random errors are indeed random and
cannot be compensated for directly. Thus, the best straight line possible
is drawn through the points (see Figure 5.4).
(a) (b)
FIGURE 5.3 (a) A plot of an instrument readout, R, vs the concentration, C, of an

analyte, showing linear relationship in the concentration range chosen,
(b) The determination of an unknown's concentration from the graph
in (a).
FIGURE 5.4 An example of a straight line fitted to a set of data.
"Eyeballing" the line through the points with a straightedge on the

graph paper may easily result in significant error. Therefore, a procedure
called the method of least squares (also called linear regression analysis)
is best applied to the data, and this method results mathematically in the
best straight line possible through a given set of data points. By this
method, the best straight line fit is obtained when the sum of the squares
of the individual y-axis value deviations (deviations between the plotted
y values and the values on the proposed line) are at a minimum. This
"proposed line" is actually calculated from the given data, a slope and
y-intercept ("m" and "b," respectively, in the equation for a straight line,
y = mx + b) are then obtained, and the deviations ( y ^ f - yline) for each
given x are calculated. Finding the values of the slope and the y-intercept
that minimize the sum of the squares of the deviations involves some
complicated mathematics that is beyond the scope of this text. Computers
and programmable calculators, however, handle this routinely in the
modem laboratory, and the results are very important to a given analysis
since the line that is determined is statistically the most correct line that
can be drawn with the data obtained. The concentrations of unknown
samples are also readily obtainable on the calculator or computer since
the equation of the straight line, including the slope and y-intercept, are
known as a result of the least squares procedure. Other statistically
important parameters are readily obtainable as well, including the "cor
relation coefficient."
5.2.4 The Correlation Coefficient
The correlation coefficient is one measure of how well the straight line
fits the analyst's data — how well a change in one variable correlates with
a change in another. Laboratories can establish their own criteria as to
what numerical value for such a coefficient is required for the accuracy
desired. A correlation coefficient of exactly 1 indicates perfectly linear
data. This, however, rarely occurs in practice. It occurs if all instrument
readings increase by exactly the same factor as the concentration level
increases, as in the following sample data:
R C
4 2
8 4
12 6
24 12
Due to random errors, this data is more ideal rather than real. Data that
approaches such linearity will show a correlation coefficient less than 1 ,
but very nearly equal to 1. Numbers such as 0.9997 or 0.9996 are considered
excellent and attainable correlation coefficients for many instrumental
techniques. Good pipetting and weighing technique when preparing
standards and well-maintained and calibrated instruments can minimize

random errors and can produce excellent correlation coefficients and
therefore accurate results. The analyst usually strives for at least two nines,
or possibly three, in their correlation coefficients, depending on the
particular instrumental method used. Again, these coefficients can be
determined on programmable calculators and laboratory computers.
5.3 METHODS FOR QUANTITATIVE ANALYSIS
As we determined in the last section, the usual procedure for

instrumental analysis is to plot the instrument readout for a series of
standard solutions vs the concentration of those solutions, establish the
linear range, and then, after determining an unknown readout, find its
concentration by interpolation within this linear range. We now describe
further details and several options that are useful when specific instrument
and experiment designs dictate.
5.3.1 Series of Standard Solutions (or Serial Dilution) Method
When the instrument readout is totally free of matrix effects, sample

loss effects, or any other such potential error-causing effects, the method
described thus far works well. The usual procedure is to accurately
prepare a stock standard solution of the analyte and then to prepare the
series of standards by diluting this stock. Chapter 3 gives the specifics
of various solution preparation schemes which may be employed here
for both the stock and the series of standards, including preparing solutions
by weighing a pure solid chemical and also by dilution. A possibility is
to prepare the stock, dilute it to make the first standard, dilute the first
to make the second, and the second to make the third, etc. This is called
"serial dilution." In either case, the solution to be diluted is pipetted into
a volumetric flask and diluted to the mark with either the pure solvent,
often distilled water, or a solution determined to be an approximation
of the sample matrix, if that is important, or having other additives, such
as for pH adjustment. It should be mentioned that stock solutions of many
analytes, especially metals (atomic absorption standards), are available
commercially, and this eliminates the need to devote lab time to such
preparation. This not only saves time, but also minimizes the possibility
of error in this part of the procedure.
5.3.2 Internal Standard Method
In some instrumental procedures, the instrument readout can vary

from standard to standard not only because of the concentration differences,
but also because of uncontrollable experiment parameters. Such parameters
may involve, for example, the problem of irreproducibility of sample
injection volume in gas chromatography or nonlinear changes in solution
viscosity as analyte concentration increases in flame AA. The solution to
these problems is the Internal Standard Method.
In this method, all standards and the sample are spiked with a constant
known amount of a substance to act as what is called an internal standard.
The purpose of the internal standard is to serve as a reference point for
instrument readout values for the analyte such that in a ratio of the
readout for the analyte to the readout for the internal standard, the effect
cancels out. Thus, if one were to plot this ratio vs the analyte concentra
tion, one expects a linear relationship, and the problem goes away. Slight
variations in gas chromatography (GC) injection volume, for example, are
compensated for by the fact that the internal standard peak size and the
analyte peak size (refer to Chapter 9) are both affected proportionally by
such variations. A plot of the peak size ratio vs analyte concentration
would be free of peak size variations due to irreproducible sample volume
and thus would give accurate results.
Solution preparation here is identical with the Series of Standard
Solutions Method, except that the constant known amount of internal
standard is added (usually pipetted) to all standards and also to the
unknown.
5.3.3 Method of Standard Additions
It is possible for the analyte readout in an instrumental procedure to

be either suppressed or enhanced by some components of the sample.
There is therefore a potential for error when these components are present
in the sample but not in the standards. One solution to this problem
would be to add the interfering substances to the standards as well so
that the effect would be measured in both the standards and the sample.
This procedure is called "matrix matching." The requirements of matrix
matching are (1 ) that the interfering substances be identifiable and (2 ) that
the concentrations of the interfering chemicals in the sample be known
so that they can be matched in the preparation of the standards. The
absence of either one or both of these would make matrix matching
impossible. The answer is the Method of Standard Additions.
In this method, small amounts of a standard solution of (or pure)
analyte are added to the sample, and the absorbance is measured after
each addition. In this way, the interfering components need not be
identified, and the sample matrix is always present at the same component
concentrations as in the sample, with only perhaps a minor dilution.
The plotting procedure and the determination of the unknown
concentration is altered somewhat, however. The plot is a graph of
instrument readout vs concentration added. The first point to be plotted
would be for 0 added (the sample readout), and the readout would
increase (presumably linearly) for each addition of analyte. Extrapolation
of the resulting line to zero readout (the x intercept), as shown in Figure
5.5, results in a length of x-axis on the negative side of zero added, which
represents the concentration in the unknown as shown. In this method,
one must presume that the plot is linear between the real zero and zero
added, since the standards will not encompass that concentration region.
This method can be used in cases in which there is some sample
preparation as well; for example, in cases in which lanthanum needs to
be added in an atomic absorbance analysis for calcium. Once the
pretreatment establishes the sample matrix, the standard additions can
be performed and data graphed.
Since some sample may be consumed, such as in sample aspiration
in AA or sample injection in GC, prior to making the second and subsequent
additions of the analyte, the standard additions method could result in
an error due to concentration changes that result. One way to partially
compensate is to prepare a series of standard solutions using the sample
matrix as the diluent. With either method, volumes of highly concentrated
(or pure) analyte can be quite small (on the order of microliters) so that
the dilution effect is negligible. A correction factor for the dilution can
also be calculated. (Figure 5.6 shows the data from a GC experiment
which was standard additions.)
FIGURE 5.5 An example of a graph for a "standard additions" absorbance

experiment.
5.4 EFFECT OF SAMPLE PRETREATMENT ON CALCULATIONS
The concentration obtained from the various graphs of the instrument

readout (or ratio, etc.) vs concentration is rarely the final answer in an
instrumental analysis. In most procedures, the sample has undergone
some form of preanalysis treatment prior to the actual measurement. In
some cases, the sample must be diluted prior to the measurement. In
other cases, a chemical must be added prior to the measurement, possibly
changing the analyte's concentration. In still other cases, the sample is a
solid and must be dissolved prior to the measurement.
The instrument measurement is the measurement of the solution tested,
and the concentration found is the concentration in that solution. What
the concentration is in the original, untreated solution or sample must
then be calculated based on what the pretreatment involved.
Frequently this is merely a "dilution factor." At other times, it is a
calculation of the grams of the constituent from the molar concentration
of the solution. Some examples of this follow. Refer to Chapter 3 for a
discussion of the parts per million unit.
FIGURE 5.6 A series of chromatograms superimposed to illustrate standard ad

ditions in GC. The peak size of the analyte grows (arrow) while the
others remain unchanged. (From Kenkel, Analytical Chemistry for
Technicians, Lewis Publishers, Inc., Chelsea, MI, 1988. With permis
sion.)
Example:
A water sample was tested for iron content, but was diluted prior to
obtaining the instrument reading. This dilution involved taking 10 mL
of the sample and diluting it to 100 mL. If the instrument reading gave
a concentration of 0.891 ppm for this diluted sample, what is the
concentration in the undiluted sample?
Solution:
mg x 0.100L = 0.0891 mg Fe in flask
This is also the milligram of Fe in the 10 mL of undiluted sample. Thus,
0.0891 mg Fe „ n< „ . . . .
-------------- — = 8.91 ppm Fe in original water
0.010 L water 6
Alternate Solution:
The "Dilution Factor" is
0.100 L
or
0.010 L
Therefore, 0.891 x 10 = 8.91 ppm Fe.
Example:
A 1 .0 0 0 -g soil sample was analyzed for potassium content by extracting
the potassium using 10.00 mL aqueous ammonium acetate solution. The
soil was then rinsed, and the solution was diluted to exactly 50.00 mL.
If the concentration in this 50 mL was found to be 5.27 ppm, what is the
concentration of the potassium in the soil in ppm?
Solution:
Ijr mg K 5.27 m g /L x 0.05000 L

ppm K —---------- —--------------------------------------
kg soil 0 .0 0 1 0 0 0 kg
5.27 mg
x 50 = 264 ppm in soil
5.5 USE OF COMPUTERS
The role that computers play in the analytical laboratory is discussed

both here and in Chapter 13. We will describe here four important
functions that microprocessors, microcomputers, and computers in general
perform in the analytical laboratory. These are (1) data acquisition, (2)
data manipulation and storage, (3) graph plotting, and (4) experiment
control.
5.5.1 Data Acquisition
It is very commonplace today for the analyst to use a computer for

acquiring data. The term data acquisition refers to the fact that the data
that is normally fed into a recorder (like so many of the instrumental

techniques described in this text) can also simultaneously be fed into and
acquired by a computer, perhaps even alongside the recorder and sharing
the same output line from the instrument. Such data is thereby stored
in short term in the computer's memory and, in the long term, on disk
or tape. The actual experiment, however, requires the use of another piece
of hardware most of the time because most computers will accept only
"digital" signals, while the output signal from an instrument is most often
of the continuous or "analog" variety. Thus one needs what is called an
"analog to digital" ("A to D") converter, sometimes referred to as an
"interface".* The continuously changing nature of an analog signal is
evident on the chart recording of a chromatography peak or molecular
absorption spectrum, for example. The A to D converter samples this data
at predetermined time intervals and feeds these "digital" values into the
computer where it is stored and/or displayed on the cathode-ray tube
(CRT) screen. Individual discreet "digital" values are what the computer
then sees.
The A to D converter, while described here as a separate piece of
hardware, can be incorporated either into a circuit board designed to plug
into a slot inside the microcomputer or within the instrument itself such
that an output for computer data acquisition is provided. In either
configuration, it appears externally that the computer is actually accepting
an analog signal, which in a sense it is.
When in actual use, a program is often run on the computer which
activates the A to D converter, establishes the sampling time interval and
other parameters at the discretion of the operation, and begins the data
acquisition at the touch of a key on the computer keyboard.
5.5.2 Data Manipulation and Storage
Any voltage signal generated by laboratory instruments and output

to a recorder can be fed into a computer in the manner just described.
This includes pH meters, spectrophotometers, chromatographs,
polarographs, etc. Even automated analysis recorder signals can be fed
into a computer. The real value of this, however, lies in the fact that the
The term "interface" is also used to describe the connection of any piece of hardware to
a computer and not just an A to D converter.
data can be permanently stored on a computer disk or tape and recalled

later for the purpose of manipulation, calculation, or whatever form of
"reduction" is needed for a given type of data.
The value of being able to permanently store data on disk or tape is
in the achievement of a "paperless" laboratory. Before the advent of the
computer, huge cumbersome volumes of chart recordings had to be
stored in the laboratory. Today, if the analyst wants to review data
acquired even years earlier, he/she can call the data from disk or tape
and view it in a matter of seconds.
5.5.3 Graph Plotting
Most instrumental analytical procedures require the plotting of a linear

graph discussed earlier in this chapter for the purpose of establishing the
exact relationship between the instrument reading and the concentration
of the analyte. The least squares fit of the data points to the line is
important in order to be as accurate as possible with the location of the
line. Also, the calculation of the correlation coefficient is important. We
have already indicated that a computer can be used to perform these
functions. Indeed, the use of a computer to plot data, perform a least
square fit, print out a copy of the fitted curve, and also calculate the
correlation coefficient are very important computer functions in the modem
lab. The improved accuracy of the determination can be demonstrated
by manual plotting and curve fitting and comparing the answer to what
a computer found. You will find that the speed of the determination also
improves dramatically.
5.5.4 Experiment Control
Finally, entire experiments can be controlled with the use of

microcomputers and microprocessors. Pressing a key on a computer
keyboard can cause an experiment to begin, be altered, or to end. Indeed
the more sophisticated modem instruments have the microprocessor for
this built right into the instrument (essentially a microcomputer complete
with disk drive and CRT and the instrument all in one unit) such that
experiment control is quite easily accomplished.
Essentially any instrument can be told when to begin executing its

function in this manner, even GC chromatographs in which the experiment
begins when the operator pushes in the syringe plunger. An electrical
contact closure can be incorporated into the syringe or injection port such
that the syringe injection actually activates the computer for data
acquisition.
C hapter 6
M olecular S pectroscopy
6.1 INTRODUCTION
More than half of all instrumental methods of analysis involve the

measurement of either light absorption or light emission by a sample. For
this reason, the analyst needs to have a basic understanding of the modern
theory of light and parameters of light, namely wavelength, frequency,
wave number, and energy, as well as the concept of how light interacts
with matter and the theory behind the emission and absorption
phenomena. This chapter introduces these concepts and continues with
perhaps the oldest and some of the most popular of all instrumental
analysis techniques, that of ultraviolet, visible, and infrared molecular
absorption analysis. The final sections deal with NMR and mass
spectrometry.
6.1.1 Nature and Parameters of Light
The modern theory of light says that light has a "dual nature." This
means that light exhibits both the properties of waves (the wave theory)
103
and the properties of particles (the particle theory). The wave theory says
that light travels from its source via a series of repeating waves, much
like waves of water move across the surface of a body of water. The
particle theory says that light consists of a stream of particles called
"photons" emanating from the source. For our purposes, the wave theory
seems to have the most applicability.
The one important difference between waves of light and waves of
water moving across a body of water is that the light waves are not
mechanical waves like water waves. Light waves do not require matter
to move or to exist. Rather than being mechanical disturbances, they are
electromagnetic disturbances, and as such, they can travel through a
vacuum, from the sun to the earth, for example. Since light waves are
electromagnetic disturbances, light is often referred to as electromagnetic
radiation. It has an electrical component and a magnetic component. Of
particular importance in analytical chemistry are the wavelength,
frequency, and energy of light as described by the wave theory.
Since light consists of a series of repeating waves, the physical distance
from a point on one wave to the same point on the next wave is an
important parameter. This distance is termed the wavelength. It is given
the Greek symbol lower case lambda (X) (see Figure 6.1). Wavelength can
vary from distances as little as fractions of atomic diameters to several
miles. This suggests the existence of a very broad spectrum of wavelengths.
Indeed, the distinction between visible (vis) light, ultraviolet (UV) light,
infrared (IR) light, etc. is the magnitude of the wavelength. Each of these,
along with the others mentioned previously, encompass a particular
"region" of the total electromagnetic spectrum. Thus, we have the vis
region, the UV region, the IR region, etc. Figure 6.2 shows a representation
of the entire electromagnetic spectrum, wavelength increasing left to
right, and indicates the approximate wavelength borders of the various
regions in nanometers. In the UV, vis, and IR regions, which are the
regions that we will be emphasizing, the nanometer (nm) and the
micrometer (or micron — pm) are the most commonly used units of
wavelength.
Another parameter of light derived from the wave theory is frequency.
Frequency is defined as the number of waves that pass a fixed point in
1 sec (cycles per second). It is given the Greek symbol nu (v). Frequency
would obviously depend on how fast the wave travels. However, all light
travels at the same speed in a vacuum, 3.00 x 101 0 cm/sec, the "speed
Molecular Spectroscopy 105
FIGURE 6.1 The definition of wavelength.
■Visible
Gamma X- UV IR Micro Radio

rays rays waves waves
10° 10 ' 103 105 107 10 '*
A (nm)
FIGURE 6.2 The electromagnetic spectrum.
of light." * Thus, a change in wavelength is the only change that would

produce a different frequency. In other words, a particular wavelength
corresponds to a particular frequency. Figure 6.3 should help clarify this
concept. The most common unit of frequency is the hertz (hz) or sec-1.
Speed, wavelength, and frequency are mathematically related by the
following equation.
A,= - (6 .1 )
D
in which c is the speed of light, 3.00 x 101 0 cm/sec.
A parameter closely related to frequency is the wave number. Wave
number is defined as the reciprocal of the wavelength when the wavelength
is expressed in centimeters and is given the symbol t> (read "nu bar").
In a medium other than a vacuum, the speed of light is different from that in a vacuum.
This creates a phenomenon known as refraction, a parameter known as refractive index,
and a technique known as refractometry. These will be discussed in Chapter 10.
FIGURE 6.3 An illustration of how frequency changes with wavelength. A light

with a shorter wavelength will have more waves (cycles) per second.
Wave number is an important parameter for the IR region, as we will

see.
Last, but certainly not least, is energy. Different energies are associated
with the different wavelengths and frequencies of light. Mathematically,
we define energy as follows:
E = hu (6-3)
and in terms of wavelength,
E = h (-) (6.4)
X
The symbol "h" is a proportionality constant known as "Planck's

Constant" which has the value 6.62 x 10~2 7 erg sec/photon. The photon,
as indicated previously, is the name given to a "particle" of light in the
particle theory. The erg is a unit of energy in the metric system.
As we consider the theories of the interaction of UV, visible (vis), and
infrared (IR) light with matter and the theories of molecular and atomic
absorption in this chapter and the next, it is important to understand what
the above four equations mean in terms of how one parameter relates
to any of the others (see Table 6.1). In addition, we can indicate which
regions of the electromagnetic spectrum are high energy regions, which
are low energy regions, which are high frequency regions, which are low
frequency regions, etc. (see Figure 6.4 for this). Concerning the three
regions to be emphasized in this chapter, then, we can say the following:
Table 6.1 Statements of Interpretation of Equations 6.1, 6.2, 6.3, and
Equation Interpretation
6.1 Frequency is inversely proportional to wavelength.

As frequency increases, wavelength decreases and vice
versa. Long wavelength corresponds to low frequency.
6.2 Wave number is inversely proportional to wavelength.
As wavelength increases, wave number decreases and vice
versa. Long wavelength corresponds to low wave number.
6.3 Energy is directly proportional to frequency.
As frequency increases, energy increases and vice versa.
High frequency corresponds to high energy.
6.4 Energy is inversely proportional to wavelength.
As wavelength increases, energy decreases and vice versa.
Long wavelength corresponds to low energy.
- 1 - 1-
gamma X-rays UV vis IR microwaves radio and
rays TV waves
high energy--------------- low energy

high frequency---------- — low frequency
short wavelength------- ■long wavelength
FIGURE 6.4 Comparisions of the various regions of the electromagnetic spectrum

in terms of energy, frequency, and wavelength.
energy: UV > vis > IR

wavelength: UV < vis < IR
frequency: UV > vis > IR
The visible region of the spectrum (vis) as shown in Figures 6 . 2 and

6.4 is a very "narrow" region compared to the others. Visible of course
means that to which our human eye is sensitive. Thus it is apparent that
we can see only a very small fraction of all the wavelengths of light that
continuously criss-cross our atmosphere and universe. In addition, those
individual wavelengths of visible light that enter our eye are
characteristically interpreted, i.e., different wavelengths correspond to
different colors. Thus we have red light, yellow light, green light, etc.
Since the different colors correspond to different wavelengths, they also
correspond to different energies and frequencies (see Figure 6.5).
I I I I I I I
I I I I I I I
I I I I I I I
Violet Blue Green Yellow Orange Red
I I I I I I I
I I I I I I I
I I I I I I I
350 450 495 570 590 620 750
Wavelength, nm
FIGURE 6.5 The visible region of the electromagnetic spectrum and the approximate
wavelength boundries of the colors.
6.1.2 Light Absorption and Emission
Evidence of light absorption abounds in our world. All colored objects

constitute such evidence. Visible white light that comes from the sun, a
light bulb, or any visible light source actually consists of all the visible
wavelengths represented in Figure 6.5. Consider a rainbow, for example.
White light coming to earth from the sun can, under the right conditions,
be dispersed into the rainbow colors. The right conditions usually means
a large concentration of water droplets in the atmosphere. This is evi
dence that white light consists of all the colors and that these colors are
separable. Why does a red sheet of paper appear red? All the wavelengths
of visible light from the visible light source in the room are absorbed
except for the red wavelengths, and these are reflected to our eye. Why
does a solution of potassium permanganate appear to be a deep purple
color? The wavelengths of visible light which are incident on the solution
from the light in the room are all absorbed except for those in the violet
and red ends of the visible region. The result is an intense purple color.
How is it that light can be absorbed? What is happening in the molecular
structure of a colored sample which results in the absorption of some
wavelengths of light more than others? Elementary atomic and molecular
theory gives us the answers. First, atoms and molecules, of which all
matter is composed, can exist in a variety of energy levels depending on
the state of their electrons. The electron clouds which are present outside
the nucleus of atoms are layered in levels of various energy. Some of these
levels, called "electronic" energy levels, have electrons in them and some
do not. The reason for this is the varying number of electrons in the
different atoms. No atom has enough electrons to fill all the levels. The
electron configuration of lithium, for example, is ls ^ s 1, while that of
carbon is ls 2 2 s2 2 p2. The 2p level is present in a lithium atom, but, unlike
carbon, does not have any electrons in it. Theoretically, all atoms have
an infinite number of electron energy levels and so all atoms have many
vacant levels above the level of the outermost electron. This is an important
point because it means that atoms can "absorb" energy from an energy
source so as to have its electrons promoted to higher vacant levels. Since
a wavelength of light is associated with a certain amount of energy, this
amount of energy in the form of light can be absorbed in order to promote
an electron. Such a promotion is called an electronic energy transition.
Thus, if light consisting of exactly the same energy as the difference
between two electronic levels shines on a sample containing an atom
which has these two levels, the light will "disappear" or be absorbed. The
energy that once was light has been transferred to the atom. The original
level, if it is the lowest possible level in which the electron can be found,
is called the "ground state." All other levels are called "excited states"
(see Figure 6 .6 ).
Secondly, molecules exist in "vibrational" and "rotational" energy
levels, as well as electronic levels. First consider vibrational levels. The
bonds between the atoms in a molecule can behave as springs. When two
balls connected by a spring are pulled apart and let go, the balls will move
back and forth as the spring is stretched and contracted, constituting a
"vibration" or, in the case of a molecular bond, a new vibrational energy
level. Thus, a molecule can absorb energy resulting in the promotion to
a higher vibrational energy level. Such a vibrational energy transition can
mean a number of different vibrational modes for the bonds (see Figure
6.7). A molecule can therefore have a number of vibrational levels to
which it can be promoted. As with the electronic energy level model, the
energy needed to promote a molecule to a higher vibrational level can
be provided by a wavelength of light, and this light can be absorbed,
resulting in the promotion to this higher level. Similarly, a molecule can
be made to rotate with the absorption of the appropriate energy, and thus
there are also rotational energy levels.
Absorptions in the UV/vis/IR regions of the electromagnetic spectrum
result in just these kinds of transitions — electronic, vibrational, and
rotational. Electronic transitions are higher energy transitions (more energy
required) than vibrational or rotational transitions. Electronic transitions
are caused by UV and vis light; vibrational and rotational transitions are
caused by IR light.
FIGURE 6.6 The absorption of light by an electron causing the promotion from
an energy level Eo (ground state) to an energy level Ej (excited state).
The energy that once was light now belongs to the electron.
FIGURE 6.7 Three vibrational modes for a 3-atom molecule. (From Kenkel, J.,
Analytical Chemistry for Technicians, Lewis Publishers, Inc., Chelsea,
MI, 1988. With permission.)
Vibrational levels are present in all electronic levels of molecules, and

rotational levels are present in all vibrational levels. This means that when
UV or vis light is absorbed, the transition can be from any vibrational
or rotational level in any electronic level to any vibrational or rotational
level in any higher electronic level. However, when IR light is absorbed,
the transition can only be between two vibrational levels (short wavelength
or "fundamental" IR) in the same electronic level or between two rotational
levels within the same vibrational level (long wavelength or "far" IR). The
initial electronic level is always the ground electronic state (see Figure 6 .8 ).
This places a greater restriction on the absorption possibilities for IR light
and opens up a very large number of possibilities for UV /vis light. Only
certain wavelengths of IR light have a chance of being absorbed for a
particular molecule. However, many different wavelengths can be ab
sorbed in certain parts of the UV and vis regions.
These facts have an effect on the manner in which these techniques
are used in the lab, as we will see. Similar statements can be made
concerning the applicability of light absorption techniques to atoms as
opposed to molecules. The absorption possibilities with atoms are greatly
restricted due to the total absence of vibrational and rotational levels.
However, "atomic" absorption techniques are no less important than
molecular techniques, as should be evidenced by the discussions in Chapter
7.
Finally, emission of light by a molecular sample can also occur. The
term for such an emission is "fluorescence." Fluorescence occurs when
v2
V, e 2
Vo
(a) (b)
FIGURE 6.8 (a) The vibrational transitions possible in the first three vibrational
levels of a molecule, (b) The electronic transitions possible in a
molecule's first three electronic levels in which there are three
vibrational levels each.
an electron that has been promoted to an excited state loses the energy
it has gained and drops back to a lower state, although not all the way
back to the original state. The energy that is lost is on the order of light
energy, but light of a different wavelength than was originally absorbed,
since it does not drop all the way back. The sample thus appears to
"glow" due to the different wavelength. An analysis technique utilizing
this phenomenon, called fluorometry, will be described later in this chapter.
6.2 NAMES OF TECHNIQUES AND INSTRUMENTS
The absorption of light at particular wavelengths is obviously an

important measurement. It is therefore important for all instruments
designed to measure such absorption to have, as part of the absorption
measurement process, the capability of resolving an entire spectral region
(polychromatic light) into these particular wavelengths ("monochromatic"

light). Two general terms which describe all techniques involving light,
including those requiring monochromatic light as just described, are
"spectroscopy" and "spectrometry." The general term for the instrument
used is "spectrometer." Any spectrometer that resolves polychromatic
light into monochromatic light and utilizes a "photomultiplier" tube
(described later) as a detector of light is properly called a "spectropho
tometer" and the corresponding technique "spectrophotometry." This
latter term correctly applies to UV and vis instruments, but technically
not to IR instruments, since in that case, as we will see, the detector is
not a photomultiplier tube. Nonetheless, IR spectrometers are often re
ferred to as spectrophotometers. It is most correct to refer to an IR instru
ment as a spectrometer and to the technique as IR spectrometry. In
addition, an instrument constructed only to be used in the vis region is
often called a colorimeter, since the monochromatic light and the solu
tions measured would display a color. The technique which utilizes a
colorimeter is called colorimetry. This term is also often used to describe
a technique in which the color of an unknown is visually matched (rather
than with an instrument) to a set of standards in order to determine
concentration.
6.3 UV/vis SPECTROPHOTOMETRY
6.3.1 General Description
We will first study the analytical technique involving the absorption

of UV and vis light. This technique, like the IR technique to be discussed
in the next section, involves the use of a laboratory instrument that
measures the degree of light absorption by a sample. In addition to the
detector (Section 6.2), there are other features that distinguish the UV/
vis instruments from the IR instruments, as we will see. The design of
the UV/vis instrument is such that a monochromatic wavelength of light
from a light source (inside the instrument) is allowed to strike a sample
solution. The amount of light absorbed by this solution is electronically
measured by a photomultiplier tube and displayed on a readout device.
The analyst is able to quantitate a constituent in the sample by relating
this degree of absorbance displayed by the instrument to the constituent's
concentration. In addition to this quantitative analysis, a qualitative analysis
can also be performed by observing the pattern of absorption that a

sample exhibits over a range of wavelengths, the so-called "molecular
absorption spectrum." No two such patterns from any two chemical
species are exactly alike. We therefore have what can be called a molecular
"fingerprint," and this is what makes identification, or qualitative analysis,
possible. The modem instrument is capable of both the qualitative and
quantitative measurements. The specific schemes by which qualitative
and quantitative analysis is accomplished will be described, but let us
begin by detailing the inner workings of the instrument.
6.3.2 Instrument Design
There are two types of UV /vis instruments in common use: the single
beam instrument and the double-beam instrument. These instruments
have many common features and utilize the same components. First, the
single-beam instrument and its components will be described, and this
will be followed by a description of the double-beam instrument. A
diagram of the basic single-beam spectrophotometer is shown in Figure
6.9. The light source (polychromatic) provides the light to be directed at
the sample. The wavelength selector, or monochromator, isolates the
wavelength to be used. The sample holder/compartment is a light-tight
"box" in which the sample solution is held, and the detector/readout
components are the electronic modules which measure and display the
degree of absorption. Let us describe each of these components in more
detail.
6.3.2a The Light Source
The light source for the vis region is different from the light source for
the UV region. Instruments which have the capability of measuring the
absorption in both regions must have two independently selectable light
sources. For the vis region, a light bulb with a tungsten filament is used.
Such a source is very bright and emits light over the entire vis region and
into the near IR region. The intensity of the light varies dramatically
across this wavelength range (Figure 6.10). This creates a bit of a problem
for the analyst, and we will discuss this later.
For the UV region, the light source is usually the deuterium discharge
Source Wavelength Sample Detector Readout

selector compartment
FIGURE 6.9 The essential components of a spectrophotometer.
Wavelength ■-■»>
FIGURE 6.10 A graph showing approximately how the intensity of light from a
tungsten filament source varies according to wavelength.
lamp. Its wavelength output ranges from 185 to about 375 nm, satisfac
tory for most UV analyses. Here again the intensity varies with wave
length.
6.3.2b The Monochromator
The monochromator, or wavelength selector, consists of three main

parts: an entrance slit, a dispersing element, and an exit slit. In addition
to these, there is often a network of mirrors situated for the purpose of

aligning or collimating a beam of light before and after it contacts the
dispersing element. A slit is a small circular or rectangular opening cut
into an otherwise opaque plate, such as a painted metal plate. The size
of the opening is often variable — a "variable slit width". The entrance
slit is where light enters the monochromator from the source (see Figure
6.11). Its purpose is to create a unidirectional beam of light of appropriate
intensity from the multidirectional light emanating from the source. Its
slit width is usually variable so that the intensity of the beam can be
varied; the wider the opening the more intense the beam.
After passing through the entrance slit, the beam strikes a dispersing
element. The dispersing element disperses the light into its component
wavelengths. For vis light, for example, this would mean that a beam of
white light is dispersed into a spray of rainbow colors, the violet/blue
wavelengths on one end to the red wavelengths on the other, with the
green and yellow in between. The monochromatic light is then selected
by the exit slit. As the dispersing element is rotated, the spray of colors
moves across the exit slit such that a particular wavelength emerges at
each position of rotation. The exit slit width can be variable too, but
making it wider would result in a wider wavelength band (the "band
pass") passing through, which is usually undesirable. The light emerging
from this exit slit is therefore monochromatic and is passed on to the
sample compartment to strike the sample. The concept is the same for
UV light (see Figure 6.11). The rotation of the dispersing element is
accomplished by either manually turning a knob on the face of the
instrument or internally by programmed scanning controls. The position
of the knob is coordinated with the wavelength emerging form the exit
slit such that this wavelength is read from a scale of wavelengths on the
face of the instrument or on a readout meter. For manual control, the
operator thus simply dials in the desired wavelength.
The dispersing element is either a diffraction grating or a prism. A
prism is a three-dimensional triangularly shaped glass or quartz block
as indicated in Figure 6.9. When the light beam strikes one of the three
faces of the prism, the light emerging through another face is dispersed.
A diffraction grating is used more often than a prism. A diffraction
grating is like a highly polished mirror that has a large number of precisely
parallel lines or grooves scribed onto its surface. Light striking this surface
is reflected, diffracted, and dispersed into the component wavelengths
as indicated in Figure 6.11. See Figure 6.12 for more explicit diagrams of
a prism and a diffraction grating.
Exit Dispersing element

slit /
To V
sample
Entrance
slit
FIGURE 6.11 The monochromator component (see text for full description).
(a) (b)
FIGURE 6.12 (a) A prism and (b) a diffraction grating.
6.3.2c Sample Compartment
Next, the light from the monochromator passes through the sample
in the sample compartment. For the UV/vis instrument, this is a light
tight box in which the container holding the sample solution is placed.
The container is called a "cuvette." The material making up the cuvette
walls is of interest.
For optimum performance, this material must be transparent to all
wavelengths of light that may pass through it. For vis light, this means
that the material must be completely clear and colorless. Inexpensive
materials, such as plastic and ordinary glass, are perfectly suitable in that
case. However, such materials may not be transparent to light in the UV
and IR regions. For the UV region, the more expensive quartz cuvettes
must be used, while in the IR region, the walls of the cuvette are often
made of inorganic salt crystals (see Section 6.4).
6.3.2d Detector/Readout
The most common detector for UV/vis spectrophotometry is the

photomultiplier tube, which is comparable to a solar cell. When light of
a particular intensity strikes it, a current, or electrical signal, of propor
tional magnitude is generated, amplified, and sent on to the readout. The
process is not energy efficient like a solar cell needs to be, however, since
it requires more electricity to run it than it generates.
The photomultiplier tube consists of a "photocathode," an anode, and
a series of "dynodes" for multiplying the signal, hence its name. The
dynodes are situated between the photocathode and the anode. A high
voltage is applied between the photocathode (an electrode which emits
electrons when light strikes it) and the anode. When the light beam from
the sample compartment strikes the photocathode, electrons are emitted
and accelerated, because of the high voltage, to the first dynode where
more electrons are emitted. These electrons pass on to the second dynode,
where even more electrons are emitted, etc. When the electrons finally
reach the anode, the signal has been sufficiently multiplied as to be treated
as any ordinary electrical signal able to be amplified by a conventional
amplifier. This amplified signal is then sent on to the readout in one form
or another. Figure 6.13 illustrates this process.
Most instruments offer a choice as to what is displayed on the readout,
since there are several parameters that can have analytical importance or
convenience. These parameters, notably "transmittance," "percent trans
mittance," "absorbance," and analytical concentration, will be defined
and discussed. The display of each of these requires associated math
ematical operations so as to obtain it from the intensity of the light to
which the detector's output signal is proportional. Thus, electronic cir
cuits which compute a ratio of intensities (transmittance), the percent
transmittance, the negative logarithm of this ratio (absorbance), etc. are
also often included as part of the total circuitry of these instrument
components. In addition, the signal may be digitized, using an analog
to digital converter (Chapter 5) for display on a digital readout or for
output to a computer.
FIGURE 6.13 The process occurring in a photomultiplier tube.
6.3.2e Calibration
The calibration of the single-beam instrument that has just been de
scribed is important to consider. Such calibration involves adjustment of
the "dark current" control when all the light is physically blocked from
striking the detector and adjustment of the entrance slit opening to the
optimum value when no analyte species is present in the path of the light.
In order to understand these two steps, we need to define some param
eters involved.
The intensity of light striking the detector when a "blank" solution (no
analyte species) held in the cuvette is present is given the symbol " 1 0."
The blank is a solution that contains all chemical species that will be
present in the standards and samples to be measured (at equal
concentration levels) except for the analyte species. Such a solution should
not display any absorption and thus I0 represents the maximum intensity
that can strike the detector at any time. When the blank is replaced with
a solution of the analyte, a less intense light beam will be detected. The
intensity of the light for this solution is given the symbol "I" (see Figure
6.14). The fraction of light transmitted is thus I/I0. This fraction is defined
as the "transmittance," "T."
Absorbing species not present
Absorbing species present
FIGURE 6.14 Illustration of the definitions of I and IQ. (From Kenkel, J., Analytical
Chemistry for Technicians, Lewis Publishers, Inc., Chelsea, MI, 1988.
With permission.)
The "percent transmittance" is similarly defined.
%T = T x 100 (6.6)
Transmittance and percent transmittance (%T) are two parameters that

are able to be displayed on the readout. Calibration thus involves
adjustment to 0% T using the dark current control when the light is
physically blocked from the detector and adjustment to 100% T using the
entrance slit opening control when the blank solution is in the path of
the light. Opening or closing the entrance slit increases or decreases the
light intensity eventually striking the detector and thus is useful for such
calibration. Following calibration, the samples and standards, in the same
or identical cuvettes (in order to measure only the effect of the solution),
are consecutively placed in the path of the light and the readout recorded.
6.3.2/ Double-Beam Instruments
The discussion in this section thus far has been concerned with instru
ments in which a single beam of monochromatic light passes through the
sample compartment. While such instruments are in very common use
(they are comparatively inexpensive), an instrument design which uti
lizes two beams of the light passing through the sample compartment
is also quite common and offers some important advantages.
In any spectrophotometer, if the intensity of the light passing through
the sample changes following calibration and before a sample or standard
is read, an error will result. Such a change can occur due to fluctuations
in the power supply to the source and/or detector (the line voltage), due
to an unstable source, or when the operator selects a different wavelength
via the monochromator control. The intensity of the light changes dra
matically as the wavelength through the monochromator is changed
(refer back to Figure 6.10). If a given experiment involves a change in the
wavelength, such as when determining the pattern of absorption over a
range of wavelengths (the "molecular absorption spectrum"), or if one
suspects that the line voltage fluctuates, recalibration with the blank is
necessary periodically or each time the wavelength is changed. This can
result in an extraordinarily long and tedious procedure and/or a loss of
accuracy, since the procedure of recalibration with the blank can take up
to 5 to 10 sec and a power fluctuation can occur in this length of time.
The double-beam design allows the blank to be checked and calibration
to take place only a split second before the sample is read. Not only does
this take much less time, eliminating the need for continuous manual
monitoring of the blank, but it also increases accuracy, since the time
between blank calibration and sample measurement is dramatically
decreased. It also allows rapid wavelength scanning in order to
conveniently obtain the molecular absorption spectrum.
A schematic diagram of a typical double-beam design for UV/vis
spectrophotometry is shown in Figure 6.15. The light coming from the
monochromator is directed along either one of two paths with the use
of a "chopper." The chopper in this case is a rotating circular half-mirror
used for splitting a light beam into two beams. At one moment, the light
passes through the sample, while at the next moment it passes through
the blank. Both beams are joined again with a beam combiner, such as
another rotating half-mirror, prior to entering the detector. The detector
sees alternating light intensities, I and Iq, and thus immediately and
automatically compensates for fluctuations and wavelength changes,
usually by automatically widening or narrowing the entrance slit to the
monochromator. If the beam becomes less intense, the slit is opened; if
the beam becomes more intense, the slit is narrowed. Thus the signal
relayed to the readout is free of effects of intensity fluctuations that cause
errors.
Sample compartments in such instruments have two cuvette holders,
one for a cuvette containing the sample or standard and one for a cuvette
containing the blank. The two beams of light pass through the sample
compartment, one through the blank (the reference beam) and one through
the sample or standard (the sample beam). The two cuvettes must be
matched in terms of pathlength and reflective and refractive properties.
Reference
. / Diai ii\
.... - / ■ ■ - j- / - r -
A u
Light from
monochromator To detector
r ^ - 7 - .
beam compartment
FIGURE 6.15 A schematic diagram of a typical double-beam design.

Scanning double-beam UV /vis spectrophotometers have become very

commonplace in analytical laboratories. Modern instruments can include
substantial microprocessor control, including control of scan time, output
functions, and data storage, as well as internal storage of Beer's Law data
such that sample concentrations can be displayed on the readout. These
instruments also have output terminals capable of transferring
concentration and absorption/transmittance and wavelength data to an
external recorder or computer.
6.3.3 Qualitative and Quantitative Analysis
The transmittance and percent transmittance of a sample have been

defined (Equations 6.5 and 6.6). The unfortunate aspect of transmittance
is that it is not linear with concentration. Low concentrations give a high
transmittance, and high concentrations give a low transmittance. However,
the relationship is not linear but logarithmic (see Figure 6.16a). Without
a linear relationship, the concentration of an unknown sample cannot be
determined by the usual procedures outlined in Chapter 5. However,
since the relationship is logarithmic, the logarithm of the transmittance
can be expected to be linear. Thus, we define the parameter of "absorbance"
as being the negative logarithm of the transmittance and give it the
symbol "A":
A = -log T (6.7)
Absorbance is a parameter then that increases linearly with concentration

(see Figure 6.16b). Absorbance is thus the parameter that is important for
quantitative analysis. If transmittance is measured by an instrument, it
must be converted to absorbance via Equation 6.7 (or by semilog graph
paper) before plotting vs concentration and obtaining the unknown
concentration according to procedures outlined in Chapter 5. However,
as indicated previously, most instruments have the electronic circuitry for
calculating absorbance built into the detector/readout system, and thus
are capable of displaying absorbance on the readout, including on a
recorder, such as for a display of the molecular absorption spectrum. Such
a display is useful for a qualitative analysis, since, as indicated previously,
it is a molecular fingerprint of the system studied. Let us study quantitative
and qualitative analysis in UV/vis spectrophotometry beginning with
this latter concept.
(a) (b)
FIGURE 6.16 (a) A plot of percent transmittance vs concentration, (b) A plot of

absorbance vs concentration.
6.3.3a Qualitative Analysis
The molecular absorption spectrum, the pattern of absorption over a

range of wavelengths, and the molecular "fingerprint" of a particular
chemical species, has been referred to at several junctures in this chapter.
We are now ready to expand on this concept and look at some examples.
The absorption pattern can be displayed as either a plot of absorbance
vs wavelength or transmittance vs wavelength. Since absorbance and
transmittance are complimentary, the transmission spectrum appears as
an inversion of the absorption spectrum. Some examples of absorption
spectra are given in Figure 6.17 and some examples of transmission
spectra in Figure 6.18. All represent molecular fingerprints — no two
chemical species will display identical absorption and transmission spectra.
Qualitative analysis can simply involve a matching-up of spectra, known
with unknown.
Absorption in the UV region warrants additional comment. Certain
characteristics of the structure of organic compounds will display certain
unique characteristics in UV absorption spectra. In the first place, organic
structures with nothing but single bonds, such as alkanes and ordinary
alcohols, do not absorb at all in the UV region. In fact, they are often used
as solvents for compounds that do absorb. However, structures with
double bonds, triple bonds, or benzene rings have very strong absorption
patterns. All of these types of structures contain n electrons, and k elec
trons are especially susceptible to UV absorption. Any such group present
124
Analytical Chemistry Refresher Manual
nm nm nm
FIGURE 6.17 The molecular absorption spectra of (a) toluene in cyclohexane, (b) potassium permanganate in water, and (c) methyl
red in water.
nm nm nm
Molecular Spectroscopy
FIGURE 6.18 The transmission spectra of (a) benzene in cyclohexane, (b) iron/o-phenanthroline in water, and (c) copper sulfate
in water.
125
in a structure, causing the appearance of absorption bands in the UV (or

vis) region, is called a "chromophore." A -C =C - bond, a -C = 0 bond, a
-N = 0 bond, and a benzene ring are thus all classified as chromophores.
If two or more chromophores appear in the same molecule, the location
of one relative to the other dictates a particular pattern. If they are
separated by more than one carbon, the absorption pattern represents a
simple summation of the two individual patterns. If the two chromophores
are on adjacent carbons, an apparent "shifting" (called a "bathochromic"
shift) of the absorption maximum to a longer wavelength is usually
observed, and the absorbance is increased (a "hyperchromic" effect). If
the opposite is observed (a shift to a shorter wavelength), it is called a
"hypsochromic" effect. When the two chromophores are attached to the
same atom, the two observations just noted also occur, but to a lesser
extent. Additional shifting of the absorption maximum can occur when
an ordinarily nonabsorbing group is attached near a chromophore. Such
a group is called an "auxochrome." An example would be an -OH group
attached near a double bond. Additionally, effects of solvent and pH are
important to observe. Thus, in addition to fingerprinting (spectra matching),
observations of shifting of absorption maxima and absorbance intensity
can be very important in a qualitative analysis.
6.3.3b Quantitative Analysis
The exact relationship between absorbance and concentration is a

famous one. It is known as the Beer-Lambert Law, or simply as Beer's
Law. A statement of Beer's Law is
A = abc (6.8)
in which A is absorbance, a is "absorptivity" or "extinction coefficient,"

b the "pathlength," and c the concentration. Absorbance was defined
previously in this section. Absorptivity is the inherent ability of a chemical
species to absorb light and is constant at a given wavelength. Pathlength
is the distance the light travels through the measured solution. It is the
inside diameter of the cuvette. Pathlength is measured in units of length,
usually centimeters or millimeters. Concentration can be expressed in a
variety of units, usually, however, in molarity, parts per million, or grams
per 100 mL. The units of absorptivity depend on the units of these other
parameters, since absorbance is a dimensionless quantity. When the

concentration is in molarity and the pathlength is in centimeters, the units
of absorptivity must be liters/(mole centimeter). Under these specific
conditions, the absorptivity is called the "molar absorptivity," or the
"molar extinction coefficient," and is given a special symbol, the Greek
letter epsilon, (e). Beer's Law is therefore sometimes given as
A = ebc (6.9)
Defining the molar absorptivity parameter presents analytical chemists

with a standardized method of comparing one spectrophotometric method
with another. The larger the molar absorptivity, the more sensitive the
method. (It is not unusual for molar absorptivity values to be as large
as 10,000 L/(mol cm) and higher.) In addition, it was stated above that
the absorptivity is constant "at a given wavelength," implying that it
changes with wavelength. The greatest analytical sensitivity occurs at the
wavelength at which the absorptivity is a maximum. This is the same
wavelength that displays the maximum absorbance in the molecular
absorption spectrum for that species. In fact, the molecular absorption
spectrum is sometimes shown as a plot of absorptivity vs wavelength
rather than absorbance vs wavelength. For a given absorbing species,
such a plot would display the same characteristic shape and the same
wavelength of maximum of absorbance.
Beer's Law constitutes the spectrophotometry application of the
discussion in Chapter 5 which states that in most instrumental quantitative
analyses, an instrument readout is proportional to concentration. In this
case, absorbance is the readout. Thus, most quantitative analyses by
Beer's Law involve preparing a series of standard solutions, measuring
the absorbance of each in identical cuvettes, and plotting the measured
absorbance vs concentration, creating the so-called "standard curve." The
absorbance of an unknown solution is then measured and its concentration
determined from the graph. Such a graph is often called a Beer's Law
Plot (see Figure 6.19).
Of course, an unknown's concentration can be determined by comparing
its absorbance with just one standard,
in which cu is the concentration of the unknown, cs is the concentration

Sample Data
A C(ppm) 0 .5 0-
0.102 0.500 Aa °-40 “

0.199 1.000 0 .3 0 -
0.310 1.500
0.400 2.000 0 .2 0 -
0.506 2.500
0.375 unk 0 .1 0 -
1.0 1.5 2.0 2.5 3.0

C
FIGURE 6.19 Some sample data and a Beer's Law Plot of the data showing the
determination of the unknown concentration.
of the standard, A^ is the absorbance of the unknown, and As is the

absorbance of the standard. It may also be determined by direct calcu
lation, if the absorptivity value and pathlength are precisely known.
c =4 (6 .1 1 )
ab
See Section 5.2 for a more thorough discussion of each of these methods,
including limitations, and also the calculations that are often involved
after the unknown solution concentration is determined.
6.3.3c Interferences and Deviations '
Interferences are quite common in qualitative and quantitative analysis

by UV/vis spectrophotometry. An interference is a contaminating
substance that gives an absorbance signal at the same wavelength or
wavelength range selected for the analyte. For qualitative analysis, this
would show up as an incorrect absorption spectrum, thus possibly leading
to erroneous conclusions if the contaminant was not known to be present.
For quantitative analysis, this would result in a higher absorbance than
one would measure otherwise. Absorbances are additive. This means that
the total absorbance measured at a particular wavelength is the sum of
absorbances of all absorbing species present. Thus, if an interference is
present, the correct absorbance can be determined by subtracting the

absorbance of the interference at the wavelength used, if it is known. The
modem solution to these problems is to utilize separation procedures,
such as extraction or liquid chromatography, to separate the interfering
substance from the analyte prior to the spectrophotometric measurement.
These techniques will be discussed in Chapters 8 and 10.
Deviations from Beer's Law are in evidence when the Beer's Law Plot
is not linear. This is probably most often observed at the higher
concentrations of the analyte (see Figure 6.20). Such deviations can be
either chemical or instrumental. Instrumental deviations occur because
it is not possible for an instrument to be accurate at extremely high or
extremely low transmittance values— values that are approaching either
0% T or 100% T. The normal working range is between 15 and 80%,
corresponding to absorbance values between 0.10 and 0.82. It is
recommended that standards be prepared so as to measure in this range
and so that unknown samples be diluted if necessary. Chemical
interferences occur when a high or low concentration of the analyte
causes chemical equilibrium shifts in the solution which directly or
indirectly affect its absorbance. It may be necessary in these instances to
work in a narrower concentration range than expected. This means that
unknown samples may also need to be further diluted as in the instrumental
deviation case.
6.4 IR SPECTROMETRY
6.4.1 Introduction
As discussed in Section 6.1, the absorption of IR light causes vibrational

energy transitions in molecules. The usefulness of the technique lies
especially in the fact that only very specific wavelengths (energies) of IR
light are able to be absorbed when a single particular kind of molecule
is in the path of the light. As with UV/vis absorption, the absorbance vs
wavelength plot is a molecular fingerprint of the molecule. The difference,
however, lies in the fact that in the IR region, the absorption bands are
extremely sharp, and each such band is associated with a particular
covalent bond present in the molecule. We referred earlier to the ball and
FIGURE 6.20 A Beer's Law Plot that shows a deviation from linearity at higher
concentrations.
spring model of a molecule and how the initiation of a vibration of the

spring constitutes an energy transition caused by an IR wavelength. The
point here is that the wavelength, at which the sharp absorption band
is observed, depends on what atoms the "spring" connects and whether
the bond is a single bond, a double bond, etc. Thus, a -C -H - bond absorbs
a particular wavelength, a -C -O - bond a different wavelength, a -C = 0
yet a different wavelength, etc. Additionally, various wavelengths can be
absorbed depending on what mode of vibration is involved. These can
be stretching vibrations, rocking and bending motions, etc. (refer back to
Figure 6.7). The technique is especially useful therefore for qualitative
analysis. Not only is the IR absorption spectrum a molecular fingerprint,
but the presence of particular bonds in the structure will be manifested
in corresponding sharp absorption bands at particular wavelengths, a fact
that has considerable usefulness in the narrowing down of the possibili
ties in a qualitative analysis scheme. This topic will be expanded upon
later in this section.
The nature of the sample holder is important to consider. In Section
6.3, we referred to the need for the sample holder in IR spectrometry to
be composed of inorganic salt crystals. Since IR wavelengths are of such
energy so as to cause covalent bonds to vibrate, any covalent compound
would not be suitable as sample holder material because such material
will absorb its characteristic wavelengths of IR light. This, in turn, may
obviously cause erroneous conclusions in qualitative analysis schemes.
Glass or plastic materials therefore cannot be used. Material with ionic
bonds, however, is not a problem because ionic bonds do not absorb IR
wavelengths— they do not undergo vibrational energy transitions. Thus,
inorganic compounds, such as NaCl and KBr, are commonly used as

matrix material and sample "windows" in IR spectrometry.
6.4.2 Liquid Sampling
For pure liquids ("neat" liquids) and liquid solutions, sandwiching a

thin layer of liquid between two large NaCl or KBr crystals (windows)
is the classic procedure for mounting the sample in the path of the light.
Typical dimensions for such windows is about 2 cm wide x 3 cm long
x 0.5 cm thick. Positioning or holding the crystals in place is done using
either a "sealed cell," a "demountable cell," or a combination "sealed
demountable" cell. "Sealed" cells are permanent fixtures for the windows
and cannot be disassembled. They have a fixed pathlength and are very
useful for quantitative analysis, since the pathlength is reproduced.
"Demountable" cells can be disassembled so as to change the pathlength.
The sample is placed on the window in a space created by a "spacer"
while it is disassembled. The thickness of the spacer establishes the
pathlength. It is then reassembled with the thin layer in place between
the windows. "Sealed demountable" are demountable cells which include
inlet and outlet ports for introducing and eliminating the liquid samples.
All demountable cells are designed to allow easy disassembly for changing
the spacer between the crystals (the pathlength) or for eliminating the
spacer altogether for viscous samples, for example. Figure 6.21 shows a
drawing of a typical sealed demountable cell. The top neoprene gasket
and window have holes drilled in them to coincide with the inlet and
outlet ports to facilitate filling the space, created by the spacer, with the
liquid sample. The path of the liquid sample is shown as a dashed line
in the figure.
Filling the cells with sample and eliminating the sample when finished
can be troublesome. The sample inlet and outlet ports are tapered to
receive a syringe with a Luer (tapered, ground-glass) tip. The usual
procedure for filling is to raise the outlet end by resting it on a pencil or
similar object (to eliminate air bubbles) and then to use a pressure/
vacuum system with the use of two syringes, one in the outlet port and
one in the inlet port, as shown in Figure 6.22. While pushing on the
plunger of the syringe containing the liquid sample in the inlet port and
pulling up on the plunger of the empty syringe in the outlet port, the cell
can be filled without excessive pressure on the inlet side. This reduces
■Neoprene
FIGURE 6.21 The "sealed demountable" cell assembly. (Adapted from Chia, L. and
Ricketts, S., Basic Techniques and Experiments in Infrared and FT-IR
Spectroscopy, The Perkin-Elmer Corporation, Norwalk, CT, 1988.)
FIGURE 6.22 The recommended method of filling a sealed or demountable cell.
the possibility of damaging the cell due to the excessive pressure that may
be needed, especially when working with unusually viscous samples and
short pathlengths. Tapered Teflon plugs are used to stopper the ports
immediately after filling. The cell may be emptied and readied for the
next sample by using two empty syringes and the same push-pull method.
When refilling, an excess of liquid sample may be used to rinse the cell
and eliminate the residue from the previous one. Alternatively, the cell
may be rinsed with a dry volatile solvent and the solvent evaporated
before introducing the next sample.
The analyst must be careful to protect the salt crystals from water
during use and storage. Sodium chloride and potassium bromide are, of
course, highly water soluble, and the crystals may be severely damaged
with even the slightest contact with water. All samples introduced into
the cell must be dry. This is important for another reason, of course. Water
contamination will show up on the measured spectrum and cause erro
neous conclusions. If the windows are damaged with traces of water, they
will become "fogged" and will appear to become nontransparent. The
windows may be repolished if this happens. Depending on the extent
of the damage, various degrees of abrasive materials may be used, but
the final polishing step must utilize a polishing pad and a very fine
abrasive. Polishing kits are available for this. Figure 6.23 shows the correct
method for polishing. Finger cots should be used to protect the windows
from finger moisture.
Liquids can be sampled as either the neat liquid (pure) or mixed with
a solvent (solution). The neat liquid is more desirable since the spectrum
will show absorption due to the liquid only. However, when a solution
is run, both the analyte and the solvent will produce absorption bands,
and they must be differentiated. Some solvents have rather simple IR
spectra and are thus desirable for solvents. Examples are carbon tetra
chloride (only —C— C l bonds) and methylene chloride (CH2C12). Their IR
spectra are shown in Figure 6.24. Alternatively, a double-beam instru
ment can be used to cancel out the solvent absorption. These instruments
will be discussed later in this section.
6.4.3 Solid Sampling
IR spectrometry is one of the few analytical techniques that routinely

analyzes solid undissolved samples. The techniques to be described here
include the KBr pellet, the Nujol (mineral oil) mull, and the diffuse
reflectance method.
FIGURE 6.23 The correct method for polishing crystal materials for infrared sample
cells. Note the figure eight motion.
6.4.3a KBr Pellet
The KBr pellet technique is based on the fact that dry, finely powdered
potassium bromide has the property of being able to be squeezed under
very high pressure into transparent discs — transparent to both IR light
and vis light. It is important for the KBr to be dry both in order to obtain
a good pellet and to eliminate absorption bands due to water in the
spectrum. If a small amount of the dry solid analyte (0.1 to 2.0%) is added
to the KBr prior to pressing, then a disc (pellet) can be formed from which
a spectrum of the solid can be obtained. Such a disc is simply placed in
the path of the light in the instrument and the spectrum measured.
Two methods of pressing the KBr pellet will be described here. First,
a pellet die consisting of a threaded body and two bolts with polished
faces may be used. One bolt is turned completely into the body of the
die. A small amount of the powdered sample, enough to cover the face
of the bolt inside, is added, and the other bolt is turned down onto the
sample, squeezing it into the pellet (see Figure 6.25). The two bolts are
then carefully removed. The body of the die is placed in the instrument
so that the light beam passes directly through the center of the die.
The other method utilizes a hydraulic press and a cup die. The cup
die consists of a base, with a protruding center, and a hollow cylinder
that fits snugly over the protrusion. With the cylinder in place, the
powdered sample is added to the cylinder to cover the face of the
protrusion, and a second metal piece with a protrusion is placed on the
top of the assembly. The entire assembly is placed into a laboratory press
(see Figure 6.26).
It is important in either case for the KBr and sample to be dry, finely
powdered, and well mixed. An agate mortar and pestle is recommended
for the grinding and mixing of the KBr and sample.
6.4.3b Nujol Mull
The Nujol (mineral oil) mull is also often used for solids. In this method,
a small amount of the finely divided solid analyte (1-2 pm particles) is
mixed together with an amount of mineral oil to form a mixture with
a toothpaste-like consistency. This mixture is then placed (lightly squeezed)
between two NaCl or KBr windows which are similar to those used in
the demountable cell discussed previously for liquids. If the particles of
solid are not already the required size when received, they must be finely
ground with an agate mortar and pestle and can be ground directly with
mineral oil to create the mull to be spread on the window. Otherwise,
a small amount (about 10 mg) of the solid is placed on one window along
with one small drop of mineral oil. A gentle rubbing of the two windows
together with a circular or back and forth motion creates the mull and
distributes it evenly between the windows. The windows are placed in
the demountable cell fixture and placed in the path of the light.
A problem with this method is the fact that mineral oil is a covalent
compound and its characteristic absorption spectrum will be found
superimposed in the spectrum of the solid analyte, as with the solvents
used for liquid solutions discussed previously. However, the spectrum
is a simple one (Figure 6.27) and often does not cause a significant
problem.
6.4.3c Diffuse Reflectance
With the advent of the Fourier Transform Infrared Spectrometer (FTIR)

instrumentation (see below), a technique called diffuse reflectance is
becoming popular for solids. In this technique, the powdered KBr/analyte
mixture (about 5% analyte) is placed in a small sample cup, and the light
beam shines directly on this powdered sample. The diffuse reflectance,
or the light returning from the surface of the sample that is scattered (and
not simply reflected), is measured by the FTIR system and the spectrum
136
WAVELENGTH IN M ICRO NS
600
FIGURE 6.24 (a) The infrared spectrum of CC14. (b) The infrared spectrum of CH2 CI2 .
WAVELENGTH IN M O O N S
FIGURE 6.24b.
137
FIGURE 6.25 The procedure for making a KBr pellet with the use of a threaded
pellet die.
FIGURE 6.26 A pellet-making assembly which utilizes a laboratory hand press.

(Reproduced from Spectra-Tech, Inc., Stamford, CT. With permis
sion.)
displayed. The advantage of this method is that an analyte that does not
form a good KBr pellet can be run with little or no problem.
The traditional IR spectrometer is very similar to the UV/vis spectro

photometer, the so-called double-beam "dispersive" instrument utilizing
the traditional slit/dispersing element/slit monochromator to create a
narrow bandwidth, often referred to as the single wavelength, to shine
on the sample. The modem IR instrument, however, utilizes an inter
ferometer in which the light does not get dispersed, and yet the traditional
IR spectrum is measured — It is called the Fourier Transform Infrared
Spectrometer, FTIR. Both of these designs will now be discussed.
6.4.4a Double-Beam Dispersive
A diagram of the double-beam dispersive IR spectrometer is shown

in Figure 6.28. It is very similar to the UV /vis spectrophotometer described
in Section 6.3. There are some differences, however. The light source is
typically a nichrome wire coil that has a high electrical resistance and
emits intense heat (IR light) when an electric current is passed through
it, much like the burner on an electric stove. The source may also be a
"Globar," which is a silicon carbide rod, or a "Nemst Glower," which
is a rare earth oxide cylinder. Both of these operate on basically the same
principle as the nichrome wire. The detector is a thermocouple transducer,
a device which converts heat energy into an electric signal.
The IR installment is typically a "double-beam in space" installment,
whereas the UV/vis instrument is a "double-beam in time" installment.
"Double-beam in time" refers to the fact that the light beam from the
source is split into two beams existing alternately in time. At one moment,
the sample beam exists and passes through the sample compartment,
while at the next moment, the reference beam exists and passes through
the sample compartment. A rotating half-mirror (chopper) ahead of the
140
WAVELENGTH IN M OONS
4.5 5 5.5
PERCENT TRANSMISSION
FIGURE 6.27 The infrared spectrum of mineral oil.

Thermocouple
FIGURE 6.28 A diagram of a double-beam dispersive IR instrument. (Reproduced

from Beckman Instruments, Inc., Fullerton, CA. With permission.)
sample compartment accomplishes this splitting of the beam (see Section

6.3). In the "double-beam in space" instrument, both light beams exist
simultaneously. Two mirrors that point in different directions, as de
picted in Figure 6.28, send two light beams through the sample compart
ment. Thus, both beams exist simultaneously and pass through the sample
compartment together. The monochromator system, located between the
sample compartment and detector in such an instrument, does utilize a
chopper to combine the light beams, however, and the detection system
is synchronized with the chopping frequency so as to differentiate the
two signals and display the percent transmittance of the unknown sample
at the recorder. As mentioned under liquid sampling (Section 6.4.2),
placing the solvent of the analyte species in the reference path will cause
all solvent absorption bands to disappear from the recorded spectrum
just as the blank adjustment is automatically made in the UV/vis
instruments discussed earlier. This does, however, require two exactly
matched IR cells for accurate work. This cell matching is less important
for qualitative work, which is a more common application of the technique.
The optical system for dispersive IR instruments has an additional
requirement. The absorption of IR light by the mirrors, lenses, and
dispersing element must be minimized. Thus, various reflection and
transmittance gratings composed of nonabsorbing materials are used,
often in combination with transmission filters, such as the "filter wheel"
in Figure 6.28. Front-reflecting mirrors are used, and glass collimating
lenses are absent. In addition, more than one grating are needed to
accommodate the different regions of the IR spectrum, and this fact
requires the instrument to stop scanning to allow the realignment of these
gratings in the middle of a run and then to resume. Figure 6.28 shows
two gratings. The gratings are usually made of glass or plastic and coated
with aluminum.
6.4.4b Fourier Transform Infrared Spectrometry (FTIR)
The modem FTIR instrument is designed to perform the same func

tions as the dispersive instruments. Such an instrument, however, does
not utilize a light dispersing monochromator and associated optics. The
light from the source, typically the same type of source as in the dispersive
instruments, passes through the optical path undispersed. The result,
however, is the same — the infrared absorption spectrum, the plot of
percent T vs wavelength, but it can be obtained much faster than with
the dispersive instrument. We will undertake a simplified discussion as
to how this happens.
In the FTIR instrument, the undispersed light beam passes through the
sample, and all wavelengths and the corresponding absorption data are
received at the detector simultaneously. A computerized mathematical
manipulation known as the Fourier Transform is performed on this data
in order to obtain absorption data for each individual wavelength. To
present the data in a form that can utilize the Fourier Transform, the wave
pattern created by combined constructive and destructive interference of
all wavelengths of light from the source over time is utilized. This pattern,
known as an interferogram, is created by moving one beam of light from
the source through another. The device for doing this is known as an
interferometer.
Figure 6.29 shows a diagram of an interferometer. It consists of a beam
splitter and two mirrors, one fixed and one movable. Consider first a light
source of a single wavelength. As light from the source strikes the beam
splitter, the beam is split such that half of the intensity is transmitted to
the movable mirror while half is reflected to the fixed mirror. Both beams
are reflected back to the splitter where they join again and proceed toward
the sample. If the distance from the splitter to the movable mirror is
exactly equal to the distance to the fixed mirror, then their rejoining will
result in constructive interference and the intensity reaching the sample
will be a maximum. The position of the movable mirror, however, may
also be such that complete destructive interference occurs. This would
occur if the movable mirror distance is equal to the fixed mirror distance
plus half of the wavelength. Other movable mirror distances would result
To sample
FIGURE 6.29 A diagram of an interferometer (see text for discussion).
(a) (b)
FIGURE 6.30 The two interferograms described in the text. (Reproduced from Chia,
L. and Ricketts, S., Basic Techniques and Experiments in Infrared and FT-
IR Spectroscopy, The Perkin-Elmer Corporation, Norwalk, CT, 1988.
With permission.)
in intermediate intensities, the cycle would repeat as the distance is

increased, and thus the "interferogram," known as a cosine wave, shown
in Figure 6.30a would result. Now consider a light beam consisting of
all wavelengths in the IR — in other words, the light from the light source
in a typical IR instrument. In this case, the interferogram in Figure 6.30b
would result. The strong intensity in the center occurs when the distances
of both mirrors from the splitter are equal and we have complete con
structive interference of all wavelengths. Motor-driving the movable
mirror to small distances toward and away from the equidistance point
"codes" the light and its wavelengths and creates the possibility for it to
be "decoded" by the Fourier Transform mathematical manipulation by
computer at the detector resulting in the IR spectrum of the sample.
The advantages of the FTIR over the dispersive technique are (1) it is
faster, making it possible to be incorporated into chromatography schemes
as we will see briefly in Chapters 9 and 10, and (2) the energy reaching
the detector is much greater thus increasing the sensitivity.
6.4.5 Qualitative and Quantitative Analysis
The ultimate goal of IR analysis is, of course, the identification of the

substance measured or the determination of the quantity of the substance
measured. The usefulness of the technique lies mostly in the identification
aspects, and we will emphasize this here. However, we will also briefly
address the quantitative aspects as well.
6.4.5a Qualitative Analysis
Early on in this section (Section 6.4), we spoke of how specific bonds

present in molecules give rise to absorption bands at specific correspond
ing wavelengths in the IR region. This is the basis of qualitative analysis
using this technique, and we will now expand on this idea, especially
discussing what wavelengths are absorbed by each type of bond.
The region of the electromagnetic spectrum involved here of course
is the IR region. This spans the wavelength region from about 2.5 to about
17 pm, or, in terms of wave number, from about 4000 to about 600
cm-1. IR spectra are transmission spectra calibrated most often in wave
number, and so we will be speaking in terms of wave number for the
remainder of this discussion and depicting the spectra with a 100% T
baseline at the top of the chart and the peaks deflecting toward the
0% T level when an absorption occurs (see Figure 6.31 for an example).
The absorption pattern derived from the particular molecule present in
the path of the light is a molecular "fingerprint" just as it was in the UV/
vis region (Section 6.3). The IR spectrum, however, is useful for an
additional reason. The absorption bands, as they are typically recorded,

appear much sharper and we have the ability to conclude that a molecule
has a particular type of bond in its structure when we observe the
corresponding characteristic absorption band in the spectrum. This is
particularly true in the region from about 4000 to about 1500 cm- 1 and
less true in the 1500 to 600 cm- 1 region. This latter region is thus often
described separately as the "fingerprint" region and is used mostly to
match, peak for peak, the spectrum of an unknown with a spectrum of
a known, perhaps from a library of known spectra, such as the Sadtler
library* (see Figure 6.32). Thus, we look for characteristic bands in the
4000 to 1500 cm- 1 region to perhaps assign the unknown to a particular
class of compounds, i.e., to narrow down the possible structures, and then
look to the fingerprint region and the overall spectrum to make the final
determination, matching peak for peak.
In addition to the location (i.e., wave number) of the absorption bands,
it can also be useful to examine the width and depth of the absorption.
The descriptions "broad" and "sharp" are often used to describe the
width of the peaks, and "weak," "medium," and "strong" are used to
describe the depth of the peaks. To understand these descriptions, refer
to Figure 6.33. Spectra "a" and "c" show broad, strong peaks centered
around 3300 cm-1. Spectrum "b" shows a sharp, strong peak at 1685
cm-1. Spectrum "c" shows a sharp, medium peak at 1220 cm- 1 and a series
of weak peaks between 1700 and 2000 cm-1.
Table 6.2 correlates the location, width, and depth of various IR
absorption patterns for some common kinds of bonds. By correlating
Figure 6.33 with Table 6 .2 , it should be obvious that spectrum "a" in
Figure 6.33 is that of an alcohol with no benzene rings, spectrum "b" is
the spectrum of a compound containing a carbonyl group (such as an
aldehyde) with no benzene ring, and spectrum "c" is the spectrum of an
alcohol with a benzene ring. Figure 6.34 is a correlation chart showing
the location of the absorption peaks of most bonds.
Finally, earlier the need for using a solvent such as carbon tetrachlo
ride, dichloromethane, or nujol (mineral oil) was indicated for some
applications. The analyst needs to know what the spectra of these solvents
look like in order to be able to account for the interfering solvent peaks.
Figures 6.24 and 6.27 gave the IR spectra for these compounds.
This refers to the collection of standard infrared spectra published by Sadtler Research
Laboratories, Philadelphia, PA.
FIGURE 6.31 The infrared spectrum of toluene. Infrared spectra are transmission spectra with the peaks recorded from the top down
as shown.
Wave number (cm 1)

4000 3500 3000 2500 2000 1500 1000
PEAK ID REGION FINGERPRINT REGION
FIGURE 6.32 The "fingerprint" and "peak ID" regions of infrared spectra.
6.4.5b Qmntitative Analysis
Quantitative analysis procedures using IR spectrometry utilize Beer's

Law. Once the %T or absorbance measurements are made, the data
reduction procedures are identical with those outlined previously in this
chapter for UV/vis spectrophotometry. This means that the analyst could
prepare just one standard solution to which to compare the unknown
using the ratio and proportion scheme (Equation 6.10), determine the
concentration by direct calculation (Equation 6.11), or prepare a series of
standards and utilize a Beer's Law plot (see Figure 6.19 and accompa
nying discussion).
Reading the %T from the recorded IR spectrum for quantitative analysis
can be a challenge. In the first place, there can be no interference from
a nearby peak due to the solvent or other component. One must choose
a peak to read that is at least nearly, if not completely, isolated.
Secondly, the baseline for the peak must be well defined. Since the
baseline may not be completely straight across the wavelength region,
the analyst does not endeavor to have the instrument trace across the
100%T line. Rather, something less than 100%T is chosen at the beginning
of the run such that the baseline, even though it likely will not be straight,
remains on scale (less than 100%T) for the entire scan. Thus, two %T
readings must be taken, one for the baseline (corresponding to where the
baseline would be if the peak were absent — a "blank" reading) and one
for the minimum %T, the tip of the peak. The two readings are then
FIGURE 6.33 The IR spectra of (a) n-butyl alcohol, (b) isobutyraldehyde, and (c) benzyl alcohol (see text for discussion).
FIGURE 6.33b.
IN M
WAVELENGTH
Table 6.2 Some Easily Recognizable IR Absorption Patterns
Bond Description of Absorption Pattern
-C -H -, where C is Sharp, strong peak "on the low side of 3000

not part of a benzene cm '1," between about 2850 and 3000 cm-1
ring
-C -H -, where C is a Sharp, medium peak "on the high side of 3000
part of a benzene cm , uciwcui auuui JUUU a n j i uu cm
ring or double bond
-C -H -, where C is a Sharp, medium peak "on the high side of 3000
part of triple bond cm-1," between about 3250 and 3350 cm-1
-O -H -, in alcohols, Broad, strong peak centered at about 3300 cm'
phenols, and water,
for example
-C = 0 , in aldehydes, Sharp, strong peak at about 1700 cm-1
ketones, etc.
-C -C -, in benzene Two sharp, strong peaks near 1500 and 1600
ring cm '1, and a series of weak peaks (overtones)
between 1600 and 2000 cm-1, the latter in a
case of a monosubstituted benzene ring
converted to absorbance, and the absorbance of the baseline is subtracted

from the absorbance of the peak. The result is the absorbance of the
sample. It is recommended, for accurate work, that two or more spectra
of each standard and the unknown be recorded and an average of each
absorbance calculated.
It should also be mentioned that the pathlength of the sample cell used
must be constant for all standards and the unknown, or at least known
so that a correction can be applied if necessary. Using care when filling
and cleaning cells is also important to avoid alterations in pathlength due
to excessive pressure or leaking.
6.5 FLUOROMETRY
Fluorometry is an analytical technique which utilizes the ability of

some substances to exhibit fluorescence. Fluorescence is a phenomenon
in which the substance appears to glow when a light shines on it. In other
words, light of a wavelength different from the irradiating light is released
or "emitted" following the absorption process. Most often the irradiating
light is UV light, and the emitted light is vis light. The phenomenon is
explained based on light absorption theory and what can happen to a
Approximate infrared absorption frequencies of various groups
152
Wavenumber (cm *1)
FIGURE 6.34 A correlation chart for IR spectroscopy. (Reprinted from Zubrick, J.W., Organic Chemistry Survival Manual Copyright
© John Wiley & Sons Inc., New York, 1988. With permission.)
chemical species in order to revert back to the ground state once the
absorption — the elevation to an excited state — has taken place. Fluo
rescence can occur with both molecules and atoms. The present discus
sion will focus on molecules and complex ions. Atomic fluorescence will
be discussed in Chapter 7.
All atoms and molecules seek to exist in their lowest possible energy
state at all times. When a molecule is raised to an excited electronic energy
state through the absorption of light, it is no longer in its lowest possible
energy state and will seek to lose the energy it gained any way it can.
Most often, the energy is lost through mechanical means, such as through
collisions with other chemical species in the solution. However, there can
be a direct jump back to the ground state with only some intermediate
stops at some lower vibrational states in between. With such a jump back
to the ground state, the energy the molecule gained as a result of the
absorption process is lost in the form of light, and since it is light of less
energy due to the accompanying small energy losses in the form of
vibrational loss, the wavelength is longer. See Figure 6.35 for a graphical
picture of this process.
The instrument for measuring fluorescence intensity for quantitative
analysis is constructed with two monochromators, one to select the
wavelength to be absorbed and one to select the fluorescence wavelength
to be measured. In addition, the instrument components are configured
so that the fluorescence measurement is optimized to be free of interfer
ence from transmitted light from the source.
This latter point means that the fluorescence monochromator and
detector are not placed in a straight line with the source, absorption
monochromator, and sample (such as in an absorption spectrophotometer),
but are rather placed at a right angle as shown in Figure 6.36. Thus, there
is a "right angle configuration" in a fluorometer to avoid any interference
from the transmitted light from the light source.
Colored glass light filters are often used for the monochromators in
fluorometers. Instruments with such filters are called filter fluorometers
and are considerably less expensive than spectrophotofluorometers, which
have standard slit/dispersing element/slit monochromators. Although
these latter instruments are excellent for determining the proper wave
lengths to be used and for precise work, filter fluorometers have proved
quite satisfactory for most routine work and are much less expensive.
We have indicated that the intensity of the fluorescence emitted by the
fluorescing species is proportional to the concentration of this species in
solution. Fluorescence intensity is therefore the parameter to be measured
Absorption ^ ) Fluorescence
FIGURE 6.35 An energy level diagram showing the transitions occurring when
absorption is accompanied by fluorescence.
- O '1
x I x
source
FIGURE 6.36 The basic fluorometer. The two monochromators can be glass filters.
(From Kenkel, J., Analytical Chemistry for Technicians, Lewis Publishers,
Inc., Chelsea, MI, 1988. With permission.)
and related to concentration. This means that the fluorescence intensity

of a single standard solution, or a series of standards, is measured and
related to concentration. A graph of fluorescence intensity vs concentra
tion is expected to be linear in the concentration range studied. Proce

dures outlined in Chapter 5, Section 5.3, and also Section 6.3 for Beer's
Law, are also applicable here.
The types of compounds that can be analyzed by fluorometry are
limited. The drop of an electron to the ground state must be accompanied
by the emission of light (it must be a direct drop). The kind of electron
which is most apt to be able to do this is an electron, and, more specifi
cally, electrons found in benzene rings. Fused benzene ring systems, such
as those in Figure 6.37, are especially highly fluorescent compounds.
Metals can be analyzed by fluorometry if they are able to form complex
ions by reaction with a ligand having the required electrons. An example
is aluminum, as in Figure 6.38.
Fluorometry and absorption spectrophotometry are competing tech
niques in the sense that both are techniques for analyzing for molecular
species. Each offers its own advantages and disadvantages, however. The
number of chemical species that can exhibit fluorescence is very limited.
However, for those species that do fluoresce, the fluorescence is generally
very intense. Thus we can say that while absorption spectrophotometry
is much more universally applicable, fluorometry suffers less from inter
ferences and is usually much more sensitive. Therefore, when an analyte
does exhibit the somewhat rare quality of fluorescence, fluorometry is
likely to be chosen for the analysis. The analysis of foods for vitamin
content is an example, since vitamins such as riboflavin and niacin exhibit
fluorescence and fluorometry would be relatively free of interference and
would be very sensitive.
6.6 NUCLEAR MAGNETIC RESONANCE SPECTROSCOPY
6.6.1 Introduction
Throughout this chapter, we have studied instrumental analysis tech

niques which are based on the phenomenon of light absorption. We have
discussed techniques utilizing light in the UV/vis region involving elec
tronic energy transitions — the elevation of electrons to higher energy
states. We have also discussed techniques utilizing light in the IR region
involving molecular vibrational and rotational energy transitions. In this
naphthalene anthracene ^3 ^ ^
riboflavin
HO-CH
HO—CH
ho - ch2oh
FIGURE 6.37 Highly fluorescent compounds. (From Kenkel, J., Analytical Chemistry
for Technicians, Lewis Publishers, Inc., Chelsea, MI, 1988. With
permission.)
(highly fluorescent)
FIGURE 6.38 The formation of a complex ion of aluminum, which is highly

fluorescent. (From Kenkel, J., Analytical Chemistry for Technicians, Lewis
Publishers, Inc., Chelsea, MI, 1988. With permission.)
section, we introduce the concepts of Nuclear Magnetic Resonance Spec

trometry (NMR). This technique utilizes light in the radio wave region
of the spectrum involving nuclear spin energy transitions which occur
in a magnetic field.
The absorption to be described is based on the theory and experimental
evidence that the nuclei of the atoms bonded to each other in molecules
spin on an axis like a top. Since any given nucleus is positively charged,
a small magnetic field exists around it. If we were to bring a spinning
nucleus into an external magnetic field, such as between the poles of a
magnet, the nucleus, representing the smaller magnetic field, will align
itself to the external field. It is possible to become aligned either in the
same direction as the external field or in the opposite direction. Alignment
\ / \ /
N N
/ \ / \
(a) (b)
FIGURE 6.39 Alignment of a spinning nucleus (a) with a magnetic field and (b)
opposed to a magnetic field.
Opposed
£
Radio wave
Alligned
FIGURE 6.40 An energy level diagram showing the transition from one nuclear
spin energy state to another.
in the opposite direction represents a slightly higher energy state (see

Figure 6.39). The energy difference between the two different alignments
is on the order of the radio wave wavelengths. Thus light in the radio
wave region can be absorbed by molecules in a magnetic field so as to
cause this nuclear spin energy transition (see Figure 6.40). While the
phenomenon just described applies to nuclei of all elements, NMR has

found its most useful application in the measurement of the hydrogen
nucleus, probably because the vast majority of organic structures contain
hydrogen atoms. For this reason, it is sometimes also referred to as Proton
Magnetic Resonance. The application lies mostly in the determination of
the structure of organic compounds; thus, it is mostly a qualitative analy
sis tool for such compounds.
6.6.2 The Traditional Instrument
Central to the instrumental design is a large magnet with the north and
south poles facing each other as shown in Figure 6.39. There are several
different types of magnets in common use, but stability and the ability
to produce a precise magnetic field are common requirements. The ability
to carefully vary the strength of the field between the two poles and the
ability to output the magnitude of the field over time to a recorder are
also requirements. A pair of coils positioned parallel to the magnet poles
and connected to a "sweep generator" permits precise scanning of the
magnetic field. Additionally, incorporated into the unit are a radio fre
quency (RF) transmitter capable of emitting a precise frequency, an RF
receiver/detector for detecting absorption, and an x-y recorder to plot the
output of the detector vs the applied magnetic field. The sample is held
in a 5-mm outside diameter glass tube containing less than 0.5 mL of
liquid, which in turn is held in a fixture called the "sample probe." A
schematic diagram of the traditional NMR spectrometer is shown in
Figure 6.41.
A precise radio frequency is emitted by the transmitter, so there is no
component needed to act as a monochromator. However, the receiver/
detector warrants additional comment. There are two designs for detec
tors. One utilizes a coil wrapped around the sample tube as the trans
mitter and a second coil arranged at right angles to the transmitter coil
as the detector. This unique design will detect a signal only if absorption
has taken place. The other design utilizes a single coil wrapped around
the sample which, with the use of an appropriately designed electronic
circuit, acts both as the transmitter and the receiver. The x-axis of the
recorder is connected to the scanning mechanism as described previ
ously, and thus with absorption plotted on the y-axis, the magnetic field
strength is plotted on the x-axis. The result is a plot of absorption vs field
strength, the so-called NMR spectrum. (See the discussion of chemical
shift below for a more precise discussion of what is plotted on the x-axis.)
A = Sample Tube, B = R-F Transmitter Coils, C = Sweep Coils

D * Detector Coil, E = Magnet.
FIGURE 6.41 Schematic diagram of an NMR spectrometer as described in the text.
(Reprinted from Shriner, R.L., et al., The Systematic Identification of
Organic Compounds, 6th ed., John Wiley & Sons, New York, 1980. With
permission.)
The magnitude of the RF frequency used is variable. Some instruments

are 60 MHz (megahertz, or one million cycles per seconds). Others are
100 MHz or higher. The magnitude of the frequency dictates the magnitude
of the magnetic field strength required. The spread between the two
energy levels in question (which corresponds to the energy of the RF
frequency) depends on the strength of the field. A 60 MHz instrument
requires a field strength of 14,092 gauss (the gauss is a unit of field
strength), while a 100 MHz instrument requires 23,486 gauss.
6.6.3 Chemical Shifts
Electrons are, of course, present around and near the hydrogen nuclei
in a molecule, and they are generating individual magnetic fields too.
These very small fields oppose the applied field giving an effective applied
field somewhat smaller than expected and therefore a slightly shifted
absorption pattern, the so-called chemical shift. If all hydrogen nuclei in

a molecule were "shielded" by electrons equally, they would all give an
absorption peak at the same frequency/field combination. Due to the
different environments surrounding the different hydrogens in an or
ganic molecule, however, we find absorption peaks at locations repre
senting each of these environments. In other words, each "type" of
hydrogen in a structure would give a characteristic absorption peak at
a specific field value. Methyl alcohol, CH3 OH, for example, would give
two peaks, one representing the methyl hydrogens and one representing
the hydroxyl hydrogen, while cyclohexane would give just one peak, etc.
The importance of this information is that (1 ) from the number of different
peaks, we can tell how many different kinds of hydrogen there are and
(2 ) from the amount of shielding shown, we can determine the structure
nearby.
Since the shifts can be extremely slight, making the actual field strength
of the peak difficult to measure precisely, a reference compound, typically
tetramethylsilane (TMS), is often added to the compound measured. All
the hydrogens in the TMS structure are equivalent (Figure 6.42). It is then
convenient to modify the x-axis to show the difference between the single
TMS peak and the compound's peaks. Thus the x-axis typically represents
a "difference factor" symbolized by the Greek letter lower case delta (8 ),
in which the single TMS peak is 0.0 ppm field strength difference and
other peaks then are so many parts per million away from the TMS peak.
Figure 6.43 shows such an NMR spectrum for methyl alcohol.
6.6.4 Peak Splitting and Integration
Additional qualitative information is possible with NMR spectra. First,

a given absorption peak, while apparently arising from one particular
kind of hydrogen, may appear to be split into two or more peaks, giving
what are termed "doublets," "triplets," etc. Also, two obviously different
kinds of hydrogens may give rise to just one peak (a singlet). Each of these
phenomena is caused by effects of the immediate environment of the
hydrogen and will present specific qualitative information. Second, the
size of a peak is indicative of the number of hydrogens it represents. This,
ch3
I
C H — S i— CH3
ch 3
FIGURE 6.42 The structure of tetramethylsilane (TMS).
Methyl
^ group hydrogens
Hydroxyl 1
group
hydrogen 1
I TMS
I
w J i
. 1 , . . . I h i iljtulfcUAlxAulilnImili mi nilim liuiLunJbutluiJlM uLntl
........... I n iluui
.. .. .. I . . . . I . . . . 1 . . . . I
10.0 9.0 8.0 7.0 6.0 5.0 4.0 3.0 2.0 1.0 0
6 (ppm)
FIGURE 6.43 The NMR spectrum of methyl alcohol. (From Wade, L.G., Jr., Organic
Chemistry, © 1987. Reprinted permission of Prentice-Hall, Englewood
Cliffs, NJ.)
too, is important qualitative information, and so many NMR spectrom

eters are equipped with integrators for determining peak size. A second
pen on the recorder is used to give an integration trace, which is then
used to determine this number of hydrogens. Figure 6.44 is the spectrum
of ethylbenzene, showing a quartet and triplet, as well as the integrator
trace.
For more details on these and other concepts of NMR, please refer to
comprehensive organic chemistry texts.
Inte
ti
Ethyl
group
Aromatic —__ _ hydrogens
hydrogen
JU. I I
10.0 9.0 8.0 7.0 6.0 5.0 4.0 3.0 2.0 1.0 0
6 (ppm)
FIGURE 6.44 The NMR spectrum of ethylbenzene. (From Wade, L.G., Jr., Organic
Chemistry, © 1987. Reprinted permission of Prentice-Hall, Inc.,
Englewood Cliffs, NJ.)
6.7 MASS SPECTROMETRY
6.7.1 Introduction
The final technique of molecular spectroscopy we will discuss is mass

spectrometry. Unlike all the others in this chapter, this technique does
not use light at all. Very briefly, the instrument known as a mass
spectrometer utilizes a high energy electron beam to cause total destruction
and fragmentation of the molecules of the sample. This fragmentation
results in small charged "pieces" or fragments of the molecules which
are then made to move through a magnetic field. The magnetic field
affects each of the fragments differently according to their mass and
charge, and thus they become separated. Finally, a detector sensitive to
these fragments is placed in their path and, in combination with a readout
device, such as a recorder or monitoring screen, allows the operator to
determine the charge to mass ratio of each fragment and to identify not
only the fragment, but the entire molecule. We will briefly discuss the
details.
There are two different instrument designs we will discuss. These are
the magnetic sector mass spectrometer and the quadrupole mass
spectrometer. Both designs consist of a system for sample introduction,
the electron beam to create the fragmentation, a magnet to create the
magnetic field, and a detection system. The entire path of the fragments,
including the inlet system, must be evacuated from 10" 4 to 10- 8 torr. This
requirement means that a sophisticated vacuum system must also be part
of the setup. The reason for the vacuum is to avoid collisions of both the
electron beam and the sample ions with contaminating particles which
would alter the results.
The difference between the magnetic sector mass spectrometer and the
quadrupole mass spectrometer lies in the design of the magnet. A dia
gram of a magnetic sector mass spectrometer is shown in Figure 6.45. In
this instrument, the magnet is a powerful, variable field electromagnetic,
the poles of which are shaped to cause a bending of the path of the
fragments through a specific angle, such as 90°. It is possible to vary the
field strength in such a way as to scan the magnetic field and to "focus"
the ion fragments of variable mass to charge ratio onto the detector slit
one at a time. In this way, specific fragments created at the electron beam
can be separated from other fragments and detected individually.
A diagram of the quadrupole mass spectrometer is shown in Figure
6.46. Here, four short parallel metal rods with a diameter of about 0.5 cm
each are utilized. These rods are aligned parallel to and surrounding the
fragment path as shown. Two nonadjacent rods, such as those in the
vertical plane, are connected to the positive pole of a variable power
source, while the other two are connected to the negative pole. Thus, a
variable electric field is created, and as the fragments enter the field and
begin to pass down the center area, they deflect from their path. Varying
the field creates the ability to "focus" the fragments one at a time onto
the detector slit, as in the magnetic sector instruments. The quadrupole
instrument is newer and more popular since it is much more compact
and provides a faster scanning capability.
Magnet
FIGURE 6.45 A magnetic sector mass spectrometer. (Adapted from Wade, L.G., Jr.,
Organic Chemistry, © 1987. Reprinted permission of Prentice-Hall,
Englewood Cliffs, NJ.)
Electron Fragment Metal "rods"
Sample
introduced
FIGURE 6.46 The quadrupole mass spectrometer.
6.7.3 Mass Spectra
The scanning of the magnetic field and the detection of fragments with
a specific mass to charge ratio creates the possibility of manufacturing
a plot of the fragment count vs mass to charge ratio. In other words, the
fractions of all specific types of fragments (of particular mass and charge)
resulting from a given sample can be determined. Such a plot is called
the mass spectrum. An example is shown in Figure 6.47. Each vertical
line evident in this figure represents a fragment of a particular mass to
charge ratio given on the x-axis. The "intensity" of the lines represents
the "count," or the number of fragments detected with that ratio. When
100
80
• t 60
c
<D
1 40
20
0
10 20 30 40 50 60 70 80 90 100 110 120 130 140 150 160
Mass to Charge Ratio ---------- ►
FIGURE 6.47 An example of a mass spectrum. (From Wade, L.G., Jr., Organic
Chemistry, © 1987. Reprinted permission of Prentice-Hall, Englewood
Cliffs, NJ.)
a certain molecule is introduced, it will be fragmented in a certain way

and its characteristic mass spectrum will always be produced. One can
see that it is a "molecular fingerprint," just as absorption spectra are
molecular fingerprints, and that it is a powerful tool for identification
purposes.
Modem mass spectrometer laboratories are linked to computers banks
containing massive numbers of mass spectra obtained over the years.
Specific identification, or at least a narrowing to specific possibilities, is
often done by a computer that accesses these spectral files.
The mass spectrometer has been used as a detector in gas chromatog
raphy instruments and, more recently, in HPLC instruments. These
combinations, referred to as GCMS and HPLC-MS, are discussed briefly
in Chapters 9 and 10.
C hapter 7
A tom ic S pectroscopy
7.1 INTRODUCTION
This chapter discusses the techniques involving the absorption and

emission of light by atoms (as opposed to molecules) and builds to some
extent on the discussions of light, parameters of light, light absorption
and light emission, and other concepts presented in Chapter 6 . If these
concepts are not well understood, we suggest that you first study the
initial sections of Chapter 6 before beginning this chapter.
The analytical techniques known as atomic absorption (AA), flame
photometry (FP), inductively coupled plasma (ICP), atomic fluorescence,
and atomic emission spectrography are all included under the heading
of "atomic" techniques or "Atomic Spectroscopy." Since the chemical
species that absorb or emit light in these cases are atoms, the techniques
are limited to sample systems from which atoms can be generated,
especially including solutions of metal ions and excluding solutions of
molecular species. Techniques for molecular species come under the
heading of "Molecular Spectroscopy" and are discussed in Chapter 6 .
Atomic spectroscopy has been applied to a wide range of metals and also
to some nonmetals. In this chapter, we will describe the theory,
instrumentation, and application of each of the above listed techniques.
167
7.2 ATOMIZATION
First, how are metal atoms generated from metal ions? There are a
number of methods. The earliest discovered method was with the use
of a flame. When solutions of metal ions are placed in a flame, the solvent
evaporates leaving behind crystals of the formerly dissolved salt.
Dissociation into atoms then occurs; the metal ions "atomize" or are
transformed into atoms. Flames are used for this purpose in most atomic
absorption instruments and in all flame photometry and atomic
fluorescence instruments. Such instruments, especially the AA, are easily
recognized because of the centralized, hooded area in which a large flame,
often 6 in. wide by 6 in. or more high, is located. All "atomizers," including
the flame, are similar energy sources. Some are of the "atomic vapor"
generator variety. Examples of these include the graphite furnace, the
Delves cup, and the borohydride vapor generator (see Section 7.6). Another
type uses an inductively coupled plasma (Section 7.7), and another uses
a spark or arc across a pair of electrodes (Section 7.8). In each case, in
addition to atomization, excitation of the atoms also occurs. The concept
of excitation will be discussed in Section 7.3. In the remaining part of this
section, we discuss the details of the flame atomizer; others will be
discussed in those later sections.
7.2.1 Fuels and Oxidants
All flames require both a fuel and an oxidant in order to exist. Bunsen
burners and Meker (Fisher) burners utilize natural gas for the fuel and
air for the oxidant. The temperature of such a flame is 1800 K maximum.
In order to atomize and excite most metal ions and achieve significant
sensitivity for quantitative analysis by atomic spectroscopy, however, a
hotter flame is desirable. Most AA and FP flames today are air-acetylene
flames — acetylene for the fuel, air for the oxidant. A maximum
temperature of 2300 K is achieved in such a flame. Ideally, pure oxygen
with acetylene would produce the highest temperature (3100 K), but such
a flame suffers from the disadvantage of a high burning velocity, which
decreases the completeness of the atomization and therefore lowers the
sensitivity. Nitrous oxide (NzO) used as the oxidant, however, produces
a higher flame temperature (2900 K), while burning at a low rate. Thus,
Atomic Spectroscopy 169
Table 7.1 A Listing Showing Which Oxidant, Air or Nitrous Oxide, is

Recommended for the Various Metals and Nonmetals Analyzed
by AA
Air:
Lithium, Sodium, Magnesium, Potassium, Calcium, Chromium, Manganese,
Iron, Cobalt, Nickel, Copper, Zinc, Arsenic, Selenium, Rubidium, Ruthenium,
Rhodium, Palladium, Silver, Cadmium, Indium, Antimony, Tellurium,
Cesium, Iridium, Platinum, Gold, Mercury, Thallium, Lead, Bismuth
Nitrous Oxide:
Beryllium, Boron, Aluminum, Silicon, Phosphorus, Scandium, Titanium,
Vanadium, Gallium, Germanium, Strontium, Yttrium, Zirconium, Niobium,
Molybdenum, Tin, Barium, Lanthanum, Hafnium, Tantalum, Tungsten,
Rhenium, Praseodymium, Neodymium, Samarium, Europium, Gadolinium,
Terbium, Dysprosium, Holmium, Erbium, Thulium, Ytterbium, Uranium
From Perkin-Elmer Corporation, Norwalk, CT; product literature.
N2 0-acetylene flames are fairly popular. The choice is made based on

which flame temperature/burning velocity combination works best with
a given element. Since all elements have been studied extensively, the
recommendations for any given element are available. Table 7.1 lists most
metals and the recommended flame for each. Air-acetylene flames are
obviously the most commonly used.
7.2.2 Burner Designs
There are two designs of burners for the flame atomizer that are in
common use. These are the so-called "total consumption burner" and the
"premix burner". In the total consumption burner (Figure 7.1), the fuel,
oxidant, and sample all meet for the first time at the base of the flame.
The fuel (usually acetylene) and oxidant (usually air) are forced, under
pressure, into the flame, whereas the sample is drawn by aspiration into
the flame through a small diameter plastic tube. The rush of the fuel and
oxidant through the burner head creates a vacuum in the sample line and
draws the sample from the sample container into the flame. This type
of burner head is used in flame photometry and is not useful for atomic
absorption. The reason for this is that the resulting flame is turbulent and
nonhomogeneous — a property that negates its usefulness in AA, since
the flame must be homogeneous for the same reason that different sample
cuvettes in molecular spectroscopy must be closely matched. One would
FIGURE 7.1 A diagram of a total consumption burner.
not want the absorption properties to change from one moment to the
next because of the lack of homogeneity in the flame. The lack of
homogeneity, however, does not affect the quality of the data obtained
with a flame photometer. The reason will become clear in Section 7.4.
The premix burner does away with the homogeneity difficulty and is
the burner typically used in flame AA. The sample is again drawn from
the sample container by aspiration through a small diameter flexible
plastic tube, nebulized, (split into a fine mist) and mixed with the fuel
and oxidant with the use of a flow spoiler (such as a set of baffles) prior
to introduction into the flame. The burner head typically used is rectangular
and has a 4- to 6 -in. long slot through which the premixed fuel, oxidant,
and sample emerge and are ignited, creating the flame. Figure 7.2 is a
diagram of this design. In the nebulizer, the sample enters an even smaller
diameter tube and then impacts a glass bead, creating the fine mist. There
is an adjustment on the nebulizer which controls the aspiration rate and
thus the amount of sample reaching the flame. This adjustment is usually
set so as to obtain the maximum absorbance on the readout. Most
instruments are equipped to accept a variety of fuels and oxidants. As
the gas combinations are varied, it is usually necessary to change the
burner head to one suitable for the particular combination chosen. A
faster burning mixture would require a burner head with a smaller slot
so as to discourage drawing the flame inside the burner head causing
a flashback.
FIGURE 7.2 A diagram of a Premix Burner. (From Kenkel, J., Analytical Chemistry
permission.)
Flashbacks can also occur when air is drawn back through the drain
line illustrated at the bottom of the premix chamber in Figure 7.2. The
drain line is necessary to allow droplets of solution that do not make it
to the flame to drain out. This problem is solved by forming a trap in
the drain line (a flexible plastic tube) and keeping the end of the drain
immersed in the waste solution contained in a bottle below the instrument.
7.3 EXCITATION
Following atomization, as indicated in the previous section, excitation

of the atoms occurs to a small extent in the flame atomizer. A very small
percentage of the atoms absorb energy from the flame and are elevated
to an excited state. The technique of flame photometry is derived from
this process as we will see. The remaining high percentage of ground state
atoms that are therefore also present in the flame are subject to excitation
with a light source. Flame atomic absorption is derived from this process.
As discussed in Chapter 6 , basic atomic theory holds that electrons in
atoms exist in energy levels around the nucleus. This theory also holds
that electrons can be moved from one energy level to a higher one if
conditions are right. These conditions consist of (1) the absorption of
sufficient energy by the electrons and (2 ) a vacancy for the electron with
this greater energy in a higher level. In other words, if an electron absorbs
the energy required to be promoted to a higher vacant energy level, then
it will be promoted to that level, and the atom will have undergone a
transition from the ground electronic state to an excited electronic state.
These "electronic energy levels" are the only ones that exist in atoms (no
vibrational levels as with molecules), and thus electronic "transitions" are
the only kind of energy transitions that can occur. This fact accounts for
many of the differences between molecular spectrophotometers and atomic
spectrophotometers and between the theories associated with each.
Since all energy transitions that take place in atoms are purely electronic,
only individual, discrete, electronic energy transitions are possible. These
transitions involve the elevation of electrons from one electronic level to
another, as depicted in Figure 7.3a. Since no vibrational transitions take
place, only a limited number of energies, those corresponding to the very
specific electronic transitions, have a chance of being absorbed. For flame
atomic absorption, in which wavelengths of light get absorbed by the
atoms, the result is a "line" absorption spectrum (Figure 7.3b) rather than
a "continuous" absorption spectrum as found for molecules. The term
"line" is used here in reference to the individual lines, or wavelengths,
of absorption evident in Figure 7.3b.
Those atoms excited by the light source in flame atomic absorption are
those that are measured by this technique. Those atoms excited by the
flame are those that are measured by flame photometry. Details of these
two techniques are given in the following two sections. As mentioned
earlier, atomization can be caused by methods other than a flame, and
thus excitation can also be caused by energy sources other than a flame
or light source. Details of techniques associated with these are given in
Sections 7.6, 7.7, and 7.8.
7.4 FLAME PHOTOMETRY
7.4.1 Introduction
Flame photometry is the technique which measures the atoms excited

by a flame (and not by a light source). This measurement is possible
because atoms that find themselves in an excited state (such as those
(a) (b)
FIGURE 7.3 (a) A hypothetical energy level diagram showing four electronic levels
and the transitions that are possible within these levels, (b) A line
absorbtion spectrum. Each transition indicated in (a) corresponds to
a line in the spectrum in (b).
found naturally from a solution aspirated into a flame) will readily lose
the gained energy in order to revert back to the ground state. Since the
magnitude of the energy lost is on the order of light energy, light is
emitted. The wavelengths of the emitted light correspond to those same
wavelengths as those that were absorbed in the flame atomic absorption
technique discussed briefly in the last section, since exactly the same
energy transitions occur, except in reverse. Figure 7.4 illustrates this
phenomenon. The spectrum shown in Figure 7.4b is a "line emission"
spectrum. Thus a line spectrum can be either an absorption spectrum or
an emission spectrum depending on the process measured.
Each individual metal has its own characteristic emission and absorption
pattern — its own unique set of wavelengths emitted or absorbed and
its own unique line emission or absorption spectrum. This is because each
individual metal atom has its own unique set of electronic levels. This
fact is demonstrated in a simple laboratory test known as the "flame test."
Sodium atoms present in a simple low temperature Bunsen burner flame
will emit a characteristic yellow light. Potassium atoms present in such
a flame will emit a violet light. Lithium and strontium atoms emit a red
light. The transitions occurring in the sodium atoms are such that the line
spectrum that is emitted corresponds to yellow light, while those occurring
in the potassium atom correspond to violet light, etc. (see Figure 7.5). The
usefulness of emission spectroscopy for qualitative analysis is thus ap
parent, especially given the fact that we can utilize monochromators and
Intensity
A-
(a) (b)
FIGURE 7.4 (a) A hypothetical set of transitions from higher electronic states to
lower electronic states, (b) The line emission spectrum that results
from the transitions in (a).
detectors to precisely measure the wavelengths involved. Indeed, all

emission spectroscopy instruments, including flame photometers, ICPs,
and emission spectrographs, are useful for qualitative analysis. Flame
photometry and ICP, however, find their greatest application in quan

titative analysis, as we will see.
7.4.2 Instrumentation
The simple flame photometer consists of three parts: the flame, a

monochromator (Chapter 6 , Section 6.3), and a detector/readout (Chapter
6 , Section 6.3). As mentioned earlier, the flame originates from the total
consumption burner. Of course, a premix burner may also be used, and

indeed instruments, i.e., flame atomic absorption instruments, that utilize
the premix burner as a matter of necessity, can also function as flame
photometers, as we will discuss shortly. Instruments, however, that are
manufactured and sold as flame photometers cannot be used as flame
AA instruments, since they are designed with a total consumption burner
and no light source to excite the atoms in the flame. Why is the total
consumption burner a satisfactory burner for FP? While the flame from
this burner is not homogeneous in its makeup, the intensity of the light
emitted by the atoms in the flame is not affected. Such emission is
normally quite stable. It is, of course, the intensity of this emission that
is measured for quantitative analysis, as we will see in the next section.
The second major component is the monochromator. In molecular
absorption instruments (Chapter 6 , Section 6.3), a monochromator is
needed to isolate the wavelength of maximum absorbance from all other
wavelengths coming from the light source. In flame photometry, there
is no light source (independent of the flame itself), and thus the mono
chromator is needed for a different reason. For quantitative analysis, the
intensity of the emitted light must be measured. For maximum sensitiv
ity, we want to measure the intensity of the most intense wavelength
emitted by the flame. The monochromator is thus tuned to this wave
length. Which wavelength is it? The flame emits the line spectrum of the
element to be measured. The most intense line in the spectrum is gen
erally the wavelength to zero in on. It is possible, however, that this line
is not the optimum line due to interferences. If this is the case, then a
different wavelength may be optimum. "Primary" and "secondary" lines,
etc. are thus often defined. Generally, a flame photometry procedure has
been well researched and the wavelength suggested in such a procedure
will be the best wavelength.
The monochromator is useful for another reason. In the molecular
spectroscopy techniques discussed in Chapter 6 , the sample cuvette is

located in a light tight box (sample "compartment"), such that room light
is not a problem. With flame techniques, however, the flame must be in
an open area of the instrument. The monochromator thus also serves to
isolate the desired wavelength from room light.
The final component of a flame photometer is the detector/readout
and associated electronics. The detector is essentially the same component
described in Chapter 6 , Section 6.3. It is a device that generates an amplified
electronic signal when light strikes it. In this case, however, the associated
electronics is designed to measure the intensity of the light, rather than
a transmittance or absorbance. Thus, the readout displays relative inten
sity. Figure 7.6 shows a schematic diagram of the flame photometer.
7.4.3 Application
Application of flame photometry can be of both a qualitative and

quantitative nature. For qualitative analysis, the determination of which
wavelengths are emitted is the key, since each element emits its own
characteristics line spectrum. Thus, one would check for the emission of
an element's wavelengths by tuning the monochromator to the expected
wavelengths for the element in question and checking for a readout of
intensity. The expected wavelengths can be obtained from tables of
emission lines such as Table 7.2.
Quantitative analysis by flame photometry is encountered more fre
quently than qualitative analysis in an analytical laboratory, however. As
implied previously, the intensity of the emitted light is directly and
linearly proportional to the concentration. Thus the standard curve for
quantitative analysis is a plot of intensity vs concentration (Figure 7.7).
As discussed in this book for other techniques, the standard curve is
prepared by measuring the readout for a series of standard solutions of
the element of interest. Table 7.2 also gives the detection limit and
application of the elements typically determined by flame photometry.
It should be pointed out that other atomic techniques, especially flame
and graphite furnace AA and ICP, have distinct advantages over flame
photometry in many respects, especially sensitivity and linear concentration
ranges. However, the flame photometry technique has been around longer
FIGURE 7.6 A schematic diagram of a flame photometer. (From Kenkel,}., Ana

lytical Chemistry for Technicians, Lewis Publishers, Inc., Chelsea, MI,
1988. With permission.)
Table 7.2 The Elements Typically Determined by Flame Photometry, Their Corre
sponding Primary Lines, Their Detection Limit, and Applications
Primary Line Detection Limit

Element (nm) (ppm) Applications
Lithium 670.8 0.00003 Water, wastewater,

biological fuids, soil
extracts, etc.
Potassium 766.5 0.0005 Water, wastewater,

biological fluids, soil
extracts, etc.
Sodium 589.0 0.0005 Water, wastewater,

biological fluids, soil
extracts, etc.
Strontium 460.7 0.09 Water, wastewater,

soil extracts, etc.
and some routine analyses are well established, especially in clinical

laboratories. Also, the instruments are much less expensive.
The flame photometer also has a specific application as a detector for
gas chromatography (see Chapter 9, Section 9.7).
FIGURE 7.7 The standard curve for quantitative analysis in flame photometry.
7.5 FLAME ATOMIC ABSORPTION
7.5.1 Introduction
Of the various atomization techniques (flame, graphite furnace, Delves

cup, and borohydride vapor generator) with which we combine the use
of a light source for excitation producing an "absorption" technique, the
flame atomizer is the oldest and the most common. We refer to these
techniques collectively as atomic absorption (AA) techniques, but the one
which uses the flame atomizer is also often referred to as simply atomic
absorption. In order to make a distinction, however, between the use of
the flame and the others, we refer to the absorption technique which
utilizes a flame as the atomizer as flame atomic absorption. This is the
technique discussed in this section. The others will be discussed in Section
7.6.
As stated previously, only a very small percentage of the atoms in the
flame are actually present in an excited state at any given instant. The
exact percentage depends on the flame temperature, but at the hottest
temperature of any flame atomizer used for AA (2900 K), the proportion
of atoms actually in the excited state at a given instant is much less than
0.01% of the total. Thus there is a large percentage of atoms that are in
the ground state and available to be excited by some other means, such
as a beam of light from a light source. Flame AA takes advantage of this
fact and uses a light beam to excite these ground state atoms in the flame.
Thus AA is very much like molecular absorption spectroscopy in that
light absorption (by these ground state atoms) is measured and related
to concentration. The major differences lie in instrument design, especially
with respect to the light source, the sample "container," and the placement
of the monochromator.
FIGURE 7.8 The Basic Flame A A Instrument. (From Kenkel, J., Analytical Chemistry
permission.)
7.5.2 Instrumentation
The basic flame AA instrument is shown in Figure 7.8. The light source,
in most cases a hollow cathode tube, is a lamp that emits exactly the
wavelength required for the analysis (without the use of a monochromator).
This light beam is directed at the flame containing the sample. The flame
standing atop the slot of the premix burner (Section 7.2) is wide (the width
of the burner head slot). This width represents the pathlength for Beer's
Law (Chapter 6 , Section 6.3) considerations. The 4-6 in. width thus aids
in determining small concentrations of the metal being analyzed. The
light beam then enters the monochromator, which is tuned to the recom
mended line, the so-called "primary" line, from the metal's line spectrum.
This line emerges from an adjustable slit opening and is thus the wave
length that strikes the detector. Since this is an absorption technique, the
electronics associated with the detector/readout is designed to display
either absorbance or transmittance on the readout.
7.5.2a The Hollow Cathode Lamp
How is it that the hollow cathode lamp emits exactly the wavelength
required without the use of a monochromator? The reason is that atoms
of the metal to be tested are present within the lamp, and when the lamp
is on, these atoms are supplied with energy which causes them to elevate
to the excited states. Upon returning to the ground state, exactly the same
wavelengths that are useful in the analysis are emitted, since it is the
analyzed metal with exactly the same energy levels that undergoes
excitation. Figure 7.9 is an illustration of this. Therefore, the hollow
cathode lamp must contain the element being determined. A typical
atomic absorption laboratory has a number of different lamps in stock
which can be interchanged in the instrument, depending on what metal
is being determined. Some lamps are "multielement," which means that
several different specified kinds of atoms are present in the lamp and are
excited when the lamp is on. The light emitted by such a lamp consists
of the line spectra of all the kinds of atoms present. No interference will
usually occur, however, since the monochromator isolates a wavelength
of our own choosing.
The exact mechanism of the excitation process in the hollow cathode
lamp is of interest. Figure 7.10 is a closeup view of this lamp and of the
mechanism. The lamp itself is a sealed glass envelope containing either
argon or neon gas (neon shown in figure). When the lamp is on, neon
atoms are ionized, with the electrons drawn to the anode (+ charged
electrode), while the neon ions (Ne+) "bombard" the surface of the cathode
(- charged electrode). The metal atoms, M, in the cathode are sputtered
from the surface and raised to the excited state as a result of the
bombardment. When the atoms return to the ground state, the characteristic
line spectrum of that atom is emitted. It is this light which is directed at
the flame, where unexcited atoms of the same element absorb the radiation
and are themselves raised to the excited state. As indicated previously,
the absorbance is measured and related to concentration.
7.5.2b Electrodeless Discharge Lamp
A light source known as the electrodeless discharge lamp (EDL) is

sometimes used. In this lamp, there is no anode or cathode. Rather, a
small, sealed quartz tube containing the metal or metal salt and some
argon at low pressure is wrapped with a coil for the purpose of creating
a radio frequency (RF) field. The tube is thus inductively coupled to an
RF field and the coupled energy ionizes the argon. The generated elec
trons collide with the metal atoms, raising them to the excited state. The
characteristic line spectrum of the metal is thus generated and is directed
at the flame just as with the hollow cathode tube. EDLs are available for
the absorption of
the lampfc emission
spectrum by atoms
in the flame.
FIGURE 7.9 The light emitted by the hollow cathode lamp is exactly the wave
length needed by the atoms in the flame. (From Kenkel, J., Analytical
With permission.)
FIGURE 7.10 The hollow cathode lamp and the process of metal atom excitation
and light emission. (From Kenkel, J., Analytical Chemistry for Technicians,
Lewis Publishers, Inc., Chelsea, MI, 1988. With permission.)
17 different elements. Their advantage lies in the fact that they are capable
of producing a much more intense spectrum and thus are useful for those
elements whose hollow cathode lamps can produce only a weak spec
trum.
752c Light Chopper
Another item relating to instrument design that is different from the

molecular absorption spectrophotometers is the fact that there are two
sources of light that enter the monochromator and strike the detector.
How can the detector measure only the intensity of the light that does
not get absorbed and not measure the light emitted by the flame, since
both sources of light are of the same wavelength? Notice in Figure 7.8
that there is a light chopper placed between the lamp and the flame. The
light is "chopped" with a rotating half-mirror, a component similar to the
beam splitting device discussed for double-beam instruments in Chapter
6 . The detectors thus sees alternating light intensities. At one moment,
only the light emitted by the flame is read, since the light from the lamp
is cutoff, while at the next moment, the light from both the flame emission
and the transmission of the lamp's light is measured, since the lamp's
light is allowed to pass. The electronics of the detector is such that the
emission signal is subtracted form the total signal and this difference then
is what is measured. Absorbance is usually displayed on the readout.
7.5.2d Double-Beam AA
There is also a difference in the design of double-beam instruments.

In flame AA, the second beam does not pass through a second sample
container as it does for the molecular instruments, since a second flame
would be required and it would need to be identical to the sample flame.
Rather, the second beam simply bypasses the flame and is relayed to the
detector directly. This design eliminates variations due to fluctuations in
source intensity (the major objective of a double-beam), but does not
eliminate effects due to the flame or other components in the sample
(blank components). These must still be adjusted for by reading the blank
at a separate time. Also, the flame's emission continues to be accounted
for via the chopper as with the single-beam instrument (see Figure 7.11).
7.5.2e Monochromator, Detector, and Readout
Descriptions of the monochromator and detector/readout are similar

to that presented for molecular instruments in Chapter 6 . One important
difference is the placement of the monochromator. It is situated between
the light source and the sample in UV/vis molecular instruments, but
between the sample and the detector in atomic instruments, an arrangement
similar to IR molecular instruments. In AA instruments, the light source
(the hollow cathode lamp) does not require a monochromator in order
light chopper
FIGURE 7.11 A representation of a double-beam AA instrument.
to obtain the desired wavelength, as indicated previously. However, once

the light has passed through the flame, which, as with flame photometry,
is located in an open, exposed portion of the instrument, it is desirable
to screen out light from the flame and light from the room, as well as
extraneous lines from the source before the light strikes the detector. Thus
the monochromator is placed just before the detector. Another difference
between the AA monochromator and the monochromator in molecular
instruments is that there is a manually adjustable exit slit opening on the
AA instrument. One can thus widen or narrow the slit manually to
optimize the optics for the metal being analyzed.
The readout is in the form of a meter display on the instrument, a
recorder, or through data acquisition with a computer. In a typical
experiment, the readout consists of a series of absorbance readings with
blank readings in between, since the blank is ordinarily read before each
sample or standard. A strip-chart recording, for example, of the absor
bance values for a series of five standard solutions would appear as in
Figure 7.12.
7.5.3 Application
Quantitative analysis in atomic absorption spectroscopy utilizes Beer's

Law. The standard curve is a Beer's Law plot, a plot of absorbance vs
concentration. The usual procedure, as with other quantitative instrumental
methods, is to prepare a series of standard solutions over a concentration
range suitable for the samples being analyzed, i.e., such that the expected
sample concentrations are within the range established by the standards.
The standards and the samples are then aspirated into the flame, and the
FIGURE 7.12 Strip-chart recording of the absorbance values of a series of standard

solutions measured by an A A instrument. (From Kenkel, J., Analytical
With permission.)
absorbances read from the instrument. The Beer's Law plot will reveal
the useful linear range and the concentrations of the sample solutions.
In addition, information on useful linear ranges is often available for
individual elements and instrument conditions from manuals of analyti
cal methods available from instrument manufacturers, such as the Perkin-
Elmer Corporation, Norwalk, CT.
7.5.3a Interferences
Interferences can be a problem in the application of AA. Interferences

can be caused by chemical sources (chemical components present in the
sample matrix) or instrumental sources (so-called "spectral" interferences
arising from an instrument condition that turns out to be less than optimal
for a particular sample). Chemical interferences usually result from
incomplete atomization caused by an unusually strong ionic bond. An
example is in the analysis of a sample for calcium. The presence of sulfate
or phosphate in the sample matrix along with the calcium suppresses the
reading for calcium because of incomplete atomization due to the strong
ionic bond between calcium and the sulfate and phosphate. This results
in a low reading for the calcium in the sample in which this interference
exists. The usual solution to this problem is to add a substance to the
sample which would chemically free the element being analyzed, calcium
in our example, from the interference. With our calcium example, the
substance that accomplishes this is lanthanum. Lanthanum sulfates and
phosphates are more stable than the corresponding calcium salts, and
thus the calcium is free to atomize when lanthanum is present.
In addition to the above method for removing a chemical interference,
another possible solution may be to exactly match the matrix of the
standard and blank to that of the sample, and thus the interference would
be present in all solutions tested (at the same concentration) and this
would negate the problem, although sensitivity may be decreased due
to a smaller concentration of metal ions that get atomized. The method
of standard additions described in Chapter 5 is a way in which this may
be accomplished.
An example of a spectral interference is when the spectral line of the
element being determined nearly overlaps the line of another element in
the sample, such that some of the light from the hollow cathode lamp
will be absorbed by this interfering element, creating an absorbance
reading that is high. We call this an instrumental interference because the
slit opening setting suggested by the instrument manufacturer is too wide
to totally isolate the desired wavelength emerging from the monochro
mator. The solution is to use a narrower slit width or to zero in on a
different line, a so-called "secondary" line for the analyte element rather
than the "primary" line. Recommended secondary lines are also found
in instrument manufacturer's methods manuals and other literature
sources.
7.5.3b Safety and Maintenance
There are a number of important safety considerations regarding the

use of AA equipment. These center around the use of highly flammable
acetylene, as well as the use of a large flame, and the possible contamination
of laboratory air by combustion products. The acetylene is stored in a
compressed gas cylinder as in a welding lab (although the gas is of a
special purity for use in AA instruments). All precautions relating to

compressed gas cylinders must be enforced; the cylinders must be se
cured to an immovable object, such as a wall, and they must have
approved pressure regulators in place, etc. Tubing and connectors must
be free of gas leaks. There must be an independently vented fume hood
in place over the flame to take care of toxic combustion products. Volatile
flammable organic solvents and their vapors, such as ether and methyl
ene chloride, must not be present in the lab when the flame is lit.
Precautions should be taken to avoid flashbacks. Flashbacks result
from improperly mixed fuel and air, such as when the flow regulators
on the instrument are improperly set or when air is drawn back through
the drain line of the premix burner (see Section 7.2). Manuals supplied
with the instruments when they are purchased give more detailed
information on the subject of safety.
Finally, periodic cleaning of the burner head and nebulizer is needed
to ensure minimal noise level due to impurities in the flame. Scraping
the slot in the burner head with a sharp knife or razor blade to remove
carbon deposits and/or removing the burner head for the purpose of
cleaning in an ultrasonic bath are two commonplace maintenance chores.
The nebulizer should be dismantled, inspected, and cleaned periodically
to remove impurities that may be collected there.
7.5.3c Sensitivities, Detection Limits, and Analytical Uses
Sensitivity and detection limits have specific definitions in AA. Sen

sitivity is defined as the concentration of an element which will
produce an absorption of 1%. It is the smallest concentration that can be
determined with a reasonable degree of accuracy. Detection limit is the
concentration which gives a readout level that is double the noise level
inherit in the experiment. It is a qualitative parameter in the sense that
it is the minimum concentration that can be detected, but not accurately
determined, like a "blip" on a noisy baseline. It would tell the analyst
that the element is present, but not necessarily at an accurately determinable
concentration level. Some sensitivities and detection limits for selected
elements are given in Table 7.3.
The analytical uses of flame AA are very numerous. As indicated in
Table 7.3 Some Sensitivities and Detection Limits of Selected Elements in

Flame AA at the Primary Wavelength Using an Air/Acetylene
Flame and Both a Flow Spoiler and a Glass Impact Bead in the
Premix Chamber
Element Sensitivity (ppm) Detection Limit
Silver 0.03 0.0009

Arsenic 0.51 0.14
Calcium 0.08 0.001
Cadmium 0.02 0.0005
Chromium 0.04 0.002
Copper 0.03 0.001
Iron 0.04 0.003
Magnesium 0.003 0.0001
Manganese 0.03 0.001
Mercury 2.2 0.17
Nickel 0.04 0.004
Lead 0.19 0.01
Zinc 0.011 0.0008
Courtesy of Perkin-Elmer Corporation, Norwalk, CT.
Section 7.1, the vast majority of applications of atomic techniques is to

solutions of metal ions. Thus, whenever metals are to be determined in
a sample, atomic techniques are likely to be the chosen. Flame AA is the
most widely used of these at the present time, probably because it has
been around a long time and the instruments are less expensive than
those which perhaps offer important advantages (see later sections).
Table 7.4 lists some of the more common analytical uses of flame AA as
well as other AA techniques.
7.6 GRAPHITE FURNACE AND OTHER ATOMIZERS
There are some limitations to the use of flames as atomizers compared

to other atomizers that have been invented. First, they require relatively
large sample volumes (at least several milliliters). Second, they produce
fewer atoms in the path of the light, making them less sensitive. Third,
they are not applicable to samples in certain matrices, due to undesirable
matrix effects. This section discusses the alternative methods and espe
cially emphasizes the graphite furnace.
Table 7.4 Examples of Analytical Uses for Atomic Absorption
Field of Endeavor Examples of Samples and Analytes
Agriculture Calcium, copper, iron, magnesium, manganese,

potassium, sodium and zinc in soils, plant
tissue, feeds and fertilizers
Biochemistry Sodium, potassium, iron, lithium, copper, zinc,
gold, lead, calcium, and magnesium in
biological fluids, tissue, fingernails and hair
Environment Calcium, copper, lithium, magnesium,
manganese, potassium, sodium, strontium, and
zinc in seawater and natural water, and other
metals in airborne particulates
Food Various metals in food and food additives, meat,
seafood, cooking oils, and beverages
Heavy industry Various metals in cement, coal ash, glass,
ceramics, paint and paint additives, etc.
Metallurgy Various minor elements in alloys, such as steels,
brasses, alloys of aluminum, magnesium, lead,
tin, copper, titanium, nickel, and iron, as well as
plating solutions
Petroleum Various metals in lubricating oils and additives,
various wear metals in used lubricating oils,
lead in gasoline, and metals in other fuels, oils
and additives
Pharmaceuticals Various metals in pharmaceutical preparations
Compiled from "Analytical Methods for Atomic Absorption Spectrophotometry"

Perkin-Elmer Corporation, Norwalk, CT.
7.6.1 Graphite Furnace Atomizer
The graphite furnace method of atomization is useful and well estab

lished. The conversion of ions to atoms occurs in a small hollow graphite
cylinder (tube) which replaces the burner head in flame AA. (The flame
and graphite furnace are interchangeable in most instruments.) This
cylinder is positioned horizontally such that the light beam passes di
rectly through the center of it. There is a small hole in the center of the
top side of the cylinder for introducing the sample. The sample is not
introduced through any form of aspiration, but rather by means of an
injection through this small hole, often onto a platform inside the tube.
Thus, atoms are in the path of the light for a limited time defined by the
sample volume. Very small and carefully measured sample volumes, 5 -
Electrical Sample
contact for Inert gas in
introduction
heating \ (external tube) Inert gas
| out
j Il , Graphite tube
/
Quartz window
\ ^ U g h t beam
Quartz window
Inert gas in Platform Inert gas
(internal tube) out
FIGURE 7.13 The graphite furnace atomizer.
50 |iL, are used. This small size can be a distinct advantage when only
small volumes are available. The tube is encased in a larger tube in order
to facilitate protection against air oxidation of the graphite at the high
temperature needed for atomization. An inert gas (argon or nitrogen) is
fed both into the larger tube (but outside the inner tube) and into the inner
tube. To further protect the tube and to prevent the sample from perme
ating through, a coating of pyrolytic graphite is often applied. Contact
rings and quartz windows are attached at both ends of the tubes so as
to seal the entire system (see Figure 7.13).
The tube itself acts as a heating element. The contact rings referred to
above serve to make electrical contact to a power source. The power is
controlled according to a three-step temperature program such that the
sample is first dried, then ashed, and finally atomized at a temperature
of about 2500 K. The atomic vapor fills the tube at this temperature, and
light from the hollow cathode lamp is absorbed with the absorbance
measured as in flame AA. The drying and ashing steps are required to
destroy organic matter which produces smoke inside and would other
wise scatter the light from the source unless it is first flushed from the
system by the flowing inert gas. The programming of the temperature
then produces the time required for this to take place prior to the atomi
zation.
Since there is a limited amount of sample present and since the atomic
vapor is also eventually swept from the tube by the flowing inert gas,
the absorbance signal is transient. This means that the recording of the
absorbance is usually done as a function of time, such as on a recorder
or by computer data acquisition, and the transient signal then appears

as a sharp peak at a particular time.
The advantages of the graphite furnace are (1) the fact that only small
sample volumes are needed, as previously stated, and (2 ) the sensitivity
is substantially increased. Atomization in this furnace is nearly 100%
efficient which is a tremendous improvement over the flame (0 .1 %).
Detection limits are improved by as much as 1000 times. The major
disadvantages are ( 1 ) there often can be matrix and background absorp
tion effects, and (2) reproducibility is not as good as with flames. Standard
additions (Chapter 5) can be a solution to the matrix effect. Continuum
source correction and Zeeman Effect correction (utilizing a magnetic
field) have been used effectively to reduce background effects. Detailed
discussions of these are beyond the scope of this book.
7.6.2 Vapor Generation Methods
An alternative room temperature atomization method has been devised

for mercury known as the cold vapor mercury technique. Also, an
alternative technique for the difficult elements arsenic, bismuth,
germanium, lead, antimony, selenium, tin, and tellurium called the hydride
generation technique has been devised. These are collectively referred to
as vapor generation techniques. They utilize a chemical reduction (of the
metal ion to the metal atom or to the metal hydride) process rather than
a direct atomization in a flame or furnace.
In the cold vapor technique for mercury, mercury atoms are generated
via chemical reduction with a strong reducing agent, such stannous
chloride. A stream of nitrogen or air sweeps the mercury atoms into a
sample absorption cell that is positioned in the path of the light. A
transient absorption signal, similar to the graphite furnace signal, is
observed and measured.
In the hydride generation technique, sodium borohydride is the usual
reducing agent. Again the reaction product is swept into the path of the
light, but, unlike the mercury technique, the analyte is not yet in the free
atomic state at this stage, but rather in the molecular hydride state. Once
in the sample absorption cell, this vapor is heated with either a cool air /
acetylene flame or with an electrical heating unit to create the atoms.
Advantages of vapor generation for the above listed elements can
include (1 ) a decrease in the effects of sample matrices, (2 ) improved
precision, and (3) higher sensitivity.
7.6.3 The Delves Cup
Improvements for some elements can be achieved by simply holding

the sample in a cup in the flame (no aspiration). This cup, called the
"Delves Cup," is composed of nickel. The atomic vapor generated is
channeled into a sample absorption cell and the absorbance measured.
Such a technique can utilize a sample as small as 100 |iL (an advantage),
but often suffers from poorer precision.
7.7 INDUCTIVELY COUPLED PLASMA
A technique that is gaining in popularity due to some important

advantages over the techniques mentioned so far in this chapter is the
Inductively Coupled Plasma technique better known as ICP. ICP is strictly
an atomic emission technique most closely related to the flame photometry
technique described earlier. The ICP emission source is not a flame,
however, but a plasma. A plasma source is a flame-like system of ionized,
very hot flowing argon gas that is directed through a quartz tube wrapped
with a copper wire (coil). Radio frequency energy is applied to the coil
creating an intense magnetic field inside the tube. The ionization of the
gas is initiated by a 'Tesla" coil, or spark, upstream from the quartz tube,
causing the argon to become conductive, creating argon ions and electrons
in the flow path. As the partially ionized argon flows through the quartz
tube and the RF generated magnetic field, more ionization occurs and
the gas becomes extremely hot — 9000-10,000 K. What emerges from the
quartz tube then is a very hot spray of ionized argon that is the ICP source,
giving the appearance of a flame.
The sample and standards, in the form of solutions (as with the other
atomic techniques described in this chapter), are introduced into the
flowing argon by aspiration just prior to the initial ionization. The entire
system is pictured in Figure 7.14.
As stated previously in this chapter, a hotter source increases both
atomization efficiency and excitation efficiency. Thus the hotter an
excitation source, the more intense one would expect the emission to be,
and the smaller the concentrations of metal ions that can be detected and
accurately measured. ICP is therefore much more sensitive than all other
atomic techniques. In addition, the concentration range over which the
FIGURE 7.14 The ICP system described in the text.
emission intensity is linear is broader and simultaneous "multielement"

analysis of a sample is possible. The instruments are more costly than
AA and other systems.
7.8 OTHER ATOMIC EMISSION TECHNIQUES
7.8.1 Arc or Spark Emission Spectrography
A technique which utilizes a solid sample for light emission is Arc of

Spark Emission Spectrography. In this technique, a high voltage is used
to excite a solid sample held in an electrode cup in such a way that when
a spark or arc jumps from this electrode to another electrode in this
arrangement, atomization, excitation, and emission occur and the emitted
light again is measured. The usual configuration is such that the emitted
light is dispersed and detected with the use of photographic film, hence
the name "spectrography." The "picture" that results is that of a combined
Spark or Arc
source
FIGURE 7.15 A diagram of a spark emission instrument.
line spectrum of all the elements in the sample. Identification (qualitative

analysis) is then possible by comparing the locations of the lines on the
film to location of lines on a standard film. Figure 7.15 shows the
instrumental arrangement.
7.8.2 Atomic Fluorescence
Finally, we briefly describe a technique based on both absorption and

emission — atomic fluorescence. When atoms that have been elevated
to higher energy levels return to the ground state, the pathway could take
them to some intermediate electronic states prior to the final drop. Such
a series of drops back to the ground state, if accompanied by light
emission, is a form of fluorescence, in this case, atomic fluorescence. (See
Chapter 6 for a discussion of molecular fluorescence.) As with molecular
fluorescence, the intensity of this emitted light is measured at right angles
to the incident light and related to concentration.
7.9 SUMMARY OF ATOMIC TECHNIQUES
Table 7.5 summarizes the description and application of the techniques

discussed in this chapter.
Table 7.5 A Summary of the Techniques Described in This Chapter
Technique Principle Comments
Flame photometry Intensity of light emitted A well-established

by atoms in a flame is technique that is used
measured and related to for only a limited
concentration. number of analytes and
samples due to the
applicability and
sensitivity of other
techniques.
Flame AA Light absorbed by atoms in a A well-established
flame is measured and rela technique that remains
ted to concentration. very popular for a large
number of samples and
analytes. Instruments are
inexpensive.
Graphite furnace Transient light absorbance A newer technique that
signal from sample atomized has become popular due
in a graphite furnace (no to its greater sensitivity
flame) is measured and and applicability to
related to concentration. smaller sample size.
Suffers from lack of precision.
Vapor generation Light absorbed by chemically Excellent technique for
generated atomic vapor is a limited number of
measured and related to analytes that are diffi
concentration. cult to measure otherwise.
ICP Light emitted by atoms in an A newer technique that
inductively coupled plasma has become popular due
is measured and related to to advantages in
concentration. sensitivity, linear range, and
multielement analysis.
Instruments are costly.
Spark or Arc Light emitted by powdered An excellent technique
emission solid sample caught in an for qualitative analysis
arc or spark between a pair of metals in solids.
of electrodes is measured for
qualitative analysis.
Atomic Light emitted by atoms in a Not a popular technique,
fluorescence flame excited by the but it can offer
absorption of light is sensitivity advantages.
measured and related to
concentration.
C hapter 8
A nalytical S eparations
8.1 INTRODUCTION
Modem day chemical analysis can involve very complicated material

samples — complicated in the sense that there can be many, many
substances present in the sample creating a myriad of problems with
interferences when the lab worker attempts the analysis. These interferences
can manifest themselves in a number of ways. The kind of interference
that is most familiar is probably one in which substances other than that
the analyte generate an instrumental readout similar to the analyte, such
that the interference adds to the readout of the analyte, creating an error.
However, an interference can also suppress the readout for the analyte
(e.g., by reacting with the analyte). An interference present in a chemical
to be used as a standard (such as a primary standard) would cause an
error, unless its presence and concentration were known. Analytical
chemists must deal with these problems, and chemical procedures designed
to effect separations or purification are now commonplace.
This chapter, and also Chapters 9 and 10, describe modem analytical
separation science. First, purification procedures known as recrystallization
and distillation will be described. Then, the separation techniques of
195
extraction and chromatography are discussed. This is followed by, in

Chapters 9 and 10, instrumental chromatography techniques which can
resolve very complicated samples and quantitate usually in one easy step.
8.2 RECRYSTALLIZATION
Recrystallization is a purification technique for a solid, usually organic.

The separation is based on the solid's solubility in a liquid solvent. A
solvent is chosen such the solid is sparingly soluble at room temperature,
but whose solubility increases considerably at higher temperatures. Both
soluble and insoluble impurities are considered to be present, and the
procedure removes both if their concentrations are not too large.
The key to the procedure is to use a minimum amount of solvent, such
that the solid will just dissolve at the elevated temperature (usually the
boiling point, if the solid is stable at that temperature). While maintaining
this elevated temperature, any impurity that has not dissolved can be
filtered out. The "insoluble" impurities are thus removed.
Soluble impurities, however, are still present in the filtrate. This is
where the minimum amount of solvent comes into play. The procedure
calls for the temperature to be lowered back to the original value. The
soluble impurities will stay dissolved if their solubility has not been
exceeded (if they are present in a small amount). However, the solid being
purified will have its solubility exceeded. Since the minimum amount of
solvent was used to just dissolve the solid at the elevated temperature,
it will thus precipitate from the solution. The solid, presumably purified,
can then be filtered and dried. It may be necessary to perform the recrys
tallization several times in order to get the desired purity.
8.3 DISTILLATION
Distillation is a method of purification of liquids contaminated with

either dissolved solids or miscible liquids. The method consists of boiling
and evaporating the mixture followed by recondensation of the vapors
in a "condenser," which is a tube cooled by isolated, cold tap water. The
theory is that the vapors (and thus recondensed liquids) will be purer
Analytical Separations 197
than the original liquid. The separation is based on the fact that the
contaminants have different boiling points and vapor pressures than the
liquid to be purified. Thus, when the liquid is boiled and evaporated, the
vapors (and recondensed liquids) created have a composition different
from the original liquid. The substances with lower boiling points and
higher vapor pressures are therefore separated from substances that have
high boiling points and low vapor pressures.
Distillation of water to remove hardness minerals is an example and
probably the most common application in an analytical laboratory. Of all
the applications of distillation, it is one of the easiest to perform. While
water is known to have a relatively high boiling point and low vapor
pressure, the dissolved minerals are ionic solids that generally have
extremely high boiling points (indeed, extremely high melting points)
and extremely low vapor pressures. Thus, a simple distilling apparatus
and a single distillation, or, at most, two (doubly distilled) or three (triply
distilled) distillations, will produce very pure water.’
Organic liquids that are contaminated with other organic liquids usually
constitute a much more difficult situation. Such liquids probably have
such similar boiling points and vapor pressures that a distillation of a
mixture of two or more would result in all being present in the distillate
(the condensed vapors) — an unsuccessful purification. However, the
liquid that has the highest vapor pressure and/or lowest boiling point,
while not being completely purified, would be present in the distillate
at a higher concentration level than the other components. It follows that
if the distillate were then to be redistilled, perhaps over and over again,
further enrichment of this component would take place such that an
acceptable purity would eventually be obtained. However, the time
involved in such a procedure would be prohibitive. A procedure known
as "fractional distillation," solves the problem.
Fractional distillation involves repeated evaporation/condensation steps
before the distillate is actually collected. These repeated steps occur in a
"fractionating column" (tube) above the original heated container — a
column that contains a high surface area inert material for condensing
the vapors. As the vapors condense on this material, the material itself
heats up and the condensate reevaporates. The reevaporated liquid then
moves further up the column, contacts more cold inert material, and the
Water is often "deionized" using an ion exchange (Section 8.6) cartridge to remove hardness
minerals. Such deionization is often done in conjuction with distillation, such that the water
is both deionized and distilled prior to use. Also, if the water is contaminated with organics,
or other low boiling substances, a charcoal filter cartridge is often used as well.
process occurs again and again and again as the liquid makes its way
up the column. If a fractionating column were used that is long enough
and contains a sufficient quantity of the high surface area material, any
purification based on differences in boiling point and vapor pressure can
be affected. A schematic diagram of a distillation apparatus fitted with
a fractionating column is shown in Figure 8.1. The high surface area
packing material in a fractionating column typically consists of glass
beads, glass helices, or glass wool.
Each time a single evaporation /condensation step occurs in a
fractionating column, the condensate has passed through what has been
called a "theoretical plate." A theoretical plate is thus that segment of a
fractionating column in which one evaporation/condensation step occurs.
The name is derived from the concepts in which the condensate is actually
captured on small "plates" inside the fractionating column from which
it is again boiled and evaporated. A fractionating column used for a given
liquid mixture is then identified as having a certain number of theoretical
plates, and given liquid mixtures are known to require a certain number
of theoretical plates in order to achieve a given purity. The "height
equivalent to a theoretical plate," or HETP, is the length of fractionating
column corresponding to one theoretical plate. If the number of theoreti
cal plates required is known, then the analyst can select a height of
column that would contain the proper number of plates according to
manufacturer's specifications or according to his own measurements of
a homemade column. Height selection is not entirely experimental,
however. The use of liquid-vapor composition diagrams to predict the
theoretical plates required can help. These diagrams are based on boiling
point and vapor pressure differences in a pair of liquids. Further discus
sion of the use of these diagrams is beyond the scope of this book.
8.4 LIQUID-LIQUID EXTRACTION
8.4.1 Introduction
One popular method of separating an analyte species from a compli

cated liquid sample is the technique known as "liquid-liquid extraction"
or "solvent extraction." In this method, the sample containing the analyte
is a liquid solution, typically a water solution, that also contains other
FIGURE 8.1 A schematic diagram of a distillation assembly complete with a frac

tionating column, condenser, and collection vessel.
solutes. The need for the separation usually arises from the fact that the
other solutes, or perhaps the original solvent, interfere in some way with
the analysis technique chosen. An example is a water sample that is being
analyzed for a pesticide residue. The water may not be a desirable solvent,
and there may be other solutes that may interfere. It is a "selective
dissolution" method; a method in which the analyte is removed from the
original solvent and subsequently dissolved in a different solvent,
(extracted) while most of the remainder of the sample remains unextracted,
i.e., remains behind in the original solution.
The technique obviously involves two liquid phases— one the original
solution and the other the extracting solvent. The important criteria for
a successful separation of the analyte are (1 ) that these two liquids be
immiscible and (2 ) that the analyte be more soluble in the extracting
solvent than the original solvent.
8.4.2 The Separatory Funnel
The extraction takes place in a specialized piece of glassware known

as a separatory funnel. The separatory funnel is manufactured especially
for solvent extraction. It has a "teardrop" shape with a stopper at the top
and a stopcock at the bottom (Figure 8.2). The sample and solvent are
placed together in the funnel, the funnel is tightly stoppered and, while
holding the stopper in with the index finger, shaken vigorously for a
moment. Following this, the funnel may need to be vented, since one of
the liquids is likely to be a volatile organic solvent, such as methylene
chloride. Venting is accomplished by opening the stopcock when in
verted. This shaking/venting step is then repeated several times such that
the two liquids have plenty of opportunity for the intimate contact re
quired for the analyte to pass into the extracting solvent to the maximum
possible extent; See Figure 8.3 for shaking/venting illustrations. Follow
ing this procedure, the funnel is positioned in a padded ring in a ring
stand (Figure 8.2) and left undisturbed for a moment to allow the two
immiscible layers to once again separate. The purpose of the specific
design of the separatory funnel is mostly to provide for easy separation
of the two immiscible liquid layers after the extraction takes place. All
one needs to do is remove the stopper, open the stopcock, allow the
bottom layer to drain, then close the stopcock when the interface between
the two layers disappears from sight in the stopcock. The denser of the
two liquids is the bottom layer and will be drained through the stopcock
first. The entire process may need to be repeated several times, since the
extraction is likely not to be quantitative. This means that another quan
tity of fresh extracting solvent may need to be introduced into the separatory
funnel with the sample and the shaking procedure repeated. Even so, the
experiment may never be completely quantitative. See the next section
for the theory of extraction and a more in-depth discussion of this prob
lem.
8.4.3 Theory
The process of a solute dissolved in one solvent being "pulled out,"

or "extracted," into a new solvent actually involves an equilibrium process.
At the time of initial contact, the solute will move from the original solvent
to the extracting solvent at a particular rate, but, after a time, it will begin
to move back to the original solvent at a particular rate. When the two
rates are equal, we have equilibrium. We can thus write the following:
FIGURE 8.2 A separatory funnel containing two immiscible liquids.
FIGURE 8.3 Illustrations of the positioning of the hands and the procedure for the
shaking and venting mentioned in the text.
in which "A" refers to "analyte," and "orig" and "ext" refer to "original
solvent" and "extracting solvent," respectively. If the analyte is more
soluble in the extracting solvent than in the original solvent, then, at
equilibrium, a greater percentage will be found in the extracting solvent
and less in the original solvent. If the analyte is more soluble in the
original solvent, then the greater percentage of analyte will be found in

the original solvent. Thus, the amount that gets extracted depends on the
relative distribution between the two layers, which, in turn, depends on
the solubilities in the two layers. A distribution coefficient analogous to
an equilibrium constant (also called the "partition coefficient") can be
defined as follows:
K - [A ]-
K- — — (8.2)
[A ]ori
orig
Often, the value of K is approximately equal to the ratio of the solubilities

of "A" in the two solvents. If the value of K is very large, the transfer
of solute to the extracting solvent can be considered to be quantitative.
A value around 1.0 would indicate equal distribution, and a small value
would indicate very little transfer.
Uses of the distribution coefficient include (1) the calculation of the
amount of a solute that is extracted in a single extraction step, (2 ) the
determination of the weight of the solute in the original solute (important
if you are quantitating the solute in this solvent), (3) the calculation of
the optimum volume of both the extracting solvent and the original
solution to be used, (4) the number of extractions needed to obtain
particular quantity or concentration in the extracting solvent, and (5) the
percent extracted. The following expansion of Equation 8.2 is useful for
these.
w.
K = w„.
(8.3)
in which Wext is the weight of the solute extracted into the extracting
solvent, Vext is the volume of the extracting solvent used, Worig is the
weight of the solute in the original solvent, and Vorig is the volume of the
original solvent used.
More complicated chemical systems may require a more universally
applicable quantity called the "distribution ratio" to describe the system.
These involve situations in which the analyte species may be found in
different chemical states and different equilibrium species, some of which
may be extracted while others are not extracted. An example is an equi-
equilibrium system involving a weak acid. In such a system, there may

be one (or several) protonated species and one unprotonated species. The
"distribution ratio," D, then takes into account all analyte species present.
C , (ext)
d =^ h I) (84)
in which CA (ext) and CA (orig) represent the total concentration of all

analyte species present in the two phases regardless of chemical state.
Further treatment of this situation is beyond the scope of this text.
8.4.4 Countercurrent Distribution
Just one extraction performed on a solution of a complicated sample

will likely not result in total or at least sufficient separation of the analyte
from other interfering solutes. Not only will these other species also be
extracted to a certain degree with the analyte, but some of the analyte
species will likely be left behind in the original solvent as well. Thus, the
analyst will need to perform additional extractions on both the extracting
solvent, to remove the other solutes that were extracted with the analyte,
and the original solution, to remove additional analyte that was not
extracted the first time. One can see that dozens of such extractions may
be required to achieve the desired separation. Eventually, there would
be a separation, however. The process is called countercurrent distribu
tion.
In countercurrent distribution, the extracting solvent, after first being
in contact with the original solution, is moved to another separatory
funnel in which there is fresh original solvent, while fresh extracting
solvent is brought into the original funnel and the extractions performed.
Then, the extracting solvent from the second funnel is moved to a third
funnel containing fresh original solvent, the extracting solvent from the
first is moved to the second funnel, and fresh extracting solvent is intro
duced into the first funnel. The process continues in this manner until
the desired separation occurs. The concept is illustrated in Figure 8.4. The
top half in each segment (a, b, etc.) represents extracting solvent, while
the lower half represents original solvent. A mixture of "x" and "." is
being separated. In each segment, fresh extracting solvent is introduced
a xTW
25RS?
• *# x #
• * •
X »»
X • *•
X •
V * • X
X • • •
X •
*• X
* • • •
* X • X
X X •
X • • .
X •
4 X X X •
X X • X •
X •x • • •
X
X •
X X X V
X • • •
X X •
• • • •
X xxX •
X X X •
• •
X • •
X
x x x X •
X X X * •
FIGURE 8.4 A diagram depicting countercurrent distribution as discussed in the

text. (From Kenkel, J., Analytical Chemistry for Technicians, Lewis
on the upper left, while fresh original solvent is introduced on the low
ered right. This illustration shows a complete separation in segments "g"
and "h." Some chromatography methods are based on this concept and
will be discussed in Section 8.5.
8.4.5 Concentrators
Following an extraction procedure, it is often necessary to evaporate

a quantity of the extracting solvent in order to increase the concentration
of the analyte prior to gas chromatographic analysis (Chapter 9), for
example. Two so-called "concentrators" in common use are the rotary
evaporator and the Kudema-Danish evaporative concentrator. With the
rotary evaporator, the analyte/solvent mixture is heated in a round
bottom flask which is rotated while lying nearly on its side in a steam
bath. Solvent vapors are condensed and collected in a separate flask
under a condenser off to the side, while the analyte remains in the rotating
flask. The result is a solution of the analyte that has a higher concentration.
The procedure takes place while the system is under reduced pressure.
Bumping and superheating are avoided because of the rotation.
The Kudema-Danish apparatus is a distillation/reflux assembly that
utilizes a special fractionating column called the Snyder column. The
lighter lower-boiling solvent(s) escapes the Snyder column to a solvent
recovery condenser, while the analyte refluxes back to the original
container, the Kudema-Danish flask, and graduated concentrator tube.
The end result is a 5-10 mL fraction of analyte concentrated in this tube.
Both of these units are used when the analyte concentration is too low
to be detected otherwise.
8.5 LIQUID-SOLID EXTRACTION
There are instances in which the analyte will need to be extracted from
a solid material sample rather than a liquid. As in the above discussion
for liquid samples, such an experiment is performed either because it is
not possible or necessary to dissolve the entire sample or because it is
undesirable to do so because of interferences that may also be present.
In these cases, the weighed solid sample, preferably finely divided, is
brought into contact with the extracting liquid in an appropriate container
(not a separatory funnel) and usually shaken or stirred for a period of
time such that the analyte species is removed from the sample and
dissolved in the liquid. The time required for this shaking is determined
by the rate of the dissolving. A separatory funnel is not used since two
liquid phases are not present, but rather a liquid and a solid phase. A
simple beaker, flask, or test tube usually suffices.
Following the extraction, the undissolved solid material is then filtered
out and the filtrate analyzed. Examples of this would be soil samples to
be analyzed for metals, such as potassium or iron, and cellophane or
insulation samples to be analyzed for formaldehyde residue. The extracting
liquid may or may not be aqueous. Soil samples being analyzed for
metals, for example, utilize aqueous solutions of appropriate inorganic
compounds, sometimes acids, while soil samples, or the cellophane or
insulation samples referred to above, that are being analyzed for organic
compounds utilize organic solvents for the extraction. As with the liquid-
liquid examples, the extract is then analyzed by whatever analytical
technique is appropriate — atomic absorption for metals and
spectrophotometry or gas or liquid chromatography for organics.
Concentration methods, as with the Kudema-Danish apparatus described
in Section 8.4, may also be required, especially for organics.
It may be desirable to try to keep the sample exposed to fresh extracting
solvent as much as possible during the extraction in order to maximize
the transfer to the liquid phase. This may be accomplished by pouring
off the filtrate and reintroducing fresh solvent periodically during the
extraction and then combining the solvent extracts at the end. There is
a special technique and apparatus, however, that has been developed,
called the Soxhlet extraction apparatus, which accomplishes this auto
matically. The Soxhlet apparatus is shown in Figure 8.5. The extracting
solvent is placed in the flask at the bottom, while the weighed solid
sample is place in the solvent-permeable thimble in the compartment
directly above the flask. A condenser is situated directly above the thimble.
The thimble compartment is a sort of cup that fills with solvent when
the solvent in the flask is boiled, evaporated, and condensed on the
condenser. The sample is thus exposed to freshly distilled solvent as the
cup fills. When the cup is full, the glass tube next to the cup is also full,
and when it (the tube) begins to overflow, the entire contents of the cup
is siphoned back to the lower chamber and the process repeated. The
advantages of such an apparatus are (1 ) fresh solvent is continuously in
contact with the sample (without having to introduce more solvent,
which would dilute the extract) and (2 ) the experiment takes place
unattended and can conveniently occur overnight if desired.
FIGURE 8.5 A drawing of a Soxhlet Extraction apparatus.
8 .6 CHROMATOGRAPHY
8.6.1 Introduction
A myriad of techniques used to separate complex samples comes

under the general heading of "chromatography." The nature of chroma
tography allows much more versatility, speed, and applicability than any
of the other techniques discussed in this chapter, particularly when the
modem instrumental techniques of gas chromatography (GC) and high
performance liquid chromatography (HPLC) are considered. These latter
techniques are covered in detail in Chapters 9 and 10. In this chapter, we
introduce the general concepts of chromatography and give a perspective

on its scope. Since there are many different classifications, this will include
an organizational scheme covering the different types and configurations
that exist.
Chromatography is the separation of the components of a mixture
based on the different degrees to which they interact with two separate
material phases. The nature of the two phases and the kind of interaction
can be varied, and this gives rise to the different "types" of chromatog
raphy which will be described in the next section. One of the two phases
is a moving phase (the "mobile" phase) while the other does not move
(the "stationary" phase). The mixture to be separated is usually intro
duced into the mobile phase which then is made to move or percolate
through the stationary phase either by gravity or some other force. The
components of the mixture are attracted to and slowed by the stationary
phase to varying degrees, and as a result, they move along with the
mobile phase at varying rates and are thus separated. Figure 8 . 6 illustrates
this concept.
The mobile phase can be either a gas or a liquid, while the stationary
phase can be either a liquid or solid. One classification scheme is based
on the nature of the two phases. All techniques which utilize a gas for
the mobile phase come under the heading of "gas chromatography"
(GC). All techniques that utilize a liquid mobile phase come under the
heading of "liquid chromatography" (LC). Additionally, we have gas-
liquid chromatography (GLC), gas-solid chromatography (GSC), liquid-
liquid chromatography (LLC), and liquid-solid chromatography (LSC) if
we wish to stipulate the nature of the stationary phase as well as the
mobile phase. It is more useful, however, to classify the techniques
according to the nature of the interaction of the mixture components with
the two phases. These classifications we refer to as "types" of chroma
tography.
8.6.2 "Types" of Chromatography
8.6.2a Partition
In Section 8.4, we stated that some chromatography methods are

based on the concept of countercurrent solvent extraction. You will recall
that this is the technique in which a large number of extractions are
performed, with fresh extracting solvent being brought into contact with
S3
38 § n
n
mobile phase — '^stationary phase
two-qomponent
mixture
-V IA /L H £5
v W k
/ W V O A /W l
A /lA A s <5 0 /V A h
M AAs
W v
A /W A
FIGURE 8.6 Mixture components separate as they move through the stationary
phase with the mobile phase. (From Kenkel, J., Analytical Chemistry
for Technicians, Lewis Publishers, Inc., Chelsea, MI, 1988. With per
mission.)
previously extracted samples, while fresh sample solvent is brought into

contact with solvent extract from previous extractions (see Figure 8.4 and
accompanying discussion). The extracting solvent can be thought of as
continuously "moving" across the sample solvent, while the latter re
mains stationary. The mixture components, which were initially found
in the first segment of sample solvent, then distribute back and forth
between the two phases as the extracting liquid moves and are found
individually separated at different points along the way according to their
individual solubilities in the two solvents (see Figure 8.7).
The extracting solvent in this scenario is the chromatographic mobile
phase, while the sample solvent is the stationary phase. Liquid-liquid
partition chromatography (LLC) is based on this idea. The mobile phase
is a liquid which moves through a liquid stationary phase as the mixture
components "partition" or distribute themselves between the two phases
and become separated. The separation mechanism is thus one of the
dissolving of the mixture components to different degrees in the two
phases according to their individual solubilities in each.
FIGURE 8.7 Illustration of the movement of an extracting solvent across a sample

solvent and the separation of two compounds initially found in the
sample solvent.
It may be difficult to imagine a liquid mobile phase used with a liquid

stationary phase. What experimental setup allows one liquid to move
through another liquid (immiscible in the first), and how can one expect
partitioning of the mixture components to occur? The stationary phase
actually consists of a thin liquid film either adsorbed or chemically bonded
to the surface of finely divided solid particles as shown in Figure 8 .8 .
Chemically bonded liquid stationary phases represent a fairly recent
development in this area, as opposed to adsorbed liquid phases. The latter
is considered to be a more "classical" chromatography technique. Bonded-
phase chromatography (BPC) has a distinct stability advantage. It is not
removed from the solid substrate either by reaction or by heat. BPC has
become the most popular of the two by far.
Since the separation depends on the relative solubilities of the
components in the two phases, the polarities of the components and that
of the stationary and mobile phases are important to consider. If the
stationary phase is somewhat polar, it will retain polar components more
than it will nonpolar components, and thus the nonpolar components will
move more quickly through the stationary phase than the polar
components. The reverse would be true if the stationary phase were
nonpolar. Of course, the polarity of a liquid mobile phase plays a role
too.
The mobile phase for partition chromatography can also be a gas
(GLC). In this case, however, the mixture components' solubility in the
thin liquid film
solid
substrate
particles
4 4 /i/m i ■W\AA,
/ W W VI m m /
AAAA/V A W l/
FIGURE 8.8 Partition chromatography with a thin liquid film adsorbed or chemi
cally bonded to the surface of finely divided solid particles. (From
Kenkel, J., Analytical Chemistry for Technicians, Lewis Publishers, Inc.,
Chelsea, MI, 1988. With permission.)
mobile phase is not an issue — rather their relative vapor pressures are
important. This idea will be expanded in Chapter 9.
In summary, partition chromatography is a type of chromatography
in which the stationary phase is a liquid adsorbed or chemically bonded
to the surface of a solid substrate, while the mobile phase is either a liquid
or gas. The mixture components dissolve in and out of the mobile and
stationary phases as the mobile phase moves through the stationary
phase and the separation occurs as a result. Examples of mobile and
stationary phases will be discussed in Chapters 9 and 10.
8.6.2b Adsorption
Another chromatography "type" is adsorption chromatography. As

the name implies, the separation mechanism is one of adsorption. The
stationary phase consists of finely divided solid particles packed inside
a tube, but with no stationary liquid substance present to function as the
stationary phase, as is the case with partition chromatography. Instead,
the solid itself is the stationary phase and the mixture components, rather
than dissolve in a liquid stationary phase, adsorb or "stick" to the surface
of the solid. Different mixture components adsorb to different degrees
Molecules
adsorbed
on surface
< & p
■YWWl >§OcSPoK b W c M AA,

''lAAAAA
'V W W D O O §5 fo°0° 1 M V
q o
FIGURE 8.9 A depiction of adsorption chromatography.
of strength, which also depends on the mobile phase, and thus again they
become separated as the mobile phase moves. The nature of the adsorp
tion involves the interaction of polar molecules or molecules with polar
groups, with a very polar solid stationary phase. Thus, hydrogen bonding
or similar molecule-molecule interactions are involved.
This "very polar solid stationary phase" is typically silica gel or alumina.
The polar mixture components can be organic acids, alcohols, etc. The
mobile phase can be either a liquid or a gas. This type of chromatography
is depicted in Figure 8.9.
8.6.2c Ion-Exchange
A third chromatography "type" is ion-exchange chromatography. As

the name implies, it is a method for separating mixtures of ions, both
inorganic and organic. Tire stationary phase consists of very small polymer
resin "beads" which have many ionic bonding sites on their surfaces.
These sites selectively exchange ions with certain mobile phase
compositions as the mobile phase moves. Ions that bond to the charged
site on the resin beads are thus separated from ions that do not bond.
Repeated changing of the mobile can create conditions which will further
<tfFe"ZZ=> I > H4 Cu*‘
w vw v
M " >
AAAA
FIGURE 8.10 A depiction of ion-exchange chromatography. (From Kenkel, J.,
selectively dislodge and exchange bound ions which then are also sepa
rated. This stationary phase material can be either an anion exchange
resin, which possesses positively charged sites to exchange negative ions,
or a cation exchange resin, which possesses negatively charge sites to
exchange positive ions. The mobile phase can only be a liquid. Further
discussion of this type can be found in Chapter 10. Figure 8.10 depicts
ion-exchange chromatography.
8.6.2d Size-Exclusion
Size-exclusion chromatography, also called "gel-permeation" (GPC) or

"gel-filtration" (GFC) chromatography, is a technique for separating dis
solved species on the basis of their size. The stationary phase consists of
porous polymer resin particles. The components to be separated can enter
the pores of these particles and be slowed from progressing through this
stationary phase as a result. Thus, the separation depends on the sizes
of the pores relative to the sizes of the molecules to be separated. Small
particles are slowed to a greater extent than larger particles, some of
which may not enter the pores at all, and thus the separation occurs. The
mobile phase for this type can also only be a liquid, and it too is discussed
further in Chapter 10. The separation mechanism is depicted in Figure
8.11.
8.6.3 Chromatography Configurations
Chromatography techniques are further classified according to "con

figuration" — how the stationary phase is "contained"; how the mobile
phase is configured with respect to the stationary phase in terms of
physical state (gas or liquid); positioning; and how and in what direction
the mobile phase travels in terms of gravity, capillary action, or other
forces.
Configurations can be broadly classified into two categories: the "planar"
methods and the "column" methods. The planar methods utilize a thin
sheet of stationary phase material, and the mobile phase moves across
this sheet, either upward ("ascending" chromatography), downward
("descending" chromatography) or horizontally ("radial" chromatogra
phy). Column methods utilize a cylindrical tube to contain the stationary
phase, and the mobile phase moves through this tube either by gravity,
with the use of a high pressure pump, or by gas pressure. Additionally,
with the exception of paper chromatography, those that utilize a liquid
for the mobile phase are capable of utilizing all of the "types" reviewed
above. Paper chromatography utilizing unmodified cellulose sheets is
strictly partition chromatography (see Section 8.6.3a). If the mobile phase
is a gas (gas chromatography), the "type" is limited to adsorption and
partition methods. Table 8.1 summarizes the different configurations. Let
us consider each individually.
Table 8.1 Chromatography Configurations and Their Applicable Options
Geometry Configuration Migration Direction Applicable Types
Planar Paper Ascending, Partition

descending,
radial
Planar Thin-layer Ascending, Adsorption,
descending, partition,
radial ion exchange,
size exclusion
Column Open-column Descending Adsorption,
partition,
ion exchange,
size exclusion
Column GC N /A Adsorption,
partition
Column HPLC N /A Adsorption,
partition,
ion-exchange,
size-exclusion
8.63a Paper and Thin-Layer Chromatography
Paper chromatography and thin layer chromatography (TLC) consti

tute the planar methods mentioned above. Paper chromatography makes
use of a sheet of paper having the consistency of filter paper (cellulose)
for the stationary phase. Since such paper is hydrophilic, the stationary
phase is actually a thin film of water unintentionally adsorbed on the
surface of the paper. Thus, paper chromatography represents a form of
partition chromatography only. The mobile phase is always a liquid.
With thin-layer chromatography, the stationary phase is a thin layer
of material spread across a plastic sheet or glass or metal plate. Such plates
or sheets may either be purchased commercially already prepared or they
may be prepared in the laboratory. The thin-layer material can be any
of the stationary phases described earlier, and thus TLC can be any of
the four types, including adsorption, partition, ion-exchange, and size-
exclusion. Perhaps the most common stationary phase for TLC, however,
is silica gel, a highly polar stationary phase for adsorption chromatography,
as mentioned earlier. Also common is pure cellulose, the same material
for paper chromatography, and here also we would have partition
chromatography. The mobile phase for TLC is always a liquid.
The most common method of configuring a paper or thin-layer experi
ment is the ascending configuration shown in Figure 8.12. The mixture
FIGURE 8.12 The paper or thin-layer chromatography configurations. The drawing

on the left shows the paper or TLC plate with the spots applied. The
drawing on the right shows the chromatogram in the developing
chamber nearing complete development.
to be separated is first "spotted" (applied as a small "spot") within 1 in.

of one edge of a 10 in. square rectangular paper sheet or TLC plate. A
typical experiment may be an attempt to separate several spots represent
ing different samples and standards on the same sheet or plate. Thus, as
many as eight or more spots may be applied on one sheet or plate. So
that all spots are aligned parallel to the bottom edge, a light pencil mark
can be drawn prior to spotting. The size of the spots must be such that
the mobile phase will carry the mixture components without streaking.
This means that they must be rather small; they must be applied with
a very small diameter capillary tube or micropipet. An injection syringe
with a 25 pL maximum capacity is usually satisfactory.
Following spotting, the sheet or plate is placed spotted edge down in
a "developing chamber" that has the liquid mobile phase in the bottom
to a depth lower than the bottom edge of the spots. The spots must not
contact the mobile phase. The mobile phase proceeds upward by capillary
action and sweeps the spots along with it. At this point, chromatography
is in progress; the mixture components will move with the mobile phase
at different rates through the stationary phase and, if the mixture
components are colored, evidence of the beginning of a separation is
FIGURE 8.13 A developed paper of thin-layer chromatogram.
visible on the sheet or plate. The end result, if the separation is successful,
is a series of spots along a path immediately above the original spot
locations each representing one of the components of the mixture spotted
there (see Figure 8.13).
If the mixture components are not colored, any of a number of techniques
designed to make the spots visible may be employed. These include
iodine staining in which iodine vapor is allowed to contact the plate.
Iodine will absorb on most spots rendering them visible. Alternatively,
a fluorescent substance may be added to the stationary phase prior to the
separation (available with commercially prepared plates) such that the
spots, viewed under an ultraviolet light, will be visible because they do
not fluoresce, while the stationary phase surrounding the spots does
fluoresce.
The visual examination of the chromatogram can reveal the identities
of the components, especially if standards were spotted on the same
paper or plate. Retardation factors (so called Rf factors) can also be
calculated and used for qualitative analysis. These factors are based on
the distance the mobile phase has traveled on the paper (measured from
the original spot of the mixture) relative to the distances the components
have traveled, each measured from either the center or leading edge of
the original spot to the center or leading edge of the migrated spot.
distance mixture component has traveled

R --------------------------------------------------------------------- /Qc\
distance mobile phase has traveled ;
These factors are compared to those of standards to reveal the identities

of the components.
Quantitative analysis is also possible. The spot representing the
component of interest can be cut (in the case of paper chromatography)
or scraped from the surface (TLC) and dissolved and quantitated by some
other technique, such as spectrophotometry. Alternatively, modern
scanning densitometers, which utilize the measurement of the absorbance
or reflectance of ultraviolet or visible light at the spot location, may be
used to measure quantity.
Using the TLC concept to prepare pure substances for use in other
experiments, such as standards preparation or synthesis experiments, is
possible. This is called preparatory TLC and involves a thicker layer of
stationary phase so that larger quantities of the mixture can be spotted
and a larger quantity of pure component obtained.
Additional details of planar chromatography, such as methods of
descending and radial development, how to prepare TLC plates, tips on
how to apply the sample, what to do if the spots are not visible, and the
details of preparatory TLC, etc., are beyond the scope of this text.
8.63b Classical Open-Column Chromatography
Another configuration for chromatography consists of a vertically

positioned glass tube in which is placed the stationary phase. It is typical
for this tube to be open at the top, to have an inner diameter on the order
of 1 cm, and to have a stopcock at the bottom, making it similar to a buret
in appearance. With this configuration, the mixture to be separated is
placed at the top of the column and allowed to pass onto the stationary
phase by opening the stopcock. The mobile phase is then added and
continuously fed into the top of the column and flushed through by
gravity flow. The mixture components separate on the stationary phase
as they travel downward and, unlike the planar methods, are then col
lected as they elute from the column. In the classical experiment, a "frac
tion collector" is used to collect the eluting solution. A typical fraction
collector consists of a rotating carousel of test tubes positioned under the
column such that fractions of eluate are collected over a period of time,
FIGRUE 8.14 The classical open-column chromatography configuration with frac

tion collector.
such as overnight or a period of days, in individual test tubes (see Figure

8.14). This then makes qualitative or quantitative analysis possible through
the analysis of these fractions by some other technique, such as spectro
photometry.
The length of the column is determined according to the degree to
which the mixture components separate on the stationary phase chosen.
Difficult separations would require more contact with the stationary
phase and thus may require longer columns. Again, all four types
(adsorption, partition, ion exchange, and size exclusion) can be used with
this technique.
It is well known that classical open-column chromatography has been
largely displaced by the modern instrumental techniques of liquid
chromatography. However, open columns are still used where extensive
sample "clean-up" in preparation for the instrumental method is necessary.
One can imagine that "dirt}/' samples originating, for example, from
animal feed extractions or soil extractions, etc. may have large
concentrations of undesirable components present. Since only very small
samples (on the order of 1-20 pL) are needed for the instrumental method,
the time required for obtaining a clean sample by this method, assuming
the components of interest are not retained, is the time it takes for an initial
amount of mobile phase to pass through from top to bottom. Compared
to the "overnight" time frame, such a clean-up time is quite minimal and
does not diminish the speed of the instrumental methods.
8.6.3c Instrumental Chromatography
The concept of the finely divided stationary phase packed inside a

column allowing the collection of the individual components as they elute
as discussed in the last section presents a useful, more practical alternative.
One can imagine such a column along with a continuous mobile phase
flow system, a device for introducing the mixture to the flowing mobile
phase, and an electronic detection system at the end of the column —
all of this incorporated into a single unit (instrument) used for repeated,
routine laboratory applications. There are two such chromatography
configurations which are in common use today known as GC and HPLC.
These techniques essentially can incorporate all types of column
chromatography discussed thus far (HPLC), as well as those types in
which the mobile phase can be a gas (GC). Both add a degree of efficiency
and speed to the chromatography concept. HPLC, for example, is such
a "high performance" technique for liquid mobile phase systems that a
procedure that would normally take hours or days with open columns
actually requires much less time. The full details of these instrumental
techniques are discussed in Chapters 9 and 10.
8.7 ELECTROPHORESIS
Another separation technique is one which utilizes the varied rates and
direction with which different dissolved ions migrate through the solution
while under the influence of an electric field. This technique is called
"electrophoresis." "Zone Electrophoresis" refers to the common case in
which a medium such as cellulose or gel is used to contain the solution.
A schematic diagram of the electrophoresis apparatus resembles an
electrochemical apparatus in many respects. A power supply is needed
for connection to a pair of electrodes to create the electric field. The
medium and sample to be separated are positioned between the electrodes.
Original sample
location
To power supply To power supply

+
Positive ions Negative ions

migrate toward migrate toward
negative electrode positive electrode
F IG U R E 8.15 Illustration of the concept o f electrophoresis.
The basic concept of the technique and apparatus is illustrated in Figure

8.15.
Electrophoresis is for separating ions, since only ions will migrate
under the influence of an electric field; negative ions migrate to the
positive electrode and positive ions to the negative electrode. Scientists
have found electrophoresis especially useful in biochemistry experiments
in which charged amino acid molecules and other biomolecules need to
be separated. Thus, application to protein and nucleic acid analysis has
been popular.
The principles of separation are (1) ions of opposite charge will migrate
in different directions and become separated on that basis and (2 ) ions
of like charge, while migrating in the same direction, become separated
due to different migration rates. Factors influencing migration rate are
charge values (i.e., - 1 as opposed to - 2 , for example) and/or different
mobilities. The mobility of an ion is dependent on the size and shape of
the ion, as well as the nature of the medium through which it must
migrate. The biomolecules referred to above can vary considerably in size
and shape, and thus electrophoresis is a powerful technique for separating
them. As for the medium used, there are some options, including the use
of an electrolyte-soaked cellulose sheet (paper electrophoresis) or a thin
gel slab (gel electrophoresis), as well as the nature of the electrolyte
solution used and its pH.
Figure 8.16 represents a paper electrophoresis apparatus. The soaked
cellulose sheet is sandwiched between two horizontal glass plates with
the ends dipped into vessels containing more electrolyte solution. The
electrodes are also dipped into these vessels as shown. The sample is
spotted in the center of the sheet, and the oppositely charged ions then
Sam ple
FIGURE 8.16 A paper electrophoresis apparatus.
have room to migrate in opposite directions on the sheet. Qualitative

analysis is performed much as with paper chromatography by comparing
the distances the individual components have migrated to that for
standards spotted on the same sheet. (It may be necessary to render the
spots visible prior to the analysis, as with paper chromatography.)
Problems associated with paper electrophoresis include the siphoning
of electrolyte solution from one vessel through the paper to the other
vessel when the levels of solution in the two vessels are different causing
the spots to possibly migrate in the wrong direction. The solution to this
problem is to ensure that the levels in the two vessels are the same.
Another problem stems from the fact that oxidation/reduction processes
are occurring at the surfaces of the electrodes. This may introduce un
desirable contaminants to the electrolyte solution. These contaminants
may in turn migrate onto the sheet. The solution to this problem is to
isolate the electrodes while still allowing electrical contact, such as with
the use of baffles, to keep the contaminants from diffusing from the
vessels.
A typical gel electrophoresis apparatus is shown in Figure 8.17. The
thin gel slab referred to above is contained between two glass plates. The
slab is held in a vertical position and has notches at the top where the
samples to be separated are spotted or "streaked." In the configuration
shown in the figure, only downward movement takes place, and thus
only one type of ion, cation or anion, can be separated, since there is only
one direction to go from the notch.
A "tracking dye" can be added to the sample so that the analyst can
know when the experiment is completed (the leading edge of the sample
solvent is visible via the tracking dye). Also, the slab can be removed from
the glass plates, and a staining dye can be applied which binds to the
components, rendering them visible. The result is shown in Figure 8.18.
Components with different mobilities through the gel show up as different
Solutions initially layered here Upper
receive samples.
FIGURE 8.17 A gel electrophoresis apparatus (see text for description).
•1 II ■ §
1 1 1
1 ■
1 ■
• 0 1
•1 1 ■
FIGURE 8.18 The appearance of the gel slab after the electrophoresis experiment
and after the components are rendered visible via staining.
bands or streaks on the gel. Qualitative analysis is performed as with

paper electrophoresis; standards are applied alongside the samples, and
the components are identified by their positions relative to the standards.
Finally, isoelectric focusing has been a useful extension of basic gel
electrophoresis in protein analysis. In this technique, a series of ampholytes
is placed on the slab via electrophoresis. An ampholyte is a substance
whose molecule contains both an acidic and basic functional group.
Solutions of different ampholytes have different pH values. Different
ampholyte molecules differ in size and therefore will have varying
mobilities in the electrophoresis experiment. Thus, these molecules migrate
into the slab and take up different positions along the height of the slab
and create a pH gradient through the height of the slab. Amino acid
molecules have different mobilities in different pH environments and

also have their charges neutralized at particular pH values, rendering
them immobile at some position in the gel. The pH at which the sample
component is neutralized is called the "isoelectric point," and this technique
is called "isoelectric focusing," since samples are separated according to
their components' isoelectric points.
C hapter 9
G as C hrom atography
9.1 INTRODUCTION
One instrumental chromatography configuration mentioned in the

preceding chapter is one in which the mobile phase is gaseous and the
stationary phase is either liquid or solid. This configuration is called gas
chromatography and is abbreviated simply GC. It may also be abbreviated
either GLC or GSC in order to stipulate whether the stationary phase is
a liquid (GLC) or a solid (GSC). Most gas chromatography procedures
utilize a liquid stationary phase (GLC), and thus the chromatog
raphy "type" (see Chapter 8 for the distinction between "type" and "con
figuration") is partition chromatography most of the time.
For the novice having just read Chapter 8 , it may be difficult to visu
alize a chromatography procedure that utilizes a gas for a mobile phase.
The gas, often called the "carrier gas," is typically purified helium or
nitrogen. It flows from a compressed gas cylinder container via the
regulated pressure of the cylinder and flow controller through a "col
umn" containing the stationary phase where the separation takes place.
There are two types of columns; two methods of holding the stationary
phase in place. These will be discussed in Section 9.4.
The fact that the mobile phase is a gas creates additional unique
features, one of which has to do with the mechanism of the separation.
225
When we discussed partition chromatography in Chapter 8 , we were, for

the most part, assuming a liquid mobile phase. The partitioning mechanism
in that case involved only the relative solubilities of the mixture components
in the two phases. With GLC, however, the mechanism, while still involv
ing the solubilities of the mixture components in one phase, the liquid
stationary phase, also involves the relative vapor pressures of the com
ponents, since we must have partitioning between one phase that is a
liquid and another phase that is a gas. Further, since the mobile phase
is a gas, the mixture components must also be gases, or at least liquids
with relatively high vapor pressures (elevated temperatures are used),
in order to be carried through the column as gases using a gaseous mobile
phase.
Let us briefly review the concept of vapor pressure. Simply defined,
vapor pressure is the pressure exerted by the vapors of a liquid above
a liquid phase containing that liquid. It is a measure of the tendency of
the molecules of a liquid substance to "escape" the liquid phase and
become gaseous. If there is a strong such tendency, the vapor pressure
is high (typical of nonpolar, low molecular weight liquids). If, however,
there is a weak such tendency, then the vapor pressure is low (typical
of highly polar and/or high molecular weight substances). Thus, example
liquid substances with high vapor pressures are diethyl ether, acetone,
and chloroform. Example liquid substances with low vapor pressures are
water, high molecular weight alcohols, and aromatic halides. In GC,
substances with high vapor pressures will be strongly influenced by the
moving gaseous mobile phase and will emerge from the column quickly
if their solubilities in the stationary phase are low. If their vapor pressures
are high, but their solubilities in the stationary phase are also high, then
they will emerge more slowly. If their vapor pressure is low, but they
have a high solubility in the stationary phase, the time required for
emergence from the column will be long. The time a given mixture
component is retained by the stationary phase in the column from the
time it is first introduced is called the "retention time." Figure 9.1 and
Table 9.1 summarize the vapor pressure concepts discussed here.
9.2 INSTRUMENT DESIGN
Compared to the classical open-column configuration (Chapter 8 ),

instrumental chromatography instruments are highly automated. For
Gas Chromatography 227
gaseous mobile phase (He)
- A -------------------
FIGURE 9.1 An illustration of the vapor pressure effects discussed in the text.
Components A, B, and C have vapor pressures decreasing from A
to B to C and have solubilities in the stationary phase increasing from
A to B to C.
Table 9.1 A Summary of Retention Concepts for GC
Component's Vapor Component's Solubility Retention

Pressure in Stationary Phase Time
High Low Short

High High Intermediate
Low Low Intermediate
Low High Long
example, the mobile phase is automatically and continuously fed into the
column; the sample is introduced (injected) into the flowing mobile phase,
which immediately carries it onto the column; and the components are
automatically electronically detected as they elute from the column, greatly
simplifying identification and quantitation. With GC, the instrument
components include (1 ) the bottled compressed carrier gas, including the
pressure regulator; (2 ) a flow controller at which the carrier gas flow rate
is adjusted; (3) an injection port for introducing the sample; (4) the column
with the stationary phase; (5) the detector; and (6 ) a recording system to
accept the output of the detector. All of these components, except for the
bottle of compressed carrier gas, are often found incorporated into a
single unit. Perhaps most often, however, the recording system, which
can be a simple strip-chart recorder (Chapter 5), is also a separate unit.
A diagram of a gas chromatograph is shown in Figure 9.2.
As mentioned earlier, the carrier gas is almost always helium or ni
trogen. Helium is used most often. The regulated pressure from the
bottled helium, typically 60-80 psi, is sufficient to sustain its flow through
the system. Thus, a bottle of helium fitted with a regulator is almost
FIGURE 9 .2 ' A diagram of a gas chromatograph.
always seen standing next to the GC unit. The helium is often purified
on-line with a molecular sieve column to remove traces of water prior
to entering the instrument. It enters the instrument and flows through
the flow controller, injection port, column, and detector before exiting into
the laboratory air. Some laboratories in which GC is used extensively
utilize small fume hoods to protect the quality of the laboratory air.
9.3 SAMPLE INJECTION
The injection port is designed to introduce samples quickly and effi

ciently. Most GC work involves the separation of volatile liquid mixtures.
In this case, the injection port must be designed to flash vaporize small
amounts of such samples so that the entire amount is immediately carried
to the head of the column by the flowing helium. The most familiar design
consists of a small glass-lined or metal chamber equipped with a rubber
septum to accommodate injection with a syringe. This chamber is heated
to a very high temperature (typically >200°C) and is connected directly
to the column. As the helium "blows" through the chamber, a small
volume of injected liquid (typically on the order of 0.1-3 pL) is thus flash
vaporized and immediately carried onto the column. Small volume
syringes with sharp, beveled tips for piercing the septum are needed. A
variety of sizes and some additional features are available to make the
injection easy and accurate. Syringes manufactured by the Hamilton Co.,

Reno, NV, are common. The mechanics for such an injection are shown
in Figure 9.3. The rubber septum, after repeated sample introduction, can
be replaced easily. Sample introduction systems for gases (gas sampling
valves) and solids are also available.
A volume of liquid as small as 0.1-3 pL (appearing as an extremely
small drop) may seem to be extraordinarily small. When vaporized,
however, such a volume is much larger and will occupy an appropriate
volume in the column. Also, the detection system, as we will see later,
is very sensitive and will detect very small concentrations even in such
a small volume. As a matter of fact, too large a volume is a concern to
the operator, since columns, especially capillary columns, can become
overloaded even with volumes that are very small. Overloading means
that the entire vaporized sample will not fit onto the column all at once
and will be introduced over a period of time. Obviously, a good separation
would be in jeopardy in such an instance. To guard against overloading
of capillary columns, split injectors have been developed. In these injectors,
only a fraction of the liquid from the syringe actually is passed to the
column. The remaining portion is split from the sample and vented to
the air.
As implied above, the appropriate range of sample injection volume
depends on column diameter. As we will see in the next section, column
diameters vary from capillary size (0 .2 - 0 . 3 mm) to Vs and lU in. Table 9 . 2
gives the maximum injection volumes suggested for these column diam
eters. The capillary columns are those in which the overloading problem
mentioned above is most relevant. Injectors preceding the Vs in. or larger
columns are not split.
The accuracy of the injection volume measurement can be very
important for quantitation, since the amount of analyte measured by the
detector depends on the concentration of the analyte in the sample as well
as the amount injected. In Section 9.9, a technique known as the internal
standard technique will be discussed. (It is also discussed in Chapter 5.)
Use of this technique negates the need for superior accuracy with the
injection volume, as we will see. However, the internal standard is not
always used. Very careful measurement of the volume with the syringe
in that case is paramount for accurate quantitation. An acceptable
procedure for this is presented in Table 9.3. Of course, if a procedure calls
only for identification (Section 9.8), then accuracy of injection volume is
less important.
FIGURE 9.3 The mechanics of loading a GC syringe, showing the careful positioning
of the plunger to the correct volume, and the introduction of the
sample to the injection port.
9.4 COLUMNS
9.4.1 Instrument Logistics
GC columns, unlike any other type of chromatography column, are

typically very long. Lengths varying from 2 ft up to 300 ft or more are
Table 9.2 Suggested Maximum Injection Volumes for Various Column Diameters
Column Diameters Maximum Injection Volumes
! / 4 in. (packed column) 100 pL

Vs in. (packed column) 20 pL
Capillary (open tubular) 0.1 pL
Table 9.3 A Syringe Loading and Injection Method When Accuracy of Injection
Volume is Important
1. Flush the syringe thoroughly with clean sample solvent.

2. Expel the solvent from the syringe, then carefully retract the plunger (in air) to
the 1.0 pL mark. A little less than 1 pL of solvent will be present (needle
hold-up).
3. Transfer the syringe to the sample container and slowly draw several microliters
of sample into the syringe barrel. Remove the syringe needle from the sample
container and expel sample until the plunger is at the 2 pL mark (for a 1 pL
injection).
4. Retract the syringe plunger, pulling the needle load entirely into the barrel Two
liquid plugs will be seen — sample and solvent without sample. Note the
volume of the sample plug.
5. Insert the needle to its full length; inject the sample and quickly remove the
syringe.
Source: Volume 2, Operators Manual for Varian Model 3300/3400 Gas Chromatograph.
Used with permission of Varian Associates, Inc., Palo Alto, California.
possible. Additionally, it is important for the column to be kept at an

elevated temperature during the run in order to prevent condensation
of the sample components. Indeed, maintaining an elevated temperature
is very important for other reasons, as we shall see in the next section.
The obvious logistical problem is how to contain a column of such length
and be able to simultaneously control its temperature.
Such a long column is wound into a coil and fits nicely into a small
oven, perhaps 1-3 ft3 in size. This oven probably constitutes about half
of the total size of the instrument (see Figure 9.4). Connections are made
through the oven wall to the injection port and the detector. The tem
peratures of column ovens typically vary from 50 to 150°C, with higher
temperatures possible in procedures that require them. A more thorough
discussion of this subject is found in Section 9.5.
GC instruments are designed so that columns can be replaced easily
by disconnecting a pair of brass fittings inside the oven. This not only
facilitates changing to a different stationary phase altogether, but also
allows the operator to replace a given column with a longer one contain-
b
FIGURE 9.4 Examples of columns: (a) a 6-ft long, Vs in. diameter packed column,
and (b) a 100-ft long capillary column. Both are of such size as to fit
into a small oven. (Courtesy of Varian Associates, Palo Alto, CA.)
ing the same stationary phase. The idea here is to allow more contact with
the stationary phase, which in turn is bound to improve the separation.
If a 6 -ft column is useful for a partial separation, would not a 12-ft column
be that much better?
9.4.2 Packed, Open Tubular, and Preparative Columns
It was indicated earlier that column lengths of up to 300 ft are not

unusual. It should be mentioned here that the longer a column with
stationary phase tightly packed (a so-called "packed" column), the greater
the gas pressure required to sustain a flow. A 20-ft length is approximately
the upper limit for the length of a "packed" column. A somewhat modern
development in this area is the "open-tubular" capillary column. Instead
of tightly packed solid substrate particles holding the liquid stationary
phase inside the column (see Chapter 8 , Section 8 .6 ), the stationary phase
is made to adsorb on the inside wall of a small diameter capillary tube
so that the tube remains open to gas flow in the center. A design such
as this offers very little resistance to gas flow and can be made hundreds
of feet long without having to utilize a large pressure. It is no exaggeration
to say that such columns are so popular today that the packed column
is fast becoming obsolete (refer again to Figure 9.4 and to Figure 9.5).
In addition to the "analytical" columns (columns used mainly for
analytical work), so-called "preparative" columns may also be encountered.
Preparative columns are used when the purpose of the experiment is to
prepare a pure sample of a particular substance (from a mixture containing
the substance) by GC for use in other laboratory work. The procedure
for this involves the individual condensation of the mixture components
of interest in a cold trap as they pass from the detector and as their peak
is being traced on the recorder. While analytical columns can be suitable
for this, the amount of pure substance generated is typically very small,
since what is being collected is only a fraction of the extremely small
volume injected. Thus, columns manufactured with very large diameters
(on the order of inches) and capable of very large injection volumes (on
the order of milliliters) are manufactured for the preparative work. Also,
the detector used must not destroy the sample, like the flame ionization
detector (Section 9.7) does for example. Thus, the thermal conductivity
detector (Section 9.7) is used most often with preparative GC.
(a) (b)
FIGURE 9.5 Illustrations of packed and capillary columns, (a) Cross-section of the
packed column, (b) Cross-section of the open-tubular capillary col
umn.
9.4.3 The Nature and Selection of the Stationary Phase
The liquid stationary phase in a GLC packed column is adsorbed on

the surface of a solid substrate (also called the "support"). This material
must be inert and finely divided (powdered). The typical diameter of a
substrate particle is 125-250 p, creating a 60- to 100-mesh material. These
particles are of two general types: diatomaceous earth and Teflon.
Diatomaceous earth, the decayed silica skeletons of algae, is most
commonly referred to by the manufacturer's (Johns Manville's) trade
name "Chromosorb." Various types of Chromosorb, which have had
different pretreatment procedures applied, are available, such as
Chromosorb P, Chromosorb W, and Chromosorb 101-104. The nature of
the stationary phase, as well as the nature of the substrate material are
usually specified in a chromatography literature procedure, and columns
are tagged to indicate each of these as well.
Since the interaction of the mixture components with the liquid
stationary phase plays the key role in the separation process, the nature
of the stationary phase is obviously important. Several hundred different
liquids useful as stationary phases are known. This means that the analyst
has an awesome choice when it comes to selecting a stationary phase for
a given separation. It is true, however, that relatively few such liquids
Table 9.4 Some Stationary Phases for GLC
Useful for
Structure, Descriptive Mixtures of
Abbreviated or Nondescriptive Name, or Other Compounds
Name Description which are
FFAP A Teflon-based Highly polar

material
Casterwax 0 Highly polar
CH3(CH2)5—CH — CH — CH = C H -(C H 2)i7-
\
OH OH
Carbowax
(variety of
molecular weights) Polar
XE-60 (also CjH,
XF1150, SF-1125) S i(C H ,),-0 -S i- 0 ' Si(CH,)3 Polar
C H ,-C H ,-C H ,-C s N
OV-17 Methyl, phenyl, Somewhat

silicone (a silicone polar
oil)
OV-101 Liquid methyl silicone Nonpolar
OV-1 Methyl siloxane Nonpolar
SE-30 (also
DOW-200, CH3 CH3 Nonpolar
DOW-11, SF-96) — -Si— 0 —S i—0 —
I I
CH3 ch3
U -ip
Apiezon A grease Nonpolar
(various types)
Squalane High molecular weight Nonpolar
Hydrocarbon (C30)
Source: "Basic Gas Chromatography" by McNair, H.M., and Bonelli, E.J., Varian
Instruments, Walnut Creek, CA. 1969. Reprinted with permission, Varian Associ
ates, Inc., Varian Analytical Instruments, Palo Alto, CA.
are in actual common use. Their composition is frequently not obvious

to the analyst because a variety of common abbreviations have come to
be popular for the names of some of them. Table 9.4 lists a number of
common stationary phases, their abbreviated names, a description of
their structures, and the classes of compounds (in terms of polarity) for
which each is most useful.
The selection of a stationary phase depends largely on trial and error
or experience, with consideration given to the polar nature of the mixture,
as noted in Table 9.4 or a similar table. The usual procedure is to select
a stationary phase, based on such literature information, and attempt the
separation under the various conditions of column temperature, length,
carrier gas flow rate, etc. to determine the optimum capability for sepa
rating the mixture in question. If this optimum resolution is not satisfac
tory (see Section 9.6), then an alternate selection is apparently required.
More experienced chromatographers may refer to the McReynolds
Constants for a given stationary phase as a measure of its resolving
power. A complete discussion of this subject, however, is beyond the
scope of this text.
9.5 OTHER VARIABLE PARAMETERS
9.5.1 Column Temperature
Both the vapor pressure and the solubility of a substance in another

substance change with temperature. Figure 9.6 shows, for example, how
the vapor pressure of isobutyl alcohol and the solubility of acetanilide
in ethanol change with temperature. It should not be surprising then that
the precise control of the temperature of a GLC column is very important,
since, as we have indicated, the separation depends on both vapor pres
sure and solubility. Both isothermal (constant) and programmed (con
tinuously changing) temperature experiments are possible. For simple
separations, the isothermal mode may well be sufficient; there may be
sufficient differences in the mixture components' vapor pressures and
solubilities to affect a good separation at the chosen temperature. How
ever, for more complicated mixtures, a complete separation is less likely
in the isothermal mode.
For example, consider gasoline which has a good number of highly
volatile components, as well as a significant number of less volatile
components. It is possible that at a temperature of, say, 1 0 0 °C some of
the less volatile components will be resolved, but the more volatile ones
(a) (b)
FIGURE 9.6 (a) A graph showing how the vapor pressure of isobutyl alcohol
changes with temperature. (From Zubrick, J.W., Organic Chemistry
Survival Manual, 2nd ed. Copyright © John Wiley & Sons Inc., New
York, 1988. With permission.) (b) A graph showing how the solubility
of acetanilide in ethanol changes with temperature. (From Moore, J.,
D. Dalrymple and O. Rodig, Experimental Methods in Organic Chem
istry. Copyright © 1982 by Saunders College Publishing, reprinted by
permission of the publisher. )
will pass through unresolved and have very short retention times. A
lower temperature of, say, 40°C may cause complete resolution of these
more volatile components, but would result in unwanted long retention
times for the less volatile components and perhaps also result in poorly
shaped peaks for these. If we could increase the temperature from 40 to
100°C or higher in the middle of the run, however, we could have the
best of both worlds — complete resolution and reasonable retention times
for all peaks. Thus, temperature programmable ovens have been devel
oped and are now commonplace on virtually all modern GC units.
Temperature programming can consist of simple programs, such as that
suggested above — a single linear increase from a low temperature to
a higher temperature — but it can also be more complex. For example,
a chromatography researcher may find that several temperature increases,
and perhaps even a decrease, must be used in some instances to affect
an acceptable separation. Most modem GC units are capable of at least
a slow temperature decrease in the middle of the mn since they are
equipped with venting fans that bring ambient air into the oven to cool
it. Both a simple program and a more complex program are represented
in Figure 9.7.
Time (m in.) ► Time (m in .) ►

(a) (b)
FIGURE 9.7 (a) A simple temperature program from 40 to 100°C at 5°C per minute,
(b) A more complex temperature program.
9.5.2 Carrier Gas Flow Rate
The rate of flow of the carrier gas affects resolution. A simple analogy
here will make the point. Wet laundry hung out on a clothesline to dry
will dry faster if it is a windy day. The components of the mixture will
"blow" through the column more quickly (regardless of the degree of
interaction with the stationary phase) if the carrier gas flow rate is in
creased. Thus, a minimum flow rate is needed for maximum resolution.
It is well known, however, that at extremely slow flow rates, resolution
is dramatically reduced due to factors such as packing irregularities,
particle size, column diameter, etc. The treatment of these factors and the
quantitative determination of the optimum flow rate is beyond the scope
of this text.
It is obvious that the flow rate must be precisely controlled. The
pressure from the compressed gas cylinder of carrier gas, while sufficient
to force the gas through a packed column, does not provide the needed
flow control of itself. Thus a flow controller, or needle valve, must be part
of the GC system and is often incorporated into the face of the instrument.
In addition, the flow rate must be able to be carefully measured so that
one can know what the optimum flow rate is and be able to match it in
subsequent experiments. Various flow meters are commercially available
for this and often the instrument manufacturer builds one into the instru
ment so that the flow rate is monitored continuously and is observable
as one turns on the needle valve. In other cases, a simple soap bubble
flow meter is often used and can be constructed easily from an old
measuring pipet, a piece of glass tubing and a pipet bulb (see Figure 9.8).
With this apparatus, a stopwatch is used to measure the time it takes a
soap bubble squeezed from the bulb to move between to graduation lines,
such as the 0 and 10 mL lines. The flow rate in milliliters per minute can
thus be calculated.
9.6 THE CHROMATOGRAM
The chart recording giving the written record of the resolved sub
stances, or peaks, is called the chromatogram. All qualitative and quan
titative information obtained from a GC experiment is found in the
chromatogram. One piece of such information is the "retention time,"
symbolized as tR. From the time a substance is injected into the injection
port until it emerges from the column and passes through the detector,
it is being retained by the column. This is the span of time referred to
as the retention time. Since the chart paper is passing through the recorder
at a constant rate (for example, 1 in./min) the recorder itself becomes a
device for measuring retention time. A certain number of inches or
centimeters of chart paper corresponds to a certain number of minutes.
Figure 9.9 shows how this measurement is made on the chromatogram.
Typically, retention times vary from a small fraction of 1 min to about
2 0 min, although much longer retention times have been experienced.
Another parameter often measured is the adjusted retention time, t R.

This is the difference between the retention time of a given component
and the retention time, tM, of an unretained substance, which is often air.
You will recall the injection technique described in Table 9.3 involved the
injection of some air with the sample. Air is usually completely unretained
by a column, and thus the adjusted retention time becomes a measure
of the exact time a mixture component spends in the stationary phase.
Figure 9.10 shows how this measurement is made. The most important
use of this retention time information is in peak identification or qualitative
analysis. This subject will be discussed in more detail in Section 9.8.
Other parameters sometimes obtained from the chromatogram, which
are mostly measures of column efficiency, are "resolution" (R), the number
of "theoretical plates" (N), and the "height equivalent to a theoretical
plate" (HETP or H). These require the measurement of the width of a
FIGURE 9.8 A homemade soap bubble flow meter constructed from an old Mohr
pipet, a piece of glass tubing, and a pipet bulb.
FIGURE 9.9 Retention time is the time corresponding to the length of chart paper
measured from the point of injection to where the peak is at its apex.
peak at the peak base. This measurement is made by first drawing the
tangents to the sides of the peaks and extending these to below the
baseline, as shown for the two peaks in Figure 9.11. The width at peak
FIGURE 9.10 A chromatogram showing the definitions of tR/ t'R< and tM.
base, WB, is then the distance between the intersections of the tangents
with the baseline, as shown. Resolution is defined as the difference in the
retention times of two closely spaced peaks divided by the average
widths of these peaks, as shown mathematically in Figure 9.11. R values
of 1.5 or more would indicate complete separation.
The number of theoretical plates, N, is also mathematically defined in
Figure 9.11. The concept of theoretical plates was discussed briefly in
Chapter 8 , Section 8.3 for distillation. For distillation, one theoretical plate
was defined as one evaporation/condensation step for the distilling liq
uid as it passes up a fractionating column. In chromatography, one
theoretical plate is one "extraction" step along the path from injector to
detector. You will recall in Chapter 8 , Section 8.4 we spoke of chroma
tography as being analogous to a series of many extractions, but with one
solvent (the mobile phase) constantly moving through the other solvent
(the stationary phase), rather than being passed along through a series
of separatory funnels. The equilibration that would occur in the fictional
separatory funnel is one theoretical plate in chromatography.
The height equivalent to a theoretical plate, also mathematically de
fined in Figure 9.11, is that length of column that represents one theo
retical plate or one equilibration step. Obviously, the smaller the value
of this parameter, the more efficient the column. The more theoretical
plates packed into a length of column the better the resolution. Factors
that influence the number of theoretical plates and the resolution are
column length, column temperature, carrier gas flow rate, and other
FIGURE 9.11 The measurement of the "width at base/' which is needed for resolution
and theoretical plate calculations.
factors we have already discussed. Other parameters calculated from the

chromatogram, including capacity factor and selectivity, are defined in
Chapter 10, Section 10.5.
9.7 DETECTORS
Detectors in GC are designed to generate an electronic signal when a

gas other than the carrier gas elutes from the column. There have been
a number of detectors invented to accomplish this. Not only do these
detectors vary in design, but they also vary in sensitivity and selectivity.
Sensitivity refers to the smallest quantity of mixture component for which
it is able to generate an observable signal, and selectivity refers to the type
of compound for which a signal can be generated. The flame ionization
detector, for example, is a very sensitive detector, but does not detect
everything, i.e., it is selective for only a certain class of compounds. The
thermal conductivity detector, on the other hand, detects virtually every
thing, i.e., it is a "universal" detector, but is not very sensitive. What
follows is a brief description of the designs of the detectors that are in
common use, along with some indication of their sensitivity and selec
tivity.
9.7.1 Thermal Conductivity Detector (TCD)
The thermal conductivity detector (TCD) operates on the principle that

gases eluting from the column have thermal conductivities different from
that of the carrier gas, which is usually helium. Present in the flow channel
at the end of the column is a hot filament, hot because it has an electrical
current passing through it. This filament is cooled to an equilibrium
temperature by the flowing helium, but is cooled differently by the
mixture components as they elute, since their thermal conductivities are
different from helium. This change in the cooling process causes the
filament's electrical resistance to change and thus causes the current
flowing through it and the voltage drop across it to change each time a
mixture component elutes. The recorder, which is constantly monitoring
this voltage drop, thus records a peak.
The actual design includes a second filament within the same detector
block. This filament is present in a different flow channel, however, one
through which only pure helium flows. Both filaments are part of a
Wheatstone Bridge circuit as shown in Figure 9 .1 2 , which allows a
"comparison" between the two resistances and a voltage output to the
recorder as shown. Such a design is intended to minimize effects of flow
rate, pressure, and line voltage variations.
Most recently, a flow modulated design has become popular. In this
design, a single filament is used, and the column effluent is alternated
with the pure helium through the flow channel where the filament is
located. This eliminates the need to use two matched filaments.
The thermal conductivity detector is universal (detects everything),
and it is nondestructive (can be used with preparative GC), but less
sensitive than other detectors.
9.7.2 Flame Ionization Detector (FID)
Another very important GC detector design is the flame ionization

detector (FID). In this detector, the column effluent is swept into a hy
drogen flame where the flammable components are burned. In the burning
process, a very small fraction of the molecules becomes fragmented, and
FIGURE 9.12 The thermal conductivity detector.
the resulting positively charged ions are drawn to a "collector" (nega

tively charged) electrode, a metal cylinder above and encircling the flame,
while electrons flow to the positively charged burner head. The nega
tively charged collector and the positively charged burner head are part
of an electrical circuit in which the current changes when this process
occurs, and the change is amplified and seen as a peak on the recorder
trace. Figure 9.13 shows a schematic diagram of this detector. The design
includes the hydrogen flame burner nozzle, the collector electrode, an
inlet for air to surround the flame, and an ignitor coil for igniting the
hydrogen as it emerges from the nozzle.
Apparent on the exterior of the instrument and located near the bench
on which the GC unit sits are pressure regulated compressed gas cylinders
of hydrogen and air, as well as the helium. Metal tubing, typically V8 in.
diameter, connect the cylinders to the detector, with a needle valve for
flow control in between. These valves are located in the instrument for
easy access and control by the operator.
The FID is very sensitive, but is not universal, and also destroys (burns)
the sample. It only detects organic substances that bum and fragment in
a hydrogen flame. These facts preclude its use for preparative GC or for
inorganic substances that do not bum, such as water, carbon dioxide, etc.
Still, it is a very popular detector, given its sensitivity and given the fact
that most analytical work involves flammable organic substances.
Collector
electrode
Ignitor
coil
H ydrogen—►
:
> — Column effluent
FIGURE 9.13 The flame ionization detector.
9.7.3 Electron Capture Detector (ECD)
A third type of detector, required for some environmental and bio

medical applications, is the electron capture detector (ECD). This detector
is especially useful for large halogenated hydrocarbon molecules, since
it is the only one which has an acceptable sensitivity for such molecules.
Thus, it finds special utility in the analysis of halogenated pesticide
residues found in environmental and biomedical samples.
The electron capture detector is another type of ionization detector.
Specifically, it utilizes the beta emissions of a radioactive source, often
nickel-63, to cause the ionization of the carrier gas molecules, thus
generating electrons which constitute an electrical current. As an
electrophilic component, such as a pesticide, from the separated mixture
enters this detector, the electrons from the carrier gas ionization are
"captured" creating an alteration in the current flow in an external circuit.
This alteration is the source of the electrical signal which is amplified and
sent on to the recorder. A schematic diagram of this detector is shown
in Figure 9.14. The carrier gas for this detector is either pure nitrogen or
a mixture of argon and methane.
An additional consideration regarding pesticides warrants mentioning
here. Most of these compounds decompose on contact with hot metal
surfaces. This problem has, however, been adequately solved for most
pesticides by constructing the entire path of the sample out of glass or
glass-lined materials. Thus, glass or glass-lined injection ports and all
glass columns are available.
column
effluent
radioactive
source
FIGURE 9.14 The electron capture detector.
In terms of advantages and disadvantages, the ECD is extremely

sensitive, but only for a very select group of compounds — halogenated
hydrocarbons. Other gases will not give a peak. It does not destroy the
sample and thus may be used for preparative work.
9.7.4 Nitrogen/Phosphorus Detector (NPD)
While the ECD is useful for chlorinated hydrocarbon pesticides, the

NPD, also known as the "thermionic" detector, is useful for the phosphorus
and nitrogen-containing pesticides, the organophosphates and carbamates.
The design, however, represents a slight alteration of the design of the
FID. In the NPD, we basically have an FID with a bead of alkali metal
salt positioned just above the flame. The hydrogen and air flow rates are
lower than in the ordinary FID, and this minimizes the fragmentation of
other organic compounds. These changes result in a somewhat mysterious
increase in both the selectivity and sensitivity for the pesticides.
9.7.5 Flame Photometric Detector (FPD)
A detector that is specific for organic compounds containing sulfur or

phosphorus is the flame photometric detector (FPD). A flame photometer
(Chapter 7) is an instrument in which a sample solution is aspirated into
a flame and the resulting emissions from the flame are measured with
a phototube detector. The FPD is a flame photometer positioned to accept
the effluent from the column in place of the aspirated sample. The flame
in this case is a hydrogen flame as in the FID. The basic operating principle
is that the sulfur or phosphorus compounds bum in the hydrogen flame
and produce light emitting species. A monochromator, typically a glass
filter, makes this detector specific for the compound of interest, and thus
only one peak appears on the recorder trace. The signal for the recorder
is the signal proportional to light intensity that is produced by the pho
totube.
The advantages are that it is a very selective detector and also very
sensitive. Disadvantages include the problems associated with the need
to carefully control the flame conditions so that the correct species are
produced (S=S for the sulfur compounds and HPO for the phosphorus
compounds). Such conditions include the gas flow rates and the flame
temperature. It is a destructive detector.
9.7.6 Electrolytic Conductivity (Hall)
The Hall detector converts the eluting gaseous components into ions
in liquid solution and then measures the electrolytic conductivity of the
solution in a conductivity cell. The solvent is continuously flowing through
the cell, and thus the conducting solution is in the cell for only a moment
while the conductivity is measured and the peak recorded before it is
swept away with fresh solvent. The conversion to ions is done by chemi
cally oxidizing or reducing the components with a "reaction gas" in a
small reaction chamber made of nickel positioned between the column
and the cell. The nature of the reaction gas depends on what class of
compounds is being determined. Organic halides, the most common
application, use hydrogen gas at 850°C or higher as the reaction gas. The
strong HX acids are produced, which give highly conductive liquid
solutions.
The Hall detector has excellent sensitivity and selectivity, giving a
peak for only those components which produce ions in the reaction
chamber. It a destructive detector.
9.7.7 GC-MS and GC-IR
In Chapter 6 , we discussed the fundamentals of mass spectrometry and

IR spectrometry. The quadrupole mass spectrometer and the Fourier
Transform IR spectrometer have been adapted to and used with GC
equipment as detectors with great success in recent years. Gas
chromatography-mass spectrometry (GC-MS) and gas chromatography-
infrared spectrometry (GC-IR) are very powerful tools for qualitative
analysis in GC because they not only give retention time information, but,
due to their inherent speed, they are able to measure and record the mass
or IR spectrum of the individual sample components as they elute from
the GC column. It is like taking a photograph of each component as it
elutes (see Figure 9.15). Coupled with the computer banks of mass and
IR spectra, a component's identity is an easy chore for such a detector.
Recently fabricated GC-MS units have become very compact, in con
trast to older units which take up large amounts of space. It seems the
only disadvantage remaining is the expense, although that also seems to
be improving. The only other slight disadvantage is the fact the large
amounts of computer memory space are required to hold the amount of
spectral information required for a good qualitative analysis.
Both the GC-MS instrument and the GC-IR instrument obviously
require that the column effluent be fed into the detection path. For the
IR instrument, this means that the IR cell, often referred to as a "light
pipe," be situated just outside the interferometer (Chapter 6 ) in the path
of the light, of course, but must also have a connection to the GC column
and an exit tube where the sample may possibly be collected. The IR
detector is nondestructive. With the mass spectrometer detector, we have
the problem of the low pressure of the MS unit coupled to the ambient
pressure of the GC column outlet. A special device is needed as a "go-
between."
9.7.8 Photoionization Detector (PID)
The photoionization detector (PID), as the name implies, involves the

ionization of eluting mixture components by light, specifically, UV light.
The UV source emits a wavelength characteristic of the gas (either helium
of argon) inside. This light passes into an "ionization chamber" through
a metal fluoride window and into the path of the column effluent there.
This is where the mixture components absorb the light and ionize. The
resulting ions are detected through the use of a pair of electrodes in the
ionization chamber, the current from which constitutes the signal to the
recorder. The specific lamp and window are chosen according to the
ionization energy needed for the compounds in the sample.
Since different lamps and windows are available, this detection method
can often be selective for only some of the components present in the
sample. Its sensitivity is especially good for aromatic hydrocarbons and
inorganics. It is a very sensitive nondestructive detector.
FIGURE 9.15 The "photographs" (the MS or IR spectra) of individual mixture

components are obtained with GC-MS and GC-IR instruments.
9.8 QUALITATIVE ANALYSIS
As mentioned in Section 9.6, the parameters that are most important

for a qualitative analysis using most GC detectors are the retention time,
and the adjusted retention time, t'R. Their definitions were graphically
presented in Figure 9.10. Under a given set of conditions (the nature of
the stationary phase, the column temperature, the carrier flow rate, the
column length and diameter, and the instrument dead volume), the
retention time is a particular value for each component. It changes only
when one or more of the above parameters changes. Thus, repeated
injections into a given system under a given set of conditions should
always yield a particular retention time for a given compound, and
qualitative analysis using this system only requires accurate measurement
of tR. When one of the parameters changes, such as when an analyst in
another laboratory sets up with a different dead volume or perhaps a
slightly different stationary phase composition, for example, then the
retention time for that component will be slightly different. The adjusted
retention time will correct for changes in the dead volume, but will not
correct for any other change. A parameter defined as the "relative reten
tion," however, will adjust for other changes. This parameter compares
the retention of one component (1 ) with another (2 ) and is given the
symbol alpha (a). It is defined as follows:
*k(>)
<«>
The relative retention is thus an important parameter for qualitative

analysis if the work involves other setups with other instruments and
columns which do not exactly match the original.
The usual qualitative analysis procedure, then, is to establish the
conditions for the experiment, perhaps by trial and error in one's own
laboratory or by matching conditions outlined in a given procedure, and
then to match the retention time data, either ordinary retention time, *R'
or the relative retention, a, whichever is appropriate, for standards (pure
samples) with that for the unknown. The analyst can then proceed to
match the retention time data for the unknown to those of the pure
samples to determine which substances are present.
One caution is that there may be more than one component with the
same retention (no separation), and thus further experimentation may be
required. For example, when working with a complex mixture whose
components are perhaps not all known, it may be necessary to change
the experimental conditions to determine whether a given peak is due
to one component (known) or more (e.g., one known and one unknown).
Changing the stationary phase may prove useful. Such a change would
produce a chromatogram with completely different retention times and
even possibly a different order of elution. Thus two components that were
coeluted before may now be separated, evidence for which would be a
different peak size for the known component.
9.9 QUANTITATIVE ANALYSIS
9.9.1 Peak Size Measurement
The physical size of the peak traced out by the strip chart recorder is
directly proportional to the amount of that particular component passing
through the detector. Thus, it is imperative that we have an accurate
method for determining this peak size if it is the quantity of a component
in the mixture that is sought. There are a variety of methods that have
height method, which simply measures the height from the baseline to
the apex of the peak, the triangulation method, which measures the area
of a triangle drawn to approximate the peak, and the half-width method,
which measures the area of a rectangle drawn to approximate the peak.
These three methods are illustrated in Figure 9.16.
While the first is a peak "height" method, the others are peak "area"
methods. In the triangulation method, the height and base of the triangle
is measured as shown and the area calculated (bh/2). In the half-width
method, the height of the peak and the width of the peak at half-height
are measured, and these represent the length and width of a rectangle,
thus the area is again easily calculated (hw). None of these methods are
terribly accurate, but they are fast and do not require expensive equip
ment. The peak height method is especially useful (fast) when only a
rough indication of quantity is desired.
The most popular method of measuring peak size is by integration.
Integration is an area measuring method in which a series of "heights"
are measured from the moment the pen begins to deflect until the baseline
is completely restored, as illustrated in Figure 9.17. This is conveniently
done by computer, since a computer works with digital values derived
from the analog data output by the instrument to the recorder (Chapter
5). The method is thus easy, fast, and accurate.
The computer hardware for integration can be one of a variety of
designs, from a small unit designed only for measuring chromatography
peaks (a "computing integrator") to a larger system, such as a
microcomputer or other computer, programmed using independently
prepared software. Figure 9.18 shows an example of a computing integrator
and the printed output. Such a device often replaces the ordinary recorder,
since it prints the peaks as a recorder would. It also records the retention
time next to each peak as the peak is recorded. Note the area values in
the sample printout (under the "area" heading). These values represent
the sum of the series of digital values represented by the heights illustrated
in Figure 9.17.
9.9.2 Quantitation Methods
Several different approaches exist as to what peaks are measured and

how the mixture component of interest is actually quantitated. We now
discuss two of the more popular methods (see also Chapter 5).
252
Peak Height Half Width Triangulation
FIGURE 9.16 Three "manual" methods for measuring peak size.

FIGURE 9.17 An illustration of the measurement of peak size by integration. The

sum of a series of vertical "heights," such as illustrated, represents
the area.
9.9.2a The Response Factor Method
Consider a four-component mixture to be analyzed by GC. The chro

matogram may look something like that shown in Figure 9.19. One might
think it logical that in order to quantitate the mixture for, say, component
B, all one would need to do is to measure the sizes of all four peaks and
divide the size of the peak representing B by the total of all four.
AreaD
%B = --------------------------- s-------------------------------- /o
Area.A 4- AreaD B
+ Area^ C+ Area^ D
v * '
The problems with this approach are (1) without comparing the peaks
to a standard or a set of standards, it is not known whether the result
is a weight percent, volume percent or mole percent. (2) The instrument
detector does not respond to all components equally. For example, not
all components will have the same thermal conductivity, and thus the
thermal conductivity detector will not give equal sized peaks for equal
concentrations of any two components. Thus, the sum of all four peaks
would be a meaningless quantity, and the size of peak B by itself would
not represent the correct fraction of the total.
It is possible, however, to measure a so-called response factor for the
analyte, which is the area generated by a unit quantity injected, such as
a microliter (jxL) or microgram (pg). The procedure is to inject a known
quantity of the analyte, measured by the position of the plunger in the
CHANNEL ft INJECT 8 5 -8 6 -8 6 8 4 :2 8 :5 3
8 5 - 8 6 - •86 8 4 : 2 8 : 5 3
F IL E 1. METHOD 8. RUN 12 INDEX :
PEAK# AREflX RT AREA BC
i 33. 696 1. 48 1 5 2 71 81
0 18. 766 2. 39 4 879 01
3 55. 538 3. 21 2 5 1 7 8 8:L
TOTAL 1 88. 453^8
FIGURE 9.18 (a) A computing integrator and (b) the printout from a computing
integrator.
FIGURE 9.19 A chromatogram of a four-component mixture.
syringe (pL), or by weighing the syringe before and after filling. The peak
size that results is measured and divided by this quantity:
size of peak
response factor = ---------------------------------------------- (9 3 )
quantity of pure sample injected
The quantity of analyte in an unknown sample is then determined by

measuring the peak size of the analyte, resulting from an injection of a
known quantity of unknown sample and dividing by the analyte's re
sponse factor:
peak size
quantity of analyte = ---------------------- (9 4 )
response factor v ‘
The percent of the analyte can then by calculated as follows:
quantity of analyte (from Equation 9 .4 )

% of analyte = ---------------------— *----------- --------------- L m gx
total quantity injected
In this method, only the peak of the analyte need be measured in the four-
component mixture in order to quantitate this component.
9.9.2b Internal Standard Method
Since the peak size is directly proportional to concentration, one may

think that one could prepare a series of standard solutions and obtain
peak sizes for a plot of peak size vs concentration, a method similar to
Beer's Law in spectrophotometry, for example. But since peak size also
varies with amount injected, there can be considerable error due to the
difficulty in injecting consistent volumes as discussed above and in Section
9.3. A method that does away with this problem is the internal standard
method (see Chapter 5, Section 5.3). In this method, all standards and
sample are spiked with a constant known amount of a substance to act
as what is called an internal standard. The purpose of the internal standard
is to serve as a reference point for the peak size measurements so that
slight variations in injection technique and volume injected are
compensated by the fact that the internal standard peak and the analyte
peak are both affected by the slight variations, and thus the problem
cancels out.
The procedure is to measure the peak sizes of both the internal standard
peak and the analyte peak and then to divide the analyte peak size by
the internal standard peak size. The "area ratio" thus determined is then
plotted vs concentration of the analyte. The result is a method in which
the volume injected is not as important and, in fact, can vary substantially
from one injection to the next.
Can just any substance serve as an internal standard? There are certain
characteristics that the internal standard should have, and these are listed
below.
(1) Its peak, like the analyte's, must be completely resolved from all
other peaks.
(2) Its retention time should be close to that of the analyte.
(3) It should be structurally similar to the analyte.
9.10 TROUBLESHOOTING
Problems that arise during a GC experiment usually manifest them

selves on the chromatogram. Examples of such manifestations are peak
shapes being distorted, peak sizes diminishing for reasons other than
quantity of analyte, the baseline drifting, the retention times changing for
no apparent reason, etc. These kinds of problems can usually be traced
to injection problems, problems with the column, or problems with the
detector. There can, of course, be problems associated with the electronics
of the instrument. However, we will not be concerned with those here
because of the large number of different instrument designs that have
been manufactured over the years. The operator can usually find assis
tance for these in a troubleshooting section of the manuals that accom
pany the instrument.*
In the following paragraphs, we will address some of the most common
problems encountered, pinpoint possible causes, and suggest methods
of solving the problems.
Diminished Peak Size — We could also refer to this as reduced
sensitivity. The peaks are smaller than expected based on previous
observations when equal or greater quantities of a particular sample were
injected. Such an observation usually means a problem with injection (less
injected than assumed) or a problem with the detector such that a smaller
electronic signal is sent to the recorder. One should check for a leaky or
plugged syringe, a worn septum, a leak in the pre- and post-column
connections or a contaminated detector. Of course, detector attenuation,
recorder sensitivity settings, electrical connections, and other associated
hardware problems are potential causes.
Unsymmetrical Peak Shapes — Peak "fronting" or peak "tailing"
(Figure 9.20) are typical examples of this problem. These could be indicators
of poor injections, meaning too large an injection volume, too slow with
the syringe manipulation during injection, or not fully penetrating the
See also the "GC Troubleshooting" column published regularly in the monthly journal
LC*GC, The Magazine of Separation Science., Aster Publishing Corporation, Eugene, OR.
FIGURE 9.20 (a) A peak exhibiting fronting and (b) a peak exhibiting tailing.
septum. It may indicate a decomposition of thermal labile components

in contact with the hot system components, such as the metal walls of
the injection port and column. It may also mean contamination of the
injection port and/or column.
Altered Retention Times — This is usually caused by changes in the
carrier gas flow rate or column temperature. Flow rate changes can be
caused by leaks in the system upstream from the column inlet, such as
in the injection port (e.g., the septum); by low pressure in the system due
to an empty or nearly empty carrier supply; or by faulty hardware, such
as the flow control valve or pressure regulator. Temperature changes can
be caused by a faulty temperature controller, an improperly set temperature
program, too short a cool-down period prior to the next injection in a
temperature programmed experiment, etc. This could also be caused by
overloading the column or by diminished effectiveness (decomposition?)
of the stationary phase.
Baseline Drift — This occurs when a new column has not been
sufficiently conditioned, when the detector temperature has not reached
its equilibrium value, or when the detector is contaminated or otherwise
faulty. New columns need to be conditioned, usually with an overnight
"bakeout" at the highest recommended temperature for that column.
Detector signals may very well change when the detector temperature
changes. One should be sure that sufficient time has been given for the
detector temperature to level off. The nature of detector problems depends
of course on the detector. TCD filaments may become oxidized due to
an air leak, ionization detectors may be leaking, or there may be a crack
in the FID burner nozzle, etc.
Baseline Perturbations — If the perturbations are in the form of spikes

of an irregular nature, the problem is likely to be detector contamination.
Such spikes are especially observed when dust particles have settled into
the HD flame orifice. Of course, the problem may also be due to interference
from electrical pulses from some other source nearby. Regular spikes can
be due to condensation in the flow lines causing the carrier, or hydrogen
(FID), to "pulse" or they can be due to a bubble flowmeter attached to
the outlet of the TCD, as well as the electrical pulses referred to above.
These can also be caused by pulses in the carrier flow due to a faulty flow
valve or pressure regulator.
Appearance of Unexpected Peaks— Unexpected peaks can arise from
components from a previous injection that moved slowly through the
column; contamination from either the reagents used to prepare the
sample or standards; or from a contaminated septum, carrier, or column.
The solution to these problems include a rapid "bakeout" via temperature
programming after the analyte peaks have eluted; using pure reagents;
and replacing or cleaning septa, carrier, or column.
C hapter 10
H igh P erform ance L iquid

C hromatography
10.1 INTRODUCTION
10.1.1 Basic Concepts
High performance liquid chromatography (HPLC) is an instrumental

chromatography method in which the mobile phase is a liquid. The
principles of liquid chromatography (LC) in general, including "types"
or separation mechanisms, as well as a brief introduction to HPLC, are
presented in Chapter 8 . All types of liquid chromatography discussed in
that chapter can be utilized in the HPLC configuration. Thus we have
partition (LLC), adsorption (LSC), bonded-phase (BPC), ion-exchange
(IEC and IC), and size-exclusion (SEC), including gel-permeation (GPC)
and gel-filtration (GFC), all as commonly used types of HPLC. The
instrumental design is based on concepts similar to gas chromatography
(GC), especially in terms of the detection and recording schemes following
the separation. You are referred to Chapter 9 for a discussion of GC
principles.
261
Simply stated, HPLC involves the high pressure flow of a liquid mobile
phase through a metal tube (column) containing the stationary phase,
with electronic detection of mixture components occurring on the effluent
end. The high pressure, often reaching 4000-6000 psi, is derived from a
special pulsation-free pump, which will be described. The detection system
can be any one of several designs, as with GC, and each of these will be
discussed. In addition, special "solvent delivery" systems and injection
systems are common and will also be described.
The rise in popularity of HPLC is due in large part to the advantages
offered by this technique over the older, noninstrumental, "open column"
method described in Chapter 8 . The most obvious of these advantages
is speed. Separation and quantitation procedures that require hours and
sometimes days with the open column method can be completed in a
matter of minutes, or even seconds, with HPLC. Modern column
technology and gradient solvent elution systems, which will be described,
have contributed significantly to this advantage in that extremely complex
samples can be resolved with ease in a very short time.
The basic HPLC system, diagrammed in Figure 1 0 .1 , consists of a
solvent (mobile phase) reservoir, pump, injection device, column, and
detector. The pump draws the mobile phase from the reservoir and
pumps it through the column as shown. At the head of the column is
the injection device which introduces the sample to the system. On the
effluent end, a detector, pictured in Figure 10.1 as a UV absorption
detector, detects the sample components and the resulting signal is
displayed as peaks on a strip-chart recorder. Besides these basic
components, an HPLC unit may be equipped with a gradient programmer
(Section 10.2), an auto sampler, a "guard column" and various in-line
filters, and a computing integrator or other data handling system.
10.1.2 Comparisons with GC
There are many similarities between the HPLC configuration and the
GC configuration. First, the stationary phase consists of small solid particles
packed inside the column. Second, there is an injection device at the head
of the column through which the mixture to be separated is introduced
into the flowing mobile phase. Third, there is a detection system on the
effluent side of the column which generates an electrical signal when
something other than the mobile phase elutes. Fourth, the electronic
High Performance Liquid Chromatography 263
Concept of Liquid Chromatography
Sample containing dis Column. Dissolved components A, B, and C

solved components A, B, migrate at different rates through column
and C is injected at top of packing. Each component forms a distinct
column. zone.
Fraction Collector
Detector. UV light absor

bance of sample compo
nent is measured, appear
ing as “peak" on recorder.
FIGURE 10.1 A diagram of an HPLC system in which mixture components A, B,

and C are separated. (Courtesy of ISCO, Inc., Lincoln, NE.)
signal is fed into a strip-chart recorder where peaks are recorded — a

system identical to GC.
Because the mobile phase is a liquid, however, there are also some very
obvious differences in the two configurations. First, the mechanism of
separation in HPLC involves the specific interaction of the mixture
components with a specific mobile phase composition, while in GC, the
vapor pressure of the components, and not their "interaction" with a
specific carrier gas, is the most important consideration (see Chapter 9).
Second, the force which sustains the flow of the mobile phase is that of
a high pressure pump, rather than the regulated pressure from a
compressed gas cylinder. Third, the injection device requires a totally
different design due to the high pressure of the system and the possibility
that a liquid mobile phase may chemically attack a rubber septum. Fourth,
the detector requires a totally different design because the mobile phase
is a liquid. Finally, the injector, column, and detector need not be heated
as in GC, although the mode of separation occurring in the column can
be affected by temperature changes, and thus sometimes elevated column
temperatures are used.
10.1.3 Sample and Mobile Phase Pretreatment
The packed bed of finely divided stationary phase particles through

which the mobile phase percolates is an excellent filter for the mobile
phase and injected samples. Particles in the mobile phase as small as 5.0
x lO- 6 cm in diameter can be filtered out by the stationary phase. The
result of this is a decreased effectiveness of the column with time. In a
reasonably short period of time, the particles filtered out on the column
will (1 ) mask the stationary phase, preventing the mixture from interacting
with it and thus causing poor resolution, and (2 ) make it necessary to use
extremely high pressures in order to get the mobile phase through the
column at the prescribed flow rate. The result is a dramatic decrease in
the life of the column, an item of significant expense.
The solutions to this problem involve the prefiltering of all mobile
phases and samples before beginning the experiment. For mobile phases
and large sample volumes, this involves utilizing a vacuum apparatus,
such as that pictured in Figure 10.2. There are many choices for filter
materials. For nonaqueous solvents and their water solutions, paper is
not a good choice due to the possibility of chemical attack which may
cause contamination. For these, a Teflon-based, or other compatible
material, should be used. A common Teflon-based designation is the
PTFE designation. When these are used, if the mobile phase contains
some water, the filter must be "wetted" first with some pure organic
solvent in order to provide a reasonable filtration rate. Aqueous solutions
are often impossible to filter unless the Teflon-based filter is first wetted
in this manner. In cases in which only very small amounts of sample (or
standard solutions) are available, a small syringe-type filtering unit is
used. Here again, the filter must be wetted first if the filter is a Teflon-
based material and the sample is partially or 1 0 0 % aqueous.
In addition to these prefiltering steps, a series of in-line filters and a
"guard column" are often used. The first such filter is at the very beginning
of the mobile phase flow stream in the mobile phase reservoir, while
others are at other strategic points, such as immediately following the
injector. The guard column is usually placed just before the "analytical"
column. Its function is often to remove not just particles, but other
contaminating substances— substances that perhaps have long retention
times on the analytical column and that eventually interfere with the
detection in later experiments. Guard columns are inexpensive and
disposable and are changed frequently.
FIGURE 10.2 A vacuum filtration apparatus for mobile phases and large volume
samples. (Courtesy of Waters Division of Millipore, Inc., Milford,
MA.)
Another problem is the appearance of air pockets in the HPLC system.

If a sample or mobile phase has a significant amount of gas (air) dissolved
in it, a pressure drop, which sometimes is experienced in an HPLC line,
can cause these gases to withdraw from the solution. The air pockets that
result are void spaces that can cause erratic readings from detectors, cause
problems in pumps, and decrease the effectiveness of columns. The
problem is alleviated by degassing the mobile phase and sample in
advance of the their entering the system. Some instruments are equipped
with degassing units between the mobile phase reservoirs and the pump.
Usually, however, the analyst will degas the mobile phase and samples
in a separate experiment in advance by creating a vacuum over the liquid

with the use of a vacuum pump and/or agitating the liquid with use of
an ultrasonic bath. A time-saving technique is to filter and degas in one
step, since both procedures can involve the use of a vacuum. To do this,
the vacuum (receiving) flask can be placed in the ultrasonic bath as the
filtration proceeds.
10.2 SOLVENT DELIVERY
10.2.1 Pumps
The pump that is used in HPLC cannot be just any pump. It must a
special pump that is capable of very high pressure in order to pump the
mobile phase through the tightly packed stationary phase at a reasonable
flow rate, usually between 0.5 and 4.0 mL/min. It also must be nearly
free of pulsations so that the flow rate remains even and constant
throughout. Only manufacturers of HPLC equipment manufacture such
pumps.
There are several pump designs in common use. Probably the most
common is the "reciprocating piston" pump shown in Figure 10.3. In this
pump, a small piston is driven rapidly back and forth, drawing liquid
in through the inlet check valve during its backward stroke and expelling
the liquid through the outlet check valve during the forward stroke.
Check valves allow liquid flow only in one direction. This design is often
a "twin piston" design in which a second piston is 180° out of phase with
the first. This means that when one piston is in its forward stroke, the
other is in its backward stroke. The result is a flow that is free of pulsations.
With the single piston design, a pulse damping device following the
pump is desirable.
10.2.2 The Gradient Programmer
There are two mobile phase elution methods that are used to elute
mixture components from the stationary phase. These are referred to as
isocratic elution and gradient elution. Isocratic elution is a method in
t
Piston
To
column
Check valves
FIGURE 10.3 A diagram of a reciprocating piston HPLC pump.
which a single mobile phase composition is in use for the entire separation
experiment. A different mobile phase composition can be used, but the
transfer to a new composition can only be done by stopping the flow,
changing the mobile phase reservoir, and restarting the flow. Gradient
elution is a method in which the mobile phase composition is changed,
often gradually, in the middle of the run, analogous to temperature
programming in GC.
In any liquid chromatography experiment, the composition of the
mobile phase is very important in the entire separation scheme. In Chapter
8 , we discussed the role of a liquid mobile phase in terms of the solubility
of the mixture components in both phases. Rapidly eluting components

are highly soluble in the mobile phase and insoluble in the stationary
phase. Slowly eluting components are less soluble in the mobile phase
and more soluble in the stationary phase. Retention times, and therefore
resolution, can be altered dramatically by a change in the mobile phase
composition. The chromatographer can use this fact to his/her advantage
by being able to change the mobile phase composition in the middle of
the run. This is the basis for the gradient elution method.
The gradient programmer is a hardware module used to achieve this
goal. The gradient programmer is capable of drawing from at least two
mobile phase reservoirs at once and gradually, in a sequence programmed
by the operator in advance, changing the composition of the mobile phase
delivered to the HPLC pump. A schematic diagram of this system is
shown in Figure 10.4a, and a sample "program" is shown in Figure 10.4b.
(a) (b)
FIGURE 10.4 (a) A schematic diagram of a gradient programming system and (b)
a sample program.
Solvent "strength" is a designation of the ability of a solvent (mobile

phase) to elute mixture components. The greater the solvent strength, the
shorter the retention times.
10.3 SAMPLE INJECTION
As mentioned previously, introducing the sample to the flowing mobile

phase at the head of the column is a special problem in HPLC due to
the high pressure of the system and the fact that the liquid mobile phase
may chemically attack the rubber septum. For these reasons, the use of
the so-called "loop injector" is the most common method for sample
introduction.
The loop injector is a two position valve which directs the flow of the
mobile phase along one of two different paths. One path is a sample loop,
which when filled with the sample causes the sample to be swept into
the column by the flowing mobile phase. The other path bypasses this
loop while continuing on to the column, leaving the loop vented to the
atmosphere and able to be loaded with the sample free of a pressure
differential. Figure 10.5 is a diagram of this injector, showing both the
"LOAD" position and the "INJECT" positions and the flow of the mobile
phase in both positions. The sample loop has a particular volume which
FIGURE 10.5 The loop injector for HPLC. (top) In the "LOAD" position, the sample
is loaded into the loop via a syringe at atmospheric pressure, (bottom)
In the "INJECT" position, the mobile phase sweeps the contents of
the loop onto the column.
is of such accuracy that measuring the sample volume with the syringe
loader is unnecessary, unless volumes smaller than this loop volume are
required. This feature aids in the injection of an accurate, reproducible
sample volume, which can increase the accuracy of a quantitation.
Automated injectors are often used when large numbers of samples
are to be run. Most designs involve the use of the loop injector coupled
to a robotic needle which draws the samples from vials arranged in a
carousel-type auto sampler. Some designs even allow sample preparation
schemes, such as extraction and derivatization, to occur prior to injection.
10.4 COLUMN SELECTION
The stationary phases available for HPLC are as numerous as those

available for GC. As mentioned previously, however, adsorption, partition
(including adsorbed and bonded-phase, BPC — see Chapter 8 ), ion-

exchange, and size-exclusion are all LC methods. We can therefore at least
classify the stationary phases according to which of these four types of
chromatography they represent. Additionally, partition HPLC, which is
the most common, is further classified as "normal-phase" HPLC or "re-
verse-phase" HPLC. Let us begin with these.
10.4.1 Normal-Phase Columns
Normal-phase HPLC consists of methods which utilize a nonpolar

mobile phase in combination with a polar stationary phase. Adsorption
HPLC actually fits this description, too, since the adsorbing solid stationary
phase particles are very polar. (See discussion of adsorption columns in
this section.) Normal-phase partition chromatography makes use of a
polar liquid phase chemically bonded to these polar particles, which
typically consists of silica, Si-O-, bonding sites. Sometimes called bonded-
phase chromatography, this is also the method by which reverse-phase
stationary phases and indeed some GC stationary phases are held in
place. Typical examples of normal-phase bonded phases are those in
which a cyano group (-CN), an amino group (-NH2), or a diol group
(-CHOH-CH 2 OH) are part of the structure of the bonded-phase. Des
ignation of such structural features are often given in the manufacturer's
names. Some examples of these are Chromegabond DIOL, LiChrosorb
DIOL, MicroPak-CN, uBondapak-CN, Nucleosil-NH2, and Zorbax-NH2.
Typical mobile phases for normal-phase HPLC are hexane, cyclohexane,
carbon tetrachloride, chloroform, benzene, and toluene.
10.4.2 Reverse-Phase Columns
Reverse-phase HPLC describes methods which utilize a polar mobile

phase in combination with a nonpolar stationary phase. As stated earlier,
the nonpolar stationary phase structure is a bonded-phase — a structure
that is chemically bonded to the silica particles. Here, typical column
names often have the carbon number designation indicating the length
of a carbon chain to which the nonpolar nature is attributed. Typical
designations are C8 or C 1 8 (or ODS, meaning "octadecyl"), etc. Some of
Table 10.1 Some Adsorption HPLC Stationary Phases
Names Types
Partisil Silica, irregular

Hypersil Silica, regular
Chromasep PAA Alumina, irregular
Spherisorb A-Y Alumina, regular
Corasil I and II Silica, pellicular
Pellumina HS Alumina, pellicular
Zipax Silica, pellicular
these and other examples of reverse-phase stationary phases are Partisil

ODS-2, uBondapak C18, Spherisorb ODS, uBondapak Phenyl, Hyposil-
SAS, and Nucleosil C-8 . Common mobile phase liquids are water,
methanol, acetonitrile (CH3 CN), and acetic acid buffered solutions.
10.4.3 Adsorption Columns
Adsorption HPLC is the classification in which the highly polar silica

particles are exposed (no adsorbed or bonded liquid phase). Aluminum
oxide particles fit this description too and are also readily available as the
stationary phase. As mentioned earlier, this classification can also be
thought of as normal-phase chromatography, but LSC rather than LLC.
Typical normal-phase mobile phases (nonpolar) are used here. The sta
tionary phase particles can be irregular, regular, or "pellicular" in which
a solid core, such as a glass bead, is used to support a solid porous
material. Examples of this classification are shown in Table 10.1.
10.4.4 Ion-Exchange and Size-Exclusion Columns
As discussed in Chapter 8 , ion-exchange stationary phases consist of

solid resin particles which have positive and/or negative ionic bonding
sites on their surfaces at which ions are exchanged with the mobile phase
(see Chapter 8 , Figure 8.10). Cation exchange resins have negative sites
so that cations are exchanged, while anion exchange resins have positive
sites at which anions are exchanged. Typical examples are given in Table
10.2. A popular modem name for HPLC ion-exchange is simply "Ion
Table 10.2 Some Ion-Exchange Chromatography Stationary Phases
Names Type
Ion-X-SC Cation
Partisil 10 SCX Cation
Amberlite IR-120 Cation
Dowex 50W Cation
Ion-X-SA Anion
Partisil 10 SAX Anion
Amberlite IRA-400 Anion
Dowex 1 Anion
Chromatography." Detection of ions eluting from the HPLC column has

posed special problems which are described in Section 10.6. The mobile
phase for ion chromatography is always a pH-buffered water solution.
Size-exclusion columns, as discussed in Chapter 8 , separate mixture
components on the basis of size by the interaction of the molecules with
various pore sizes on the surfaces of porous polymeric particles. Size-
exclusion chromatography is subdivided into two classifications, gel-
permeation chromatography (GPC) and gel-filtration chromatography
(GFC). GPC utilizes nonpolar organic mobile phases, such as
tetrahydrofuran (THF), trichlorobenzene, toluene, and chloroform, to
analyze for organic polymers such as polystyrene. GFC utilizes mobile
phases that are water-based solutions and is used to analyze for naturally
occurring polymers, such as proteins and nucleic acids. GPC stationary
phases are rigid gels, such as silica gel, whereas GFC stationary phases
are soft gels, such as Sephadex. Neither technique utilizes gradient elution
because the stationary phase pore sizes are sensitive to mobile phase
changes.
10.4.5 Column Selection
Since each type of HPLC just discussed utilizes a different separation

mechanism, the selection of a specific column packing (stationary phase)
depends on whether or not the planned separation is possible or logical
with a given mechanism. For example, if a given mixture consists of
different molecules all of approximately the same size, then size-exclu
sion chromatography will not work. If a mixture consists only of ions,
then ion chromatography is the logical choice. While the conclusions
Table 10.3 Summary of Applications of the Different Types of HPLC
Type Useful for Components which
Normal and reverse-phase Have a low formula weight (<2000)

Are nonionic
Are either polar or nonpolar
Are water or organic soluble
Adsorption Have a low formula weight (<2000)
Are nonpolar
Are organic soluble
Ion-exchange Have a low formula weight (<2000)
Are ionic
Are water soluble
Size-exclusion Have a high or low formula weight
Are nonionic
Are water or organic soluble
drawn from these examples are obvious, others are less obvious and
require a study of the variables and the mechanisms in order to be able
to logically choose a particular stationary phase.
Table 10.3 presents some guidelines about each choice which would
be helpful in deciding which to use. While these guidelines may prove
helpful as a starting point, additional facts about the planned separation
need to be determined in order to select the most appropriate
chromatographic system, including facts that can only be discovered
through experimentation, or by searching the chemical literature. Several
different mobile phase/stationary phase systems may work. Comparing
reverse-phase with normal-phase, for example, one can see that there
would only be a reversal in the order of elution. Polar components would
elute first with reverse-phase, whereas nonpolar components would elute
first with normal-phase. Experimenting with various mobile phase com
positions, which may include a mixture of two or three solvents in various
ratios, would be a logical starting point. Some considerations which
would involve such experimentation are
1. The mixture components should have a relatively high affinity for

the stationary phase compared to the mobile phase. This would
mean longer retention times and thus probably better resolution.
2. The various separation parameters should be adjusted to provide
optimum resolution. These include mobile phase flow rate, stationary
phase particle size, gradient elution, and column temperature (using
an optional column oven).
3. Use partition chrom atography for highly polar mixtures and adsorp
tion chrom atography for very nonpolar mixtures.
10.5 THE CHROMATOGRAM
As with GC, the chart recording, which presents the written record
of the separation, is called the chromatogram. Please refer to Chapter 9,
Section 9.6 for a brief related discussion and for definitions of retention
time (tR), adjusted retention time (t'R), resolution (R), the number of
theoretical plates (N), and the height equivalent to a theoretical plate (H).
In the HPLC definition of t'R, the reference substance is not air, but the
sample solvent, which usually gives a slight perturbation to the baseline
at a very short retention time as it emerges from the column.
In addition to these parameters, liquid chromatographers are also
concerned with the capacity factor, k', and selectivity, a. The capacity
factor is the adjusted retention time divided by the retention time of the
solvent, t0.
k' = t'R/ t 0 (10.1)
The capacity factor is a measure of the retention of a component per

column volume, since the retention time is referred to the time for the
unretained solvent. The greater the capacity factor, the longer that
component is retained and the better the chances for good resolution.
This, however, must be weighed against the high speed advantage of
HPLC. While a large capacity factor is desirable, the experiment should
be completed within about 15 min. An optimum range for k' values is
between 2 and 6 .
Selectivity, a, is defined as the ratio of the adjusted retention time for
one component to the adjusted retention time for another
<x = t'R(A )/t'R(B) (10.2)
and is a measure of the "quality" of a separation. Selectivity values greater

than about 1.2 are considered good. A selectivity equal to 1 would mean
that the two retention times are equal, which means no separation at all.
10.6 DETECTORS
The function of the HPLC detector is of course to examine the solution

that elutes from the column and output an electronic signal proportional
to the concentrations of individual components present there. In Chapter
9 , we discussed a number of detector designs that serve this same purpose
for GC. The design of the HPLC detectors, however, are more
"conventional" in the sense that components present in a liquid solution
can be determined with conventional instrum ents, such as a
spectrophotometer. Thus, spectrophotometric and fluorometric detectors
are common. Let us discuss some of the more popular HPLC detectors
individually.
10.6.1 UV Absorption
The UV absorption HPLC detector is basically a UV spectrophotometer

that is capable of measuring a flowing solution rather than a static solution.
It has a light source, a wavelength selector, and a phototube as does an
ordinary spectrophotometer. The sample compartment, however, is
equipped with a "flow cell" through which the column effluent flows,
and the absorbance is monitored continuously (see Figure 10.6). The
output of the phototube is sent to the recorder or integrator where the
absorbance is continuously displayed with time. Peaks are recorded as
the UV absorbing components elute from the column.
The monochromator can be one of two different designs, either a
simple light filter (a so-called fixed wavelength detector) or a full-blown
slit/dispersing element/slit monochromator (a variable wavelength
detector), which has a control on the face of the detector for dialing in
the wavelength as in a standard spectrophotometer. The fixed wavelength
detector can be made to be variable in the sense that the light filter can
be changed, but one cannot tune to the wavelength of maximum
absorbance, and thus some sensitivity can be lost. The fixed wavelength
version, however, is less expensive and is fine for many applications for
which UV absorbance detection is appropriate. Filters for 254 and 280 nm
are common.
While such a detector is fairly sensitive, it is not universally applicable.
FIGURE 10.6 A schematic diagram of a UV absorbance detector. (From Kenkel, J.,

The mixture components being measured must absorb light in the UV

region and, at least in the fixed wavelength design, they must absorb at
the wavelength used for a particular experiment in order for a peak to
appear on the recorder. Also, the mobile phase must not absorb an
appreciable amount at the selected wavelength.
10.6.2 Diode Array
A diode array UV detector is a "multichannel" detector in which the

light beam from the UV source is not dispersed into its component
wavelengths until after it has passed through the flow cell. The dispersed
light then sprays across an array of photodiodes, each of which detects
only a narrow wavelength band. With the help of a computer, the entire
UV absorption spectrum can be immediately measured as each individual
component elutes. With computer banks containing a library of UV
absorption spectral information, a rapid, definitive qualitative analysis is
possible in a manner similar to GCMS or GCIR (Chapter 9). In addition,
the peak displayed on the recorder/integrator can be the result of a rapid
changeover of the wavelength by the computer. Thus, the peaks dis
played can represent the maximum possible sensitivity for each compo
nent. Finally, a diode array detector can be used to "clean up" a chro
matogram so as to only display the peaks of interest. This is possible since
we can rapidly change the wavelength giving rise to the peaks.
10.6.3 Fluorescence
The basic theory, principles, sensitivity, and application of fluorescence

spectrometry (fluorometry) were discussed in Chapter 6 . Like the UV
absorption detector described above, the HPLC fluorescence detector is
based on the design and application of its parent instrument, in this case
the fluorometer, and you are referred to Chapter 6 for a review of the
fundamentals of the fluorescence technique.
In summary, the basic fluorometer, and thus the basic fluorescence
detector, consists of a light source and a monochromator (usually a filter)
for creating and isolating a desired wavelength, a sample "compart
ment," and a second monochromator (another filter) with a phototube
detector for isolating and measuring the fluorescence wavelength. The
second monochromator and detector are lined up perpendicular to the
light beam from the source (so-called "right angle" configuration).
As with the UV absorption detector, the sample compartment consists
of a special cell for measuring a flowing, rather than static, solution. The
fluorescence detector thus individually measures the fluorescence
intensities of the mixture components as they elute from the column (see
Figure 10.7). The electronic signal generated at the phototube is sent to
the recorder or integrator where a peak is recorded each time a fluorescing
species elutes.
The advantages and disadvantages of the fluorometry technique in
general hold true here. The fluorescence detector is not universal (it will
give a peak only for fluorescing species), but it is thus very selective
(almost no possibility for interference) and very sensitive.
10.6.4 Refractive Index
The refractive index of a liquid or liquid solution is defined as the ratio

of the speed of light in a vacuum to the speed of light in the liquid.
n = c v a c //c ..liq (10.3)

Column effluent
FIGURE 10.7 An illustration of a fluorescence detector.
Since the speed of light in any material medium is less than the speed
of light in a vacuum, the numerical value of the refractive index for any
liquid is greater than 1 .
An instrument known as a refractometer has been invented and used
for many years to measure the refractive index of liquids and liquid
solutions for the purpose of both quantitative and qualitative analysis.
A refractometer measures the degree of refraction (or "bending") of a
light beam passing through a thin film of the liquid. This refraction occurs
when the speed of light is different from a reference liquid or air. The
refractometer measures the position of the light beam relative to the
reference and is calibrated directly in refractive index values. It is rare
for any two liquids to have the same refractive index, and thus this
instrument has been used successfully for qualitative analyses.
The refractive index detector in HPLC is a modification of this basic
instrument and actually can be purchased in two different designs,
depending on the manufacturer. In probably the most popular design,
both the column effluent and the pure mobile phase (acting as a reference)
pass through adjacent flow cells in the detector. A light beam, passing
through both cells, is focused onto a photosensitive surface, and the
location of the beam when both cells contain pure mobile phase is taken
as the reference point and the recorder pen is zeroed. When a mixture
component elutes, the refractive index in one cell changes, and the light
beam is "bent" and becomes focused onto a different point on the
photosensitive surface, causing the recorder pen to deflect and trace a
peak (see Figure 10.8).
The major advantage of this detector is that it is almost universal. All
Column effluent
Pure mobile phase
Light :
source ^ Mirror
Indicates light beam

when column effluent
Photosensitive is pure mobile phase
surface
Indicates light beam
when a mixture component
is present in column
effluent
FIGURE 10.8 A representation of a retractive index detector (see text for description).
substances have their own characteristic refractive index (it is a physical

property of the substance). Thus, the only time that a mixture component
would not give a peak is when it has a refractive index equal to that of
the mobile phase, a rare occurrence. The disadvantages are that it is not
very sensitive and the output to the recorder is subject to temperature
effects. Also, it is difficult to use this detector with the gradient elution
method because it is sensitive to changes in the mobile phase composition.
10.6.5 Electrochemical
Various detectors which utilize electrical current or conductivity

measurements for detecting eluting mixture components have been
invented. These are called electrochemical detectors. Let us now examine
some of the basic designs.
10.6.5a Conductivity
Perhaps the most important of all electrochemical detection schemes

currently in use is the electrical conductivity detector. This detector is
specifically useful for ion-exchange, or ion chromatography, in which the
analyte is in ionic form. Such ions elute from the column and need to be
detected as peaks on the recorder trace.
A well-known fact of fundamental solution science is that the presence

of ions in any solution gives the solution a low electrical resistance and
the ability to conduct an electrical current. The absence of ions means that
the solution would not be conductive. Thus, solutions of ionic compounds
and acids, especially strong acids, have a low electrical resistance and are
conductive. This means that if a pair of conductive surfaces is immersed
into the solution and connected to an electrical power source, such as a
simple battery, a current can be detected flowing in the circuit.
Alternatively, if the resistance of the solution between the electrodes is
measured (with an ohmmeter), it would be low. Conductivity cells based
on this simple design are in common use to determine the quality of
deionized water, for example. Deionized water should have no ions
dissolved in it and thus should have a very low conductivity. The
conductivity detector is based on this simple apparatus.
For many years, the concept of the conductivity detector could not
work, however. Ion chromatography experiments utilize solutions of
high ion concentrations as the mobile phase. Thus, changes in conductivity
due to eluting ions are not detectable above the already high conductivity
of the mobile phase. This was true until the invention of so-called ion
"suppressors." Today, conductivity detectors are used extensively in
HPLC ion chromatography instruments which also include suppressors.
A suppressor is a short column (tube) that is inserted into the flow
stream just after the analytical column. It is packed with an ion-exchange
resin itself; a resin that removes mobile phase ions from the effluent, much
like a deionizing cartridge removes the ions in laboratory tap water, and
replaces them with molecular species. A popular "mixed bed" ion-ex
change resin is used, for example, in deionizing cartridges, such that tap
water ions are exchanged for H+ and OH- ions, which in turn react to
form water. The resulting water is thus deionized. Of course, in the HPLC
experiment, the analyte ions must not be removed in this process, and
thus suppressors must be selective only for the mobile phase ions.
A typical design for a conductivity detector uses electrically isolated
inlet and outlet tubes as the electrodes. This design is shown schematically
in Figure 10.9.
10.6.5b Amperometric
A thorough discussion of electroanalytical techniques, including "po-

larography," "voltammetry," and "amperometry" is given in Chapter 11.
FIGURE 10.9 A drawing of a conductivity detector in which the inlet and outlet
tubes are the electrodes.
An understanding of these would be useful for understanding the

amperometric HPLC detector.
Electrochemical oxidation and/or reduction of eluting mixture
components is the basis for amperometric electrochemical detectors. The
three electrodes needed for the detection, the working ("indicator")
electrode, reference electrode, and auxiliary electrode, are either inserted
into the flow stream or imbedded in the wall of the flow stream (see
Figure 10.10). The indicator electrode is typically glassy carbon, platinum
or gold; the reference electrode is a silver/silver chloride electrode; and
the auxiliary is a stainless steel electrode. Most often, the indicator electrode
is polarized so as to cause oxidation of the mixture components as they
elute. The oxidation current is then measured and constitutes the signal
sent to the recorder/integrator.
Advantages of this detector include broad applicability to both ionic
mixture components as well as molecular components, as long as they are
able to be oxidized (or reduced) at fairly small voltage polarizations. Selec
tivity* can be improved by varying the potential. In addition, the sensitivity*
experienced with this detector is quite good — generally better than the
UV detector, but not as good as the fluorescence detector. A disadvantage
is that the indicator electrode can become fouled due to products of the
electrochemical reaction coating the electrode surface. Thus, this detector
must be able to be disassembled and cleaned with relative ease, since this
may need to be done frequently.
See Section 9.7 for definitions as related to chromatography detectors.

— '-■■■» To waste
Auxiliary
electrode
Reference electrode
Indicator (working)
electrode
Column effluent ■
FIGURE 10.10 A drawing of an amperometric HPLC detector.
10.6.6 LC-MS and LC-IR
In Chapter 9, the use of mass spectrometry and FTIR for GC detection

was discussed. Details of these techniques were individually given in
Chapter 6 . Much of the discussion presented in Chapter 9 is applicable
here. Both mass spectrometry and infrared spectrometry have been adapted
to HPLC detection in recent years.
FTIR is a "natural" for HPLC in that it is a technique that has been
used mostly for liquids. The speed introduced by the Fourier Transform
technique allows, as was mentioned for GC, the recording of the complete
IR spectrum of mixture components as they elute, thus allowing the IR
"photograph" to be taken and interpreted for qualitative analysis. Of
course, the mobile phase, and its accompanying absorptions, is ever
present in such a technique and water must be absent if the NaCl windows
are used, but IR holds great potential, at least for nonaqueous systems,
as a detector for HPLC in the future.
The mass spectrometer is also incompatible with the HPLC system, but
for a different reason. The ordinary mass spectrometer operates under
very low pressure (a high vacuum — see Chapter 6 ), and thus the liquid
detection path must rapidly convert from a very high pressure and large
liquid volume to a very low pressure and a gaseous state. Several
approaches to this problem have been used, but probably the most
popular is the "thermospray" (TS) technique. In this technique, the column
effluent is converted to a fine mist (spray) as it passes through a small-
diameter heated nozzle. The analyte molecules, which must be thermally

stable, are preionized with the presence of a dissolved salt. A portion of
the spray is introduced into the mass spectrometer. The analyte and
mobile phase must be polar if the TS technique is used because the mobile
phase must dissolve the required salt and the components must interact
with the analyte molecule.
10.7 QUALITATIVE AND QUANTITATIVE ANALYSIS
Qualitative and quantitative analysis with HPLC are very similar to

that with GC (Chapter 9, Sections 9.8 and 9.9). In the absence of diode
array, mass spectrometric, and FTIR detectors which give additional
identification information, qualitative analysis depends solely on retention
time data, tR and t'R. (Remember that t'R here is the time from when the
solvent front is evident to the peak). Under a given set of HPLC conditions
— namely the mobile and stationary phase compositions, the mobile
phase flow rate, the column length, temperature (when the optional
column oven is used), and instrument dead volume — the retention time
is a particular value for each component. It changes only when one of
the above parameters changes. Refer to Section 9.8 for further discussion
of qualitative analysis.
Peak size measurement and quantitation methods outlined for GC in
Section 9.9 are also applicable here. The reproducibility of the amount
injected is not nearly the problem with HPLC as it is with GC. Roughly
ten times more sample is typically injected (5-20 pL), and there is no loss
during the injection since the sample is not loaded into a higher pressure
system through a septum. In addition, the sample loop is manufactured
to have a particular volume and is often the means by which a consistent
amount is injected, which means reproducibility is maximized through
the consistent "overfilling" of the loop via the injection syringe. In this
way, the loop is assured of being filled at each injection, and a reproduc
ible volume is always introduced. Sometimes, however, the analyst chooses
to inject varying volumes of a single standard to generate the standard
curve (Chapter 5) rather than equal volumes of a series of standard
solutions. In this case, the injection syringe is used to measure the volumes
— a less accurate method, but better than an identical method with GC,
since the sample volume is larger and there is less chance for sample loss.
With this type of quantitation, the standard curve is a plot of peak size
vs amount injected, rather than concentration.
The most popular quantitation "method," then, is the series of standard
solutions method with no internal standard (i.e., "serial dilution" — see
Chapter 5) or the variable injection of a single standard solution as
outlined above.
1 0 .8 TROUBLESHOOTING
Problems that arise with HPLC experiments are usually associated

with abnormally high or low pressures, system leaks, worn injector parts,
air bubbles, and/or blocked in-line filters. Sometimes these manifest
themselves on the chromatogram and sometimes they do not. In the
following paragraphs, we will address some of the most common problems
encountered, pinpoint possible causes, and suggest methods of solving
the problems.*
Unusually High Pressure — A common cause of unusually high
pressure is a plugged in-line filter. In-line filters are found at the very
beginning of the flow line in the mobile phase reservoir, immediately
before and/or after the injector and just ahead of the column. With time,
they can become plugged due to particles that are filtered out (particles
can appear in the mobile phase and sample even if they were filtered
ahead of time), and thus the pressure required to sustain a given flow
rate can become quite high. The solution to this problem is to backflush
the filters with solvent and/or clean them with a nitric acid solution in
an ultrasonic bath. Other causes of unusually high pressure are an injector
blockage, mismatched mobile and stationary phases, and a flow rate that
is simply too high. An injector that is left in a position between "LOAD"
and "INJECT" can also cause a high pressure, since the pump is pumping,
but their can be now flow.
Unusually Low Pressure — A sustained flow that is accompanied by
low pressure may be indicative of a leak in the system. All joints should
be checked for leaks (see next paragraph).
System Leaks — Leaks can occur within the pump, at the injector, at
various fittings and joints, such as at the column, and in the detector.
See also the "LC Troubleshooting" column published regularly in the monthly journal
GC*LC, A Magazine of Separation Science., Aster Publishing Corporation, Eugene, OR.
Leaks within the pump can be due to failure of pump seals and diaphragms
and loose fittings, such as at the check valves, etc. Leaky fittings should
be checked for mismatched or stripped ferrules and threads, or perhaps
they simply need tightening. Leaks in the injector can be due to a plugged
internal line, or other system blockage, gasket failure, loose connections,
or use of the wrong size syringe if the leak occurs as the sample is loaded.
Detector leaks are most often due to a bad gasket seal or a broken flow
cell. Of course, loose or damaged fittings and a blockage in the flow line
beyond the detector are possible causes.
Air Bubbles— An air pocket in the pump can cause low or no pressure
or flow, erratic pressure, and changes in retention time data. It may be
necessary to bleed air from the pump or prime the pump according to
system start-up procedures. Air pockets in the column will mean decreased
contact with the stationary phase and thus shorter retention times and
decreased resolution. Tailing and peak splitting on the chromatogram
may also occur due to air in the column. Air bubbles in the detector flow
cell are usually manifested on the chromatogram as small spikes due to
the periodic interruption of the light beam (e.g., in a UV absorbance
detector). Increasing the flow rate, or restricting and then releasing the
post-detector flow, so as to increase the pressure, should cause such
bubbles to be "blown" out.
Column "Channeling" — If the column packing becomes separated
and a channel is formed in the stationary phase, the tailing and splitting
of peaks will be observed on the chromatogram. In this case, the column
needs to be replaced.
Recorder Zero vs Detector Zero — Both the detector and the recorder
have "zeroing" capability. The recorder zero control is used when there
is zero recorder input, such as when the input terminals are shorted. The
detector zero control is used to zero the recorder pen when only the
mobile phase is eluting. The two can be "mismatched." This problem is
obvious on the chromatogram when changing the attenuation setting on
the detector. At a less sensitive attenuation, the detector output may
appear to be zero, but when the attenuation is changed to a more sensitive
setting, there may be a sudden jump in the baseline, indicating that the
detector output is actually not zero since the more sensitive setting is able
to show a small offset from zero. The solution to this problem is to initially
zero the detector at the more sensitive setting.
Decreased Retention Time — When retention times of mixture com
ponents decrease, there may be problems with either the mobile or
stationary phases. It may be that the mobile phase composition was not
restored after a gradient elution, or it may be that the stationary phase

was altered due to irreversed adsorption of mixture components, or
simply chemical decomposition. Use of guard columns (see previous
discussion) may avoid stationary phase problems.
Baseline Drift — A common cause of baseline drift is a slow elution
of adsorbed substances on the column. A column clean-up procedure
may be in order, or it may need to be replaced. This problem may also
be caused by temperature effects in the detector. Refractive index detectors
are especially vulnerable to this. In addition, a contaminated detector can
cause drift. The solution here may be to disassemble and clean the detector.
You are also referred to the troubleshooting guide in Chapter 9 (GC)
for possible solutions to problems.
C hapter 11
E lectroanalytical M ethods
11.1 INTRODUCTION
The subject of electroanalytical chemistry encompasses all analytical

techniques which are based on electrode potential and current
measurements at the surfaces of electrodes immersed in the solution
tested. Either an electrical current flowing between a pair of immersed
electrodes or an electrical potential developed between a pair of immersed
electrodes is measured and related to the concentration of some dissolved
species.
Electroanalytical techniques are an extension of classical oxidation-
reduction chemistry, and indeed oxidation and reduction processes occur
at or within the two electrodes, oxidation at one and reduction at the
other. Electrons are consumed by the reduction process at one electrode
and generated by the oxidation process at the other. The complete system
is often called a "cell," the individual electrodes "half-cells," and the
individual oxidation and reduction reactions are the "half-reactions."
Electrons flow on a conductor between the half-cells, and this flow
constitutes the electrical current that is often measured. A "galvanic" cell
is one in which this current flows spontaneously because of the strong
tendency for the chemical species involved to give and take electrons. A
287
Recharging unit
• Wall socket
for recharger
With external poles Current forced to flow

+ - connected, current in opposite direction
flows spontaneously (electrolytic cell)
9V (galvanic cell)
k Battery Rechargeable battery
(a) (b)
FIGURE 1 1 .1 (a) A battery with its negative and positive poles connected is a
galvanic cell, (b) A rechargeable battery, when positioned in its
recharging unit, is an example of an electrolytic cell.
battery that has its positive and negative poles externally connected is
an example of a such a cell. An "electrolytic" cell is one in which the
current is not a spontaneous current, but rather is the result of connecting
an external power source, such as a battery, to the system. A rechargeable
battery, when it is positioned in the recharging unit, would be an example
of such a cell (see Figure 1 1 .1 ). Electroanalytical techniques utilize both
general types of cells.
11.2 POTENTIOMETRY
Electroanalytical techniques which measure or monitor electrode

potential utilize the galvanic cell concept. Such techniques fall under the
general heading of "potentiom etry." Examples include the pH
measurement, ion-selective electrode measurement, and potentiometric
titrations. In these techniques, a pair of electrodes is immersed, and the
potential, or voltage, of one of the electrodes is measured, hence the name
potentiometry. To understand how and why these techniques work, a
fundamental knowledge of the Nemst Equation is needed.
Electroanalytical Methods 289
11.2.1 The Nemst Equation
All oxidation and reduction half-reactions have a certain tendency to

occur. Relative tendencies are often listed in text and reference books in
the form of a table of "standard reduction potentials," symbolized E° and
having the units of volts. If the tendency of a particular species to be
reduced is high, then it will have a positive E°. For example, the reduction
of fluorine, F2 (to F~), which is perhaps the species with the strongest of
all tendencies to reduce, has an E° of +2.87 V. This represents what may
be the upper limit to the E° values. If the tendency of a particular species
to reduce is low, and in fact is more likely to be formed when another
species is oxidized, then its E° value will be negative. Such is the case,
for example, with all ions of alkali metals and alkaline earth metals. The
E° value of lithium ion, Li+ 1 (being reduced to Li metal), is perhaps the
most negative of all, -3.05. Thus the E° values for typical chemical species
range from about -3 to about +3 V. These values are "relative" values,
not absolute. This means that they are a measure of the relative tendency
of each reaction to occur. They are based on the hydrogen ion reduction
to hydrogen gas. This reaction is assigned an E° value of 0.00 V. Table
11.1 lists some half-reactions and accompanying E° values.
It is possible to determine the tendency, or E value, for an overall
oxidation-reduction reaction. This is done by adding together the E° value
for the reduction half-reaction and the (~)E° value for the oxidation half
reaction. The result is the "overall" oxidation-reduction reaction tendency,
symbolized E°cell. If the sign of this E° is positive, the reaction will proceed
spontaneously to the right as written. If it is negative, it will proceed to
the left.
Standard reduction and overall potentials are based on standard
conditions of temperature (25°C), concentration of dissolved ionic species
(1 M), and, when gases are involved, pressure (1 atm). If there is a
deviation from these conditions, then the actual reduction and overall
potential will be different from the standard. The relationship between
the E value and these parameters is given by the Nemst Equation. For
the general half-reaction
qQr + ne -+ qQr n (11.1)

Table 11.1 Standard Reduction Potentials for Selected Half-Reactions
Half-Reactions E° (volts)
F2 + 2e- -» 2F- +2.87

H20 2 + 2H+ + 2e" -> 2H20 +2.07
M n04" + 8H+ + 5e" -> Mn+2 + 4H20 +1.49
Ce+4 + le~ -+ Ce+3 +1.44
Cl2 + 2e“ -> 2CT +1.36
Cr20 7' 2 + 14H+ + 6e- -+ 2Cr+3 + 7H20 +1.33
0 2 + 4H+ + 4e- -> 2H20 +1.23
Hg+2 + 2e- -> Hg +0.85
Ag+ + le" -+ Ag +0.80
Fe+3 + le~ -+ Fe+2 +0.77
I2 + 2e‘ -» 21- +0.54
Cu+2 + 2e" -> Cu +0.34
Hg2Cl2 + 2e- -» 2Hg + 2C1" (SCE) +0.24
AgCl + le" -» Ag + Cl" (Ag/AgCl Ref.) +0.22
2H+ + 2e- -» H2 0.00
Fe+3 + 3e" -+ Fe -0.04
Fe+2 + 2e" -> Fe -0.41
Cr+3 + 3e" -» Cr -0.74
Zn+2 + 2e- -» Zn -0.76
Mg+2 + 2e" —> Mg -2.38
Na+ + le- -» Na -2.71
K+ + le“ -> K -2.92
Li+ + le~ -» Li -3.05
Reprinted with permission from the CRC Handbook of Chemistry and Physics,
Weast, R., Ed., CRC Press, Inc., Boca Raton, FL.
the Nemst Equation* is
„ „ 0.059, [ Q 'i’
g_i ^ f <n'2)
and for the general overall reaction
aA + bB -» cC + dD (113)
Strictly speaking, the Nernst Equation involves "activity" rather than concentration. Activity
is directly proportional to concentration — the "activity coefficient" is the proportional
constant. For most applications, the activity coefficient is equal to 1 and thus the activity
equals the concentration. Further discussion of activity and activity coefficient is beyond
the scope of this book.
the Nernst Equation is
F _ r° 0 059 w [C]°[P]d ni /n
cell ceU n [A ]a[B ]b (1 L 4 )
If any species involved is a gas, the partial pressure of the gas is

substituted for the concentration. If the temperature is different from
25°C, the "constant" 0.059 changes. The symbol "n" represents the num
ber of electrons involved.
It is obvious from the foregoing discussion that the potential of elec
trodes and half-cells that one would actually measure is dependent on
the concentration of the dissolved species involved. This is the basis for
all quantitative potentiometry techniques and measurements to be dis
cussed in this chapter.
11.2.2 Reference Electrodes
The measurement of any voltage is a relative measurement and re

quires an unchanging reference point. For voltage measurements in most
electronic circuitry, this reference is usually "ground," which is often a
wire that is connected to the frame of the electronic unit and the third
prong in an electrical outlet, which in turn is connected to a rod that is
pushed into the earth, hence the name "ground." Thus an electronics
technician measures voltages relative to ground.
In electroanalytical chemistry, the unchanging reference is a portable
half-cell that is designed to develop a potential that is constant. There is
more than one design for this half-cell, and we now proceed to describe
several that have become popular over the years.
11.2.2a The Saturated Calomel Reference Electrode (SCE)
The saturated calomel reference electrode is one such constant-potential

electrode. A typical SCE available commercially is shown in Figure 11.2.
It consists of two concentric glass tubes, which we will refer to here as
an outer tube and an inner tube, each isolated from the other except for
a small opening for electrical contact. The outer tube has a porous fiber
FIGURE 1 1 .2 The saturated calomel reference electrode. (From Kenkel, ]., Analytical
With permission.)
plug in the tip, which acts as the "salt bridge" to the analyte solution.
A saturated solution of potassium chloride is in the outer tube. The
saturation is evidenced by the fact that there is some undissolved KC1
present. Within the inner tube is a paste-like material known as calomel.
Calomel is made by thoroughly mixing mercury metal (Hg) with
mercurous chloride (Hg2 Cl2), a white solid. When in use, the following
half-cell reaction occurs:
Hg 2 Cl 2 + 2e~ -> 2Hg + 2C1“
The Nemst Equation for this reaction is:
E = E» - « j p log[c i-f
or
E = E° -0 .0 5 9 lo g [c r]
Obviously the only parameter on which the potential depends is

[Cl-]. The saturated KC1 present provides the [Cl-] for the reaction and,
since it is a saturated solution, [Cl-] is a constant at a given temperature
represented by the solubility of KC1 at that temperature. If [Cl-] is constant,
the potential of this half-cell, dependent only on the [Cl-], is also a
constant. As long as the KC1 is kept saturated and the temperature kept
constant, the SCE is useful as a reference against which all other potential
measurements can be made. Its standard reduction potential at 25°C (see
Table 11.1) is +0.24 V. An advantage of this electrode is that, unlike its
unsaturated counterparts, the [Cl-] does not change with evaporation of
the water, since the solution remains saturated. Unsaturated calomel
electrodes, however, are not affected by temperature changes like the
SCE.
The SCE is usually used by dipping it into the analyte solution which
is part of another half-cell, including an electrode typically referred to as
an "indicator" electrode. A voltmeter is then externally connected across
the leads to the two electrodes and the potential of the indicator electrode
recorded vs the constant reference. While the SCE is dipped into the
solution, there will be a slight leakage of the potassium and chloride ions
through the porous tip. In order for the SCE to be used accurately, the
experiment must not be adversely affected by the slight contamination
from these ions. It is a good idea to slide the movable sleeve (Figure 11.2)
downward so that the outer tube is vented while the electrode is in use
so that the ions do indeed freely diffuse through the porous tip. Also,
the electrode, under these circumstances, should not be immersed into
the solution so deep that the level of solution in the external tube is lower
than the level of the solution tested. This would cause the solution to
diffuse into the SCE rather than the reverse, and thus would contaminate
the KC1 solution inside and possibly damage the SCE.
The vent hole may also be used prior to the experiment to refill the
outer tube with more saturated KC1 solution as this solution is lost with
time. In addition, if the undissolved KC1 disappears, more solid KC1 can
be added through the vent hole. The SCE should be stored in a beaker
of distilled water (with the vent hole covered) for short-term storage and
in a removable plastic cover usually provided with the electrode for long
term storage.
&
I Movable sleeve
Vent
Ag wire
—Saturated AgCI/KCI
AgCI "paste"
Porous fiber plug
FIGURE 11.3 A typical silver-silver chloride reference electode.
11.2.2b The Silver-Silver Chloride Electrode
The commercial silver-silver chloride electrode is similar to the SCE

in that it is enclosed in glass, has nearly the same size and shape, and
has a porous fiber tip for contact with the external solution. Internally,
however, it is different. There is only one glass tube and a solution
saturated in silver chloride and potassium chloride is inside. A silver wire
coated at the end with a silver chloride "paste" extends into this solution
from the external lead (see Figure 11.3). The half-reaction which occurs
is
AgCl(s) + le -+ Ag(s) + Cl
and the Nemst Equation for this is
The standard reduction potential for this half-reaction is +0.22 V. The

* H2 out
Hg in-
S\
-platinum
-porous glass
FIGURE 11.4 The standard hydrogen electrode. (From Kenkel, J., Analytical Chem
istry for Technicians, Lewis Publishers, Inc., Chelsea, MI, 1988. With
permission.)
potential is only dependent on the [Cl-], as was the SCE, and once again
the [Cl-] is constant because the solution is saturated. Thus this electrode
is also appropriate for use as a reference electrode. Refer to the segment
above on the SCE for a discussion as to how it is physically handled while
in use and in storage.
11.2.2c The Standard Hydrogen Electrode
The ultimate reference electrode utilizes the half-reaction on which

Table 11.1 is based:
2H + + 2 e - —» H2 E° =O . OO V
This half-cell consists of a glass envelope housing a solution of constant

pH in which a platinum wire or foil is immersed (see Figure 11.4). The
platinum provides a surface for the exchange of electrons. Hydrogen gas
bubbles through this solution at a constant pressure from some external
source. Thus, since [H+] is a constant, and since the partial pressure of
H2 is a constant, the potential is a constant. The use of this cell as a
reference electrode, however, is not very practical. It is cumbersome to

use because it is not portable and requires a venting system for the
hydrogen which otherwise is released into the laboratory.
11.2.3 Indicator Electrodes
As stated previously, the reference electrode represents half of the

complete system for potentiometric measurements. The other half is the
half at which the potential of analytical importance— the potential which
is related to the concentration of the analyte — develops. There are a
number of such "indicator" electrodes and analytical experiments which
are of importance. Let us discuss these.
11.2.3a The pH Electrode
The measurement of pH is very important in many aspects of chemical

analysis. Curiously, the measurement is based on the potential of a half
cell, the pH electrode.
The pH electrode, also called the glass electrode, consists of a closed-
end glass tube that has a very thin fragile glass membrane at the tip. Inside
the tube is a saturated solution of silver chloride that has a particular pH.
It is typically a l M solution of HC1. A silver wire coated with silver
chloride is dipped into this solution to just inside the thin membrane.
While this is almost the same design as the silver-silver chloride reference
electrode, the presence of the HC1 and the fact that the tip is fragile glass
and does not have a porous fiber plug points out the difference (see Figure
11.5).
The purpose of the silver-silver chloride combination is to prevent the
potential that develops from changing due to possible changes in the
interior of the electrode. The potential that develops is a "membrane"
potential. Since the glass membrane at the tip is thin, a potential develops
due to the fact that the chemical composition inside is different from the
chemical composition outside. Specifically, it is the difference in the
concentration of the hydrogen ions on opposite sides of the membrane
that causes the potential — the membrane potential — to develop. There
^silver
membrane
FIGURE 11.5 The pH or "glass" electrode. (From Kenkel, J., Analytical Chemistry for
Technicians, Lewis Publishers, Inc., Chelsea, MI, 1988. With permission.)
is no half-cell reaction involved. The Nernst Equation is
or, since the internal [H+] is a constant, it can be lumped into E°, which
is also a constant giving a modified E°, E*, and eliminating [H+](internal):
E = E* + 0 .0 5 9 log[H+](extemal)
In addition, we can recognize that pH = -log[H+] and substitute this into

the above equation:
E = E* - 0 .0 5 9 pH
The beauty of this electrode is that the measured potential (measured

against a reference electrode) is thus directly proportional to the pH of
the solution into which it is dipped. A specially designed voltmeter, called
a pH meter, is used. In a pH meter, the meter readout is typically calibrated
in both pH units and volts making the pH meter a rather versatile device.
The pH meter is standardized (calibrated) with the use of buffer so

lutions. Usually, two buffer solutions are used for maximum accuracy.
The pH values for these solutions "bracket" the pH value expected for
the sample. For example, if the pH of a sample to be measured is expected
to be 9.0, buffers of pH = 7.0 and 10.0 are used. Buffers with pH values
of 4.0, 7.0, and 10.0 are available commercially specifically for pH meter
standardization. Alternatively, of course, homemade buffer solutions (see
Chapter 3) may be used. In either case, the meter is adjusted, using the
appropriate standardization method (refer to manufacturer's literature),
to read the pH of the buffer into which the electrodes (the pH electrode
and a reference electrode) are dipped. The solution of unknown pH is
then determined. The pH electrode should be stored in a solution of
distilled water or with a protective plastic sleeve over the tip.
11.2.3b The Combination pH Electrode
In order to use the pH electrode described above, two half-cells (some

times called "probes") are needed — the pH electrode itself and a ref
erence electrode, either the SCE or the silver-silver chloride electrode —
and two connections are made to the pH meter. A modern development
in this area is the invention of the "combination" pH electrode. This
electrode incorporates both the reference probe and pH probe into a
single probe. The reference portion is a silver-silver chloride reference.
A drawing of the combination pH electrode is given in Figure 11.6.
The pH portion of this electrode is found in the center of the probe
as shown. It is identical with the pH electrode described above — a silver
wire coated with silver chloride immersed in a solution saturated with
silver chloride and having a [H+] of 1.0 M. This solution is in contact with
a thin glass membrane at the tip. The reference portion is in an outer tube
concentric with the inner pH portion. It has a silver wire coated with silver
chloride in contact with a solution saturated with silver chloride and
potassium chloride. A porous fiber plug in the wall of the outer tube
connects the outer tube with the solution tested as shown. This electrode
either has two wires protruding from the lead (for connection to older
style pH meters) or makes both connections using the "bnc" type or
similar connector.
FIGURE 11.6 The combination pH electrode.
11.2.3c Ion-Selective Electrodes
The concept of the pH electrode has been extended to include other

ions as well. Considerable research has gone into the development of
these "ion-selective" electrodes over the years, especially in studying the
composition of the membrane that separates the internal solution from
the analyte solution. The internal solution must contain a constant
concentration of the ion analyzed for, as with the pH electrode. Today
we utilize electrodes with (1) glass membranes of varying compositions,
(2) crystalline membranes, (3) liquid membranes, and (4) gas-permeable
membranes. In each case, the interior of the electrode has a silver-silver
chloride wire immersed in a solution of the analyte ion.
Examples of electrodes which utilize a glass membrane are those for
lithium ions, sodium ions, potassium ions, and silver ions. Varying
percentages of A120 3, S i0 2, along with oxides of the metal analyte are
often found in the membrane as well as other metal oxides. Selectivity
and sensitivity of these electrodes vary.
With crystalline membranes, the membrane material is most often an
- Internal solution
of analyte ion
,A g/A gC I electrode
.Organic liquid
ion-exchanger
'Sir Porous membrane saturated with

liquid ion-exchanger
FIGURE 11.7 A liquid membrane electrode.
insoluble ionic crystal cut to a round flat shape and having a thickness
of 1 or 2 mm and a diameter of about 10 mm. This flat "disc" is mounted
into the end of a Teflon or PVC tube. The most important of the electrodes
with crystalline membranes is the fluoride electrode. The membrane
material for this electrode is lanthanum fluoride. The fluoride electrode
is capable of accurately sensing fluoride ion concentrations over a broad
range and to levels as low as 10-6 M. Other electrodes that utilize a
crystalline membrane but with less impressive success records are chloride,
bromide, iodide, cyanide, and sulfide electrodes. The main difficulty with
these is problems with interferences.
Liquid membrane electrodes utilize porous polymer materials, such as
PVC or other plastics. An organic liquid ion exchanger, immiscible with
water, contacts and saturates the membrane from a reservoir around the
outside of the tube containing the water solution of the analyte and the
silver-silver chloride wire (see Figure 11.7). Important electrodes with this
design are the calcium and nitrate ion-selective electrodes.
Finally, gas-permeable membranes are used in electrodes which are
useful for dissolved gases, such as ammonia, carbon dioxide, and hydrogen
cyanide. These membranes are permeated by the dissolved gases, but not
by solvents or ionic solutes. Inside the electrode is a solution containing
log [F ’)
FIGURE 11.8 A calibration curve and unknown determination for an experiment

using an ion-selective electrode.
the reference wire as well as a pH probe, the latter positioned so as to

create a thin liquid film between the glass membrane of the pH probe
and the gas-permeable membrane. As the gases diffuse in, the pH of the
solution constituting the thin film changes, and thus the response of the
pH electrode changes proportionally to the amount of gas diffusing in.
Calibration of ion-selective electrodes for use in quantitative analysis
is usually done by preparing a series of standards as in most other
instrumental analysis methods (see Chapter 5), since the measured potential
is proportional to the logarithm of the concentration. The relationship is
_ 0 . 0 5 9 , fT ,
E =E + log[IonJ
z
in which z is the signed charge on the ion. The analyst can measure the
potential of the electrode immersed in each of the standards and the
sample (vs the SCE or silver-silver chloride reference), plot E vs log[Ion],
and find the unknown concentration from the graph. A typical plot and
unknown determination is shown in Figure 11.8 for some real data using
a fluoride ion-selective electrode. Alternatively, the standard additions
technique (Chapter 5) may be used.
FIGURE 11.9 A potentiometric acid-base titration using a combination pH electrode.
11.2.4 Potentiometric Titrations
It is possible to monitor the course of a titration using potentiometric

measurements. The pH electrode, for example, is appropriate for
monitoring an acid-base titration and determining an end point in lieu
of an indicator. The procedure has been called a "potentiometric titration,"
and the experimental setup is shown in Figure 11.9. The end point occurs
when the measured pH undergoes a sharp change — when all the acid
or base in the titration vessel is reacted. The same procedure can be used
for any ion for which an ion-selective electrode has been fabricated and
for which there exists an appropriate titrant.
In addition, potentiometric titration methods exist in which an electrode
other than an ion-selective electrode is used. A simple platinum wire
surface can be used as the indicator electrode when an oxidation-reduction
reaction occurs in the titration vessel. An example is the reaction of Ce(IV)
with Fe(II):
C e+4 + Fe+2 Ce+3 + Fe+3
If this reaction were to be set up as a titration, with Ce+4 as the titrant,

the Fe+2 in the titration vessel, and the potential of a platinum electrode
dipped into the solution monitored (vs a reference electrode) as the titrant
is added, the potential would change with the volume of titrant added.
This is because as the titrant is added, the measured E would change as
the [Fe+2] is decreased, the [Fe+3] is increased, and the [Ce+3] is increased.
At the end point and beyond, all the Fe+2 is consumed and the [Fe+3] and
[Ce+3] change only by dilution, and thus the E is dependent mostly on
the change in the [Ce+4]. At the end point, there would be a sharp change
in the measured E.
Automatic titrators have been invented which are based on these
principles. A sharp change in a measured potential can be used as an
electrical signal used to activate a solenoid and stop a titration (see also
Chapter 12).
11.3 POLAROGRAPHY AND VOLTAMMETRY
11.3.1 Introduction
Techniques which utilize the electrolytic cell concept, rather than the
galvanic cell concept as in potentiometry, are techniques which, as stated
in Section 11.1, utilize an external power source to drive the cell reaction
one way or the other. In these techniques, the current that results from
this, and not the potential, is measured and related to the concentration
of the analyte. Techniques in this category fall under the general heading
of "amperometry" and include "polarography," which utilizes a special
nonstationary mercury electrode to be described shortly, and
"voltammetry," which utilizes only stationary electrodes.
Electrolytic cells can involve significantly larger currents than galvanic
cells. The two-electrode system and the solution composition discussed
for potentiometry must usually be modified in order to obtain desirable
results. A reference electrode, such as the SCE or the silver-silver chloride
electrode, cannot handle larger currents. In addition, the solution into
which the electrodes are dipped must be able to sustain such a current.
This was not true with potentiometric methods. Thus, for amperometric
methods, a three-electrode system is usually used, and a "supporting
electrolyte" must be added to the solution.
11.3.1a The Three Electrode System
The current measured is usually the current due to a specific oxidation

or reduction process involving the analyte species at the surface of one
of the electrodes. A three-electrode system includes a "working" electrode,
at which the oxidation or reduction process of interest occurs, a reference
electrode, such as the SCE or the silver-silver chloride electrode, and an
"auxiliary" or "counter" electrode, which carries the bulk of the current
(instead of the reference electrode) and "counters" the process that occurs
at the working electrode. The three electrodes are connected to the power
source which is a specially designed circuit for precise control of the
potential applied to the working electrode and is often called a
"potentiostat" or "polarograph." The working electrode can be made
positive or negative with the flip of a polarity switch on the instrument.
The use of the reference electrode is crucial for the precise control of the
potential of the working electrode. The process occurring at the auxiliary
electrode is unimportant analytically. However, it is important not to
allow the products of the reaction occurring there to interfere with the
process occurring at the working electrode. For this reason, the auxiliary
electrode is often placed in a separate chamber with a fritted glass disc
allowing electrical contact with the rest of the cell, but not allowing
diffusion of undesirable chemical species to the working electrode (see
Figure 11.10).
The typical experiment consists of scanning a potential range within
which the process of interest is forced to occur at the surface of the
working electrode. The current that flows as a result is measured by the
instrument and displayed, usually on a recorder. This visual display has
characteristics which depend on the particular potential application used,
as we will see, and also on whether the electrode is stationary or
nonstationary. The current due to the process of interest is determined
from this display and related to the concentration of the analyte species.
11.3.1b The Supporting Electrolyte
The need for the supporting electrolyte mentioned above is obvious

from the standpoint that the solution tested must be an electrolyte solution
if it is to sustain a measurable current. Real-world sample solutions and
standards which have very low analyte concentrations often do not have
FIGURE 11.10 A drawing of a three-electrode system typical of most amperometric

experiments.
a sufficient electrolyte content. In addition, the presence of a supporting

electrolyte can help bring the potential of the electrode process into the
desired range and free from interferences because of the possibility of
complexation and other reactions between the analyte species and the
supporting electrolyte species. Such a reaction can create a complex ion,
for example, which could shift the potential required to an interference-
free range. The possibility of the supporting electrolyte species itself
interfering with the oxidation or reduction of the analyte is eliminated
by choosing an electrolyte species that is difficult to oxidize or reduce
compared to the analyte species or the complex ion. The instrument is
capable of precise control of the potential, and thus capable of differentiating
between the different species present in the solution. Thus, the analyte
species can be "electroactive" (oxidized or reduced at the electrode sur
face) when the supporting electrolyte is not.
11.3.2 Polarography
11.3.2a Introduction
We have previously defined polarography as an amperometric tech

nique which utilizes a special nonstationary electrode for the working
electrode. We now proceed to describe this electrode in detail and con

tinue with a discussion of the modern techniques associated with it.
The electrode is a "dropping mercury electrode" or DME. It consists
of liquid mercury flowing through a very narrow-bore capillary tube, as
shown in Figure 11.11. The mercury flows through the capillary by
gravity from a reservoir of mercury. Drops of mercury form at the tip
of the capillary, grow, fall, and reform in fairly rapid intervals (0.5-10 sec).
This sequence of events is depicted in Figure 11.11. Figure 11.12 shows
the complete apparatus with connections made from the electrodes to the
instrument.
The DME was invented in the early 1900s by a Nobel Prize-winning
chemist by the name of Heyerovsky. Its fluid nature eliminates two
problems normally associated with amperometric measurements: (1)
contamination of the electrode surface resulting from electrode reactions
(the mercury surface is continuously renewed in the case of the DME),
and (2) the decay of the current to small values with time due to the
diffusion limited transportation of analyte to the electrode surface when
the solution is not stirred (the solution is stirred in the case of the DME
each time the drop falls).
As we will soon see in the sections to come, it is desirable for the
instrument to know the precise moment at which the drop will fall.
Mechanical drop timers have been invented which "knock" the capillary
at regular intervals such that the drop falls when required. Modern drop
timers utilize a larger bore capillary and a spring-loaded polyurethane
"plug" at the top of the capillary for controlling the drop. Figure 11.13
shows this more recently developed unit. In either case, the precise time
the drop is to fall is controlled electrically such that the instrument makes
its measurement at that exact moment, usually just before the drop falls.
11.3.2b Classical Polarography
The "classical" or "normal" polarography experiment consists of

scanning the potential range in which the analyte becomes electroactive
and monitoring the current continuously at the same time. Since the
current that will flow at any electrode surface is proportional to the area
of the electrode surface exposed to the solution (in addition to the analyte
concentration), the current vs potential display on the recorder shows a
considerably unstable (but uniform) current. As the drop grows, the
FIGURE 11.11 The dropping mercury electrode (DME) showing the growth and fall,
etc. of the mercury drop. (From Kenkel, J., Analytical Chemistry for
FIGURE 11.12 The "complete" polarography apparatus.

FIGURE 11.13 A modem apparatus for the mechanical control of the drop time and
size. (Courtesy EG&G Princeton Applied Research, Princeton, NJ.)
FIGURE 11.14 The E/tim e characteristic (left) and the recorder trace (right) for the
classical polarography experiment described in the text. The "up and
down" pattern on the right is due to the continuous growth and fall
of the mecury drop.
current increases, but when the drop falls, the current drops. The con
tinuous growing and falling of the drop is thus accompanied by a con
tinuous up and down motion of the recorder pen. The potential-time
characteristic and the resulting current display (polarogram) for the
reduction of cadmium ion in 0.1 M N aN 0 3 is shown in Figure 11.14. The
potential-time characteristic is often called a "ramp" because of the linear
increase apparent in the figure.
Parameters associated with the polarogram are the "diffusion current,"
id, the "half-wave potential," E1/2, and the "background current." The
diffusion current is the current which is due directly to the presence of
the analyte species in the solution. It is the increase in the measured
current that occurs when the potential for the reduction or oxidation of
the analyte species is reached as electrode potential is linearly increased.
The half-wave potential is the potential at which this increase in the
current is half of its full value. The background current is the current that
would be observed if the analyte species were not present in the solution.
Sometimes this background current includes the current due to other
oxidation or reduction processes occurring in the event there are other
species in solution that can be reduced or oxidized at lower potentials.
In the absence of these other species, this current would be very small
and is called the "residual" or "charging" current. The diffusion current,
half-wave potential, and residual current are all graphically defined in
Figure 11.15.
The half-wave potential is often taken to be the potential at which
FIGURE 11.15 A classical polarogram and the definitions of diffusion current, id;
half-wave potential, E1/2, and residual current, iR.
reduction or oxidation of a given species occurs. It can be thought of as

the "barrier" to be crossed in order for oxidation or reduction to occur.
This barrier can be crossed from either direction, but must be crossed in
order to observe the process desired. Usually we cross it from one direction
only and observe the magnitude of the current that results. In a few
techniques, however, it is desirable to cross it while proceeding in one
direction, then reverse the scan and cross it from the other direction too.
We will describe these techniques in the next several subsections. Prob
ably the most notable fact about the half-wave potential is that it is the
one parameter which identifies a current increase to be due to a particular
dissolved chemical species. One might think of it then as a parameter for
qualitative analysis. However, the greatest usefulness of this lies in the
fact that one can use it to identify the precise potential at which the current
due to the analyte is expected and found and thus allow the correct
diffusion current to be measured. This is important because it is the
magnitude of the diffusion current that is related to concentration, and
one must know which diffusion current to measure in the event that there
are several current increases in a given polarogram. Figure 11.16 shows
an example of one such polarogram.
Notice that the residual and background currents in Figure 11.16 are
FIGURE 11.16 A polarogram showing several diffusion currents due to the presence
of several electroactive species in the solution. The analyst may identify
the analyte species' diffusion current by the magnitude of the associated
half-wave potential.
steadily increasing slightly as the potential is increased. This is due to the

fact that the mercury drop is becoming more and more charged as the
potential is increased. The instrument sees this as a slight current flow
and displays it on the recorder. This is a "non-Faradaic" type of current.
A "Faradaic" current is one that is due to the oxidation or reduction of

a dissolved chemical species. It is due to an electron transfer process. A
non-Faradaic current is one that is not due to an electron transfer process.
Polarographic "maxima" are sometimes encountered in the current/
potential display, and these can inhibit the accurate measurement of the
diffusion current. Polarographic maxima are distortions of the polarogram
manifested by a large "peak" of current immediately beyond the half
wave potential (Figure 11.16). The cause is not yet well understood,
however, the problem can usually be solved by the addition of small
amounts of gelatin or a surfactant, such as Triton X-100, to the solution.
11.2.3c Modem Polarographic Techniques
Modern polarography instruments are capable of several modifica

tions of the classical procedure just described. These are "sampled dc
polarography," "pulse polarography," and "differential pulse
polarography." Each differs from classical polarography either because
of the nature of the potential-time characteristic, or the mode of current
measurement, or both. In most of these, the sensitivity is dramatically
increased over the classical mode. In one of them, the resolution of two
closely spaced half-wave potentials is improved. The techniques are
summarized in terms of their potential-time characteristics and their
current-potential recorder trace in Figure 11.17.
Sampled dc polarography is identical with the classical technique
except that the current is only "sampled" and displayed just before the
drop falls. The advantage is that it makes the diffusion current easier to
measure — such a current is free of the continuous up and down pattern.
Pulse polarography is like sampled in that the current is sampled just
before the drop falls. However, the potential-time characteristic is very
different; a series of increasing potential pulses is applied to the DME,
as shown, each time just before the drop falls. The current sampling is
done while the pulse is being applied, which is for less than a tenth of
a second. Thus, the following sequence of events occurs in rapid order:
(1 ) the pulse is applied, (2) the current is measured and displayed, (3)
the pulse drops back, and (4) the drop falls. After a delay (on the order
of seconds), this sequence begins again, but with a higher pulse. Once
the pulse is high enough to cross the half-wave potential, the current
begins to rise due to the reduction or oxidation of the analyte. The
FIGURE 11.17 Modern polarography techniques compared in terms of E-time

characteristic and i-E trace: (a) classical dc, (b) sampled dc, (c) pulse
polarography, and (d) differential pulse polarography.
advantage is a much higher current for the same concentration, and thus
a better sensitivity.
With differential pulse polarography, both the potential-time charac
teristic and the current sampling method are different from the others.
The potential-time characteristic is one in which pulses of constant
magnitude are superimposed on the ramp as shown. The current is
sampled both before and during the application of the pulse. These two
current values are subtracted from each other and divided by the poten
tial difference, creating a signal which represents the rate of change or
the "derivative" of the original polarogram. It is this signal that is then
displayed on the recorder as the sampled currents were for the others.
The trace resembles a peak because it is the rate of change of the original
polarogram, and in the original, there is very little change before and after
the rise in current, while a sharp change occurs where there is a rise. Thus,
a peak occurs at or near E1/2. The advantage of this technique is not only
that it is very sensitive (typically more sensitive than any of the others),
but also, two species that have similar E)/ 2 values, while not distinguished
before, can be with this technique.
11.3.2d Quantitative Analysis
In classical (and sampled dc) polarography, the diffusion current, the

current increase that occurs when the half-wave potential of the analyte
is reached, is directly and linearly related to the concentration of the
species involved. Quantitative analysis is then based on the measurement
of the diffusion current. The usual procedure is to prepare a series of
standard solutions and measure the diffusion current for each. A plot of
id vs concentration then is useful in determining the concentration in an
unknown. In pulse polarography, the current increase observed for each
of a series of standard solutions is plotted vs concentration. We have
already stated that this current is typically much higher than that for
classical and sampled dc, and thus quantitative analysis is much more
sensitive. See Figure 11.18 for a comparison between sampled dc and
pulse polarography.
The height of the peak in differential pulse polarography is proportional
to concentration. Thus, in this technique, the peak height for each of a
series of standard solutions is measured and plotted vs concentration.
FIGURE 11.18 A comparison of a pulse polarogram with a sampled dc polarogram

for the same concentration of electroactive species.
11.3.3 Voltammehy
11.3.3a Introduction
As stated previously, "voltammetry" is an amperometric technique in

which the working electrode is some stationary electrode and not the
DME. Typical electrodes here include small platinum (or gold) strips,
wire coils, or beads sealed into the tip of a glass tube so that a small area
or cross-section is exposed and polished. They may also consist of a small
bead of graphitic carbon sealed into the tip of a glass rod and also finely
polished as the others. This latter electrode is referred to as a "glassy
carbon" electrode. A third type is the so-called hanging mercury drop
electrode (the HME). This electrode is like the DME, except the mercury
is not flowing but held stationary.
The nature of the voltammetry techniques we will now describe dictates
that the electrode be stationary. In one case, the electrode serves as a
collection point for the analyte, thus precluding the need to have a
continuously renewed surface. In another case, the electrode monitors the
chemical reactivity near the electrode surface, thus precluding the need
to have the solution stirred.
11.3.3b Stripping Voltammetry
This technique is most useful for metals that are able to be electroplated
onto the surface of a working electrode that has been polarized to some
negative potential so that reduction of the metal ion to the metal atom
will take place, i.e., at a potential more negative that the half-wave potential
for such a reduction. This reduction occurs over a period of time, perhaps
even on the order of 1 or 2 h, while the solution is stirred (with the use
of a stirrer) to aid in the transport of the metal ion to the electrode surface.
Over this period of time, a fairly large quantity of metal ions are reduced,
and the metal either deposits onto the surface of the electrode or, in the
case of the HME, a very common electrode for this technique, dissolves
into the mercury drop. After the preset period of time has elapsed, a
"reverse" scan is performed so that the applied potential crosses the
E 1 / 2 "barrier" and all the metal is oxidized back to the metal ion, in effect
"stripping" it from the surface. The resulting oxidation current (usually
the total current measured in coulombs) is compared to that of a standard
that has undergone the same experiment and the concentration determined.
The potential-time characteristic and the resulting current pattern are
shown in Figure 11.19. The advantage is that a solution of a very low
concentration can be analyzed because the analyte is preconcentrated in
or on the electrode over a period of time, making the stripping current
easily measurable.
11.3.3c Cyclic Voltammetry
This is a technique that is seldom used for quantitative analysis, but

is a popular technique for other reasons, such as the determination of
electrode reactions mechanisms. The potential-time characteristic is shown
in Figure 11.20.
As the potential is increased past E1/2, the current is monitored
continuously. The potential scan is then reversed and returned at an equal
rate back to the starting point. The current/potential curve, if the electrode
reaction is reversible, is shown in Figure 11.21a. If it is not reversible, that
is, if the product of the electrode reaction has been reacted to form another
species, the result is shown in Figure 1 1 .2 1 b. The oxidation current evi
dent in (a) shows that the product of the reduction is still available at the
electrode surface to be oxidized back. In (b) it is not available.
FIGURE 11.19 The potential and current characteristics of a stripping voltammetry

experiment. (From Kenkel, J., Analytical Chemistry for Technicians, Lewis
FIGURE 11.20 The potential-time characteristic in cycle voltammetry. (From Kenkel,

J., Analytical Chemistry for Technicians, Lewis Publishers, Inc., Chelsea,
11.3.3d Amperometric Titration
The concept of current measurement in voltammetry can be applied

to a titration experiment much like potential measurements were in
Section 11.2 (potentiometric titration). Such an experiment is called an
"amperometric titration"; a titration in which the end point is detected
through the measurement of the current flowing at an electrode.
The potential of the measuring (working) electrode, which is typically
a rotating platinum disk embedded in a Teflon sheath, is held constant
at some value beyond the half-wave potential of the analyte. The solution
FIGURE 1 1 .2 1 Cyclic voltammograms of (a) a reversible electrode process and (b)

of an irreversible process. (From Kenkel, J., Analytical Chemistry for
is stirred due to the rotation of the electrode. The resulting current (the
diffusion current) is then measured as the titrant is added. The titrant
reacts with the electroactive species, removing it from the solution, and
thus decreasing its concentration. The measured current therefore also
decreases. When all of the analyte has reacted with the titrant, the decrease
will stop and this signals the end point.
C hapter 12
A u to m a tio n
12.1 INTRODUCTION
There is a continuing effort to streamline analytical laboratory proce

dures. Over the years, such procedures have developed from the time-
consuming classical wet methods of analysis to the more sophisticated
and faster instrumental methods. In turn, the instruments have undergone
continuous improvement and streamlining to the point where large
samples and standards are able to be tested in a very short period of time
through what are called "automated" procedures. Today, automatic
titrators, automatic analyzers which incorporate instrumental designs
and concepts, and automatic samplers which are attached to traditional
instruments, are commonplace. Even complex sample preparation schemes
involving sample dissolution and extraction utilize robotics to help
streamline the procedures. In this chapter, we discuss the design and
theory of automatic titrators, the Technicon AutoAnalyzer/ flow injection
analyzers (FLA), and robotics.
Registered trademark of Bran & Luebe, Inc., Buffalo Grove, Illinois.
319
12.2 AUTOMATIC TITRATIONS
When a large number of titrations is part of a laboratory's daily workload,

it is possible for the laboratory to employ a partially automated or even
a fully automated system. This system may be something simple, such
as a unit that automatically refills the buret, or refills the buret with the
use of a vacuum bulb, after a titration. In this case, a turn of the stopcock
causes the buret to refill either by gravity or by vacuum from a large
reservoir of titrant. It may also be an automatic dispensing buret with
a digital display of titrant volume. Such a unit fits into a large reservoir
of titrant solution and dispenses the solution into the titration flask. The
operator starts and stops the titration according to his observations of the
indicator, but the titrant volume is automatically measured and displayed,
thus eliminating meniscus reading errors.
A fully automated system involves what is called an automatic titrator
or autotitrator. Such a unit not only utilizes an automatic dispenser, such
as described above, but also has automatic potentiometric, amperometric,
or coulometric end point detection. The unit can even employ a computer
controlled carousel in which samples rotate one after the other
automatically through the unit. Such a system automatically activates
reagent addition, stirring, measuring, titrating, aspirating, rinsing, and
reconditioning of electrodes.
12.3 SEGMENTED FLOW METHODS
The Technicon AutoAnalyzer, developed in the late 1950s and early

1960s, represents the widely used analysis system that is based on
"segmented" flow. It is an automation system in which the sample
solutions, and various reagents with which the sample solutions need to
be mixed, flow from their original containers through a maze of plastic
tubes, the diameters of which define the rate of flow and thus the
proportions to be mixed. At appropriate and strategic locations along the
flow route, the reagents and samples come together and mix, creating
the solution to be measured. In addition, air is drawn into the system also
through one or more plastic tubes, and thus air bubbles are introduced
as a result which help to divide the flowing liquid into segments, hence
Automation 321
m/l' wjh vwx v/s/z^i

FIGURE 1 2 .1 An illustration of a plastic tube through which an air-segmented
solution is flowing. (Adapted from Publication No.TNl-0169-00,
Technicon Instrument Corporation, Tarrytown, NY; Bran & Luebbe
Analyzing Technologies, Elmsford, NY. With permission.)
FIGURE 12.2 An illustration of a peristaltic proportioning pump. (Reprinted from

Publication No.TNl-0169-00, Technicon Instrument Corporation,
Tarrytown, NY; Bran & Luebbe Analyzing Technologies, Elmsford,
NY. With permission.)
the "segmented flow" designation (see Figure 12.1). The complete system
consists of a series of specific modular units assembled for a particular
analysis. Modular units common to all analyses using this system are an
automatic sampler (an "auto-sampler"), a peristaltic proportioning pump,
a colorimeter, and a recorder. The proportioning pump is "peristaltic,"
which means the pumping action is performed through the repeated
squeezing of the tubes through the action of metal rollers (see Figure 12.2).
The system of tubes positioned in the pump along with the interconnecting
tubes, mixing coils (see below), and glass and plastic pieces is called the
"manifold." The auto sampler consists of a rotating carousel which
positions vials containing the sample solutions in such a way that they
are drawn out, one after another, through one of the tubes leading to the
proportioning pump. The length of time during which the sample is
drawn out is determined by an interchangeable cam in the carousel.
At the same time, the pump also delivers the air bubbles, a diluent if
desired, and one or more reagents needed for color development such
that all of these ultimately meet and move through a single tube. Thorough
mixing then takes place by channeling this mixture, with the air bubbles,
through coiled glass tubing for mixing. The air bubbles are of special
assistance to the mixing process. Without air bubbles, a layer of a liquid
adheres to the wall of the tubing causing contamination of the flow stream
RECORDER COLORIMETER PROPORTIONING PUMP SAMPLER

AND MANIFOLD
FIGURE 12.3 A drawing of the hardware components of the Technicon

AutoAnalyzer system as described in the text. (Adapted from
Publication No.TN0-0169-10, Technicon Instrument Corporation,
to follow. After mixing and color development, the air bubbles are drawn
off (via a "debubbler"), the solution flows through a detector, usually a
colorimeter equipped with a flow-through cuvette, and the transmittance
level is traced on the recorder and/or fed into a computer. The entire
system, assembled for an analysis such as just described, is illustrated in
Figure 12.3.
Specific methods and analytes require specific tubing diameters, junction
tubes, mixing coils, etc. To transform a system designed for phosphate
analysis, for example, into one for chloride analysis requires a complete
dismantling and reassembly of the manifold, unless the manifold systems
are stored so that they can be reinstalled later. Technicon manufactures
and sells all tubing and parts needed for a particular manifold. The tubing
is color-coded and tubing junctions and mixing coils are given part
numbers which are specified in AutoAnalyzer system flow diagrams
available from Technicon or found in method books, such as in Standard
Methods for the Examination of Water and Wastewater. Figure 12.4 gives two
such system flow diagrams, one for chloride in water and wastewater
(a) and one for phosphate in water and seawater (b). The color codes, such
as "BLK/BLK" and "ORN/GRN," symbolize the colors of plastic clips
that are found on the opposite ends of a given tube, and their associated
flow rates are given within the parentheses next to the stated color code
as shown. The clips are used to stretch and install the tubes in the pump.
Part numbers for the tubing junctions and mixing coils are also given in
these diagrams. You will notice, for example, in the chloride system
(Figure 12.4a) that a 14-turn mixing coil has part number 116-0152-02.
Automation 323
CHLORIDE IN WATER AND WASTEWATER

(RANGE: 0.2-10 mg/l)
MANIFOLD NO. 116-0061-01
T O S A M P L E R IV ^ -v G R N /G R N (2.001 W A T E R
W A SH R E C E P T A C LE
^ B L K /B L K (0 3 2 ) A IR
-^VHT/WHTj0 6 ^
6 Tum i
r01
S A M P LE R IV
170 -0 10 3 6 0 /H R
0007 La
fl1 6 ~ 0 4 8 9 -0 1
q G R Y /G R V (1 001 S A M P LE 6:1
q O R N /G R N (0.101 D IL U E N T W A T E R ,10 nitI B R IJ-35
► T O F /C -^ G R Y /G R Y (1 .0 0 ) C O L O R R E A G E N T
PUMP TUBE
C O L O R IM E T E R - v G R Y /G R Y (1 .0 0 ) F R O M F /C
IS m m F /C N O T E : F IG U R E S IN P A R E N T H E S E S S IG N IF Y
1 6 2 -8 0 4 0 FLO W R A T E S IN M L /M IN .
• 0 .0 3 4 ” P O L Y E T H Y L E N E
O R T H O P H O S P H A T E IN W A T E R A N D S E A W A T E R
RANGE- ^ P /|
0 -124 m 9 P / I (PPb)
M A N IF O L D N O . 11S -D 221-01
T o Somplor IV G R N /G R N (2 .0 0 ) W A T E R
1 57 -B 2 7 3 -0 3 Wash R a ca p ta c la - ^ “O
3 7 .5 ° C 5 T u rn a 5 Turn a B L K /B L K (0 .3 2 ) A IR
7 .7 ml 1 7 0 -0 1 0 3 1 7 0 -0 10 3
M M *.0WW B L K /B L K (0 .3 2 ) W A T E R
S A M P L E R IV
3 0 /H R
C O L O R IM E T E R P O L Y E T H Y L E N E 0 .0 3 4 ID
2:1
880 nm
50 mm F / C x 1 .5 mm 10 N O T E : F IG U R E S IN P A R E N T H E S E S S IG N IF Y
199-B 023-D 1 FLO W R A T E S IN M L /M IN .
FIGURE 12.4 (a) The AutoAnalyzer system for chloride, (b) The AutoAnalyzer
system for phosphate. (Courtesy of Technicon Instrument Corpora
tion, Tarrytown, NY; Bran & Luebbe Analyzing Technologies,
Elmsford, NY.)
You should also notice in Figure 12.4b that one of the mixing coils, the
one positioned just before the colorimeter, requires an elevated tempera
ture (37.5°C). A high temperature bath is another module that can be
added to the system. Yet another module is a dialyzer which is needed
for glucose analysis. The colorimeter (See Chapter 6 ) consists of a visible
light source, glass filters for monochromators, a flow cell (cuvette), a
photocell, and the associated electronics required to produce a percent
transmittance (%T) signal to be sent to the recorder. The air bubbles must
FIGURE 12.5 Flow cell and debubbler. See text for discussion. (Reprinted from
Publication No.TNl-0169-00, Technicon Instrument Corporation,
be removed prior to the solution entering the colorimeter. A "debubbler"

is used for this. Figure 12.5 is a diagram of the debubbler in combination
with the flow cell. The air bubbles float and pass upward and out while
the solution is drawn down into the flow cell by the vacuum created due
to a connection to the pump.
The recorder records the %T signal over time (it is a strip-chart re
corder). It is calibrated in %T units. As sample after sample (separated
by distilled water volumes) drawn from the auto sampler pass through
the system, a recorder signal for each is generated, and these signals are
then recorded one after the other on the chart paper. Figure 12.6 shows
a sample of such a recorder trace. A popular modem alternative to the
recorder is data acquisition with a computer.
A wealth of information, including additional details of the system
setup and details of specific methods and manifolds for various analytes,
is available from the Technicon Instrument Corporation.
12.4 FLOW INJECTION METHODS
More recently, an automated flow system in which the various liquid

flow streams are not segmented has been developed. To say that they are
not segmented means that air bubbles, which are characteristic of the
Technicon system described in the previous section, are not introduced
\ :__;__j___} !
Automation
325
FIGURE 12.6 A typical AutoAnalyzer strip-chart recording of a series of solutions.

at any time, and thus the various flow streams are not characterized by
segments separated by the bubbles. This nonsegmented type of system
is known as Flow Injection Analysis or FIA.
In the segmented flow method, the air bubbles serve to help mix the
various flow streams after they come together in the system. The obvious
question with regard to FIA then is how do the various flow streams
adequately mix prior to measurement. With FIA, the tubing diameters
are much smaller (0.5 mm i.d.), and a much smaller amount of sample
is "injected" with a valve similar to HPLC systems described in Chapter
10. Thus, the need for thorough mixing is greatly reduced, as adequate
mixing of such small quantities occurs more readily without the need for
air bubbles. A small "plug" of sample is thus carried through the flow
system and sharp spikes are observed on the recorder trace. The advantages
are (1 ) greater sample throughput, meaning a more rapid analysis of large
numbers of samples per unit time; (2 ) shorter time from injection to
detection; (3) faster start-up, shut-down, and changeover; and (4) less
sophisticated equipment. For an excellent review of FIA, please refer to
a paper which appeared in Analytical Chemistry in 1978* and again in
Volume 2 of Instrumentation in Analytical Chemistry in 1982.**
12.5 ROBOTICS
Given the large volume of samples to be analyzed in many analytical

laboratories, total automation of laboratory procedures has been a goal
of analytical chemists and instrument manufacturers for years. With the
advent of automation equipment, such as has been described in this
chapter, and the ever more sophisticated computer and computer software,
discussed in Chapter 5, this dream is coming closer to reality. It is now
commonplace for sample solutions and standards to be rapidly diluted
and mixed with appropriate reagents and measured one after another
quickly as they pass through a flow channel. Along with this, the resulting
data can be acquired, manipulated, statistically adjusted, and stored by
a computer all with only pressing a button or a computer key in an
operation that is not only automated, but may even be unattended for
Betteridge, D., "Flow Injection Analysis," Analytical Chemistry, 50(4), 832A (1978).
Borman, S.A., Ed., Instrumentation in Analytical Chemistry, Vol. 2, American Chemical
Society, Washington, DC, 1982.
Automation 327
FIGURE 12.7 An example of a laboratory robotics system. (Courtesy of Zymark

Corporation, Hopkinton, MA.)
a substantial period of time. All that remains is for the chemist to take
the raw samples and perform the dissolution and/or extraction techniques
described in Chapter 2 manually prior to executing the automated
procedure.
The interesting news is that even these latter operations can be
automated, since robots are available that will perform sample preparation
tasks. Such robots are not androids or other two-legged variety that may
immediately come to mind in light of some futuristic TV program or

movie. A "robot" typically consists of a robotic "arm" that is capable of
360° rotation around a table equipped with the necessary auxiliaries, such
as tube racks, balances, extraction and dissolution solvent reservoirs,
vortex mixers, shakers, dispensers, and centrifuges (see Figure 12.7). The
robotic arm is programmed to move samples from one treatment station
to another in a sample preparation scheme designed to emulate the
manual procedures. Besides having the ability to perform the routine
preparation schemes without becoming bored, requiring a lunch or coffee
break, or being limited to an 8 -h day, the robot is able to work in haz
ardous environments, which is a significant advantage given the increas
ing publicity of health risks associated with handling the chemicals needed
for such schemes. Analytical chemists can fully expect further develop
ments in this area as time goes on.
C hapter 1 3
D ata H andling and

E rror A nalysis
13.1 MODERN DATA HANDLING
13.1.1 Introduction
The molecular spectroscopic techniques of ultraviolet (UV), visible

(vis), and infrared (IR) spectrometry; the atomic spectroscopic techniques
of flame atomic absorption, graphite furnace atomic absorption, and
inductively coupled plasma; the instrumental chromatography techniques
of gas chromatography (GC) and high performance liquid chromatography
(HPLC) — all of these and the myriad of others discussed in this book
have become very popular analysis techniques of the modem analytical
chemistry laboratory. The ability to handle the quantity of data produced
and the ability to assure the quality of the decisions made as a result of
the data produced are uppermost in the minds of laboratory managers
everywhere. It is obviously important for these managers to be keenly
329
aware of modern data handling procedures and to be able to execute

them.
At the heart of modern laboratory data handling is the laboratory
computer, often termed the laboratory "workstation" or "data station."
The use of the computer in the laboratory was briefly discussed in Chapter
5. The modern laboratory utilizes computer workstations to (1 ) acquire
data from instruments, (2) store the acquired data, (3) display and
manipulate the acquired data, (4) control instruments, and finally to (5)
gather and print the results in a way that is acceptable to clients. Professional
analytical chemists must apply their knowledge of chemistry to the
laboratory methods and procedures, but they must also be knowledgeable
in computer usage for all the categories listed above.
13.1.2 Data Display and Analysis
In Chapter 5, the conversion of the analog signal typical of the output

of many types of modern laboratory instruments to the digital signal
required by a computer was discussed. The display of the digitized data
on a monitor screen or a printer/plotter for subsequent interpretation and
analysis is a common activity in these laboratories. Such a display amounts
to a reconstruction of the analog data which allows the analyst to quickly
determine maxima or minima in the data (such as may be required for
chromatography peak heights or the absorbance at the wavelength of
maximum absorption), perform simple mathematical operations on all
or part of the data (such as multiplying by a dilution factor), average all
or a portion of the data (which may be required for statistical analysis
— see the next section), and many other similar manipulations. The
computer especially enhances chromatographic data analysis, since it is
capable of quickly determining retention times (and thus quickly iden
tifying mixture components) as well as quantitation parameters, such as
where to begin and end a peak integration, whether the peak threshold
limit defined by the analyst has been reached, and the actual determi
nation of the peak sizes by the integration (see Chapter 9).
Various companies market the hardware and software required for
computer workstations. Such systems are readily available in various
price ranges and are commonplace in modern laboratories.
Data Handling and Error Analysis 331
13.1.3 Reporting and Managing Results
Modern computerized data stations are capable of automatically

accumulating and assembling data files for the purpose of preparing and
storing personal reports for the analyst, as well as customized reports for
clients. Such reports can be assembled for presentation before a group
or they can also be formatted for importing to standard microcomputer
software, most notably word processing, data base, and spreadsheet
software, which is useful if a formal written report of the results of an
analysis is required.
In laboratories in which the volume of laboratory data generated is
large, the computer is a godsend. The analyst does not even need to view
individual routine results. Often the computer takes over, accumulates
and analyzes the data, and generates reports which it automatically
imports to data base files for a formal printed laboratory report. A visual
check of the resulting report can pinpoint problems and suggest remedies.
13.1.4 The State of the Art
The strip-chart recorders and x-y recorders discussed in Chapter 5 are

fast becoming obsolete, if they are not already. They are useful for
immediate visual inspection of the analog data and are inexpensive, but
they respond slowly and are not capable of automatic range switching
and attenuation. Instrument output signals that "go off scale" or are too
small to see need to be generated a second time.
Computing integrators (Chapter 9) are in common use in GC and
HPLC laboratories. These perform retention time and peak integration
determinations quickly and accurately, but usually have limited report-
generating capability and cannot store and reconstruct chromatograms,
items we discussed above as advantages of modem computer workstations.
They are, however, relatively inexpensive and portable.
Thus, personal computers and computer workstations, as well as
minicomputer and mainframe systems, have transformed laboratory data
handling into job functions that are often as easy as pressing a key on
a keyboard, especially given the increasing speed of the computers'
central processors. Virtually all the functions listed in the preceding
discussions can be performed quickly and accurately, thus making the

handling of analytical data electronically automated. Modem chemical
literature* addresses these issues and should be scanned regularly for
modem developments.
13.2 INTRODUCTION TO ERROR ANALYSIS
" 'In laboratory testing, there are no magic black boxes where you simply
follow directions and a number flashes on the screen that guarantees the truth/
says Dr. Benjamin Turner, a pathologist in private practice in Tallahassee, Fla.
'Laboratory tests can never be 100 percent perfect. Even the finest testing
instruments can break down — and human beings aren't infallible either.'
How true Dr. Turner's words are! They dramatically point out perhaps
the most important aspect of the job of the chemical analyst — to assure
that the data and results that are reported is of the maximum possible
quality. This means that the analyst must be able to recognize when the
testing instmment is "breaking down" and when a "human" error is
suspected. The analyst must be as confident as he/she can be that the
"number that flashes on the screen" does in fact "guarantee the truth"
as much as is humanly possible. The analyst must be familiar with error
analysis schemes that have been developed and be able to use them to
the point where confidence and quality is assured.
Errors in the analytical laboratory are basically of two types:
"determinate," also called "systematic," and "indeterminate," also called
"random." Determinate errors are avoidable blunders that were known
to have occurred, or at least were determined later to have occurred, in
the procedures. They arise from such avoidable sources as contamination,
wrongly calibrated instruments, reagent impurities, instrumental
malfunctions, poor sampling techniques, incomplete dissolution of sample,
errors in calculations, etc. Sometimes correction factors can adjust for their
occurrence, at other times the procedures must be repeated so as to avoid
the error.
Indeterminate errors, on the other hand, are impossible to avoid. They
See, for example, a column entitled "The Data File" by G.I. Ouchi appearing regularly in
the journal GC«LC, which is a journal dedicated to modern instrumental chromatography.
From Heller, L.J., "How Accurate Are Medical Tests?", Parade Magazine, February 3,1991.
Reprinted with permission from Parade 1991.
are random errors; errors which either were not known to have occurred,
or which were known to have occurred, but could neither be accounted
for directly nor avoided. Examples include problems inherent in the
manner in which an instrument operates, errors inherent in reading a
meniscus, errors inherent in sample and solution handling, etc. Since they
are unavoidable and unknown, they are presumed to affect the result
both positively and negatively and are dealt with by statistics. A given
result can be rejected as being too inaccurate based on statistical analysis
indicating too great a deviation from the established norm. The procedure
to check for such errors involves running a series of identical tests on the
same sample, using the same instrument or other piece of equipment,
over and over. Those results that agree to within certain predetermined
limits are averaged, and the average is then considered the correct answer.
Any results that fall outside these predetermined limits are "rejected" and
are not used to compute the average. The "rejectability" parameters are
the standards which a given laboratory must determine and adopt for
the particular situation. Some methods for the determination of these
parameters will be discussed in this chapter.
13.3 TERMINOLOGY
In working with the statistical methods used to establish the "correct"

answer in an analysis, a number of terms appear which need to be
defined. These are as follows:
1. Mean — In the case in which a given measurement on a sample is

repeated a number of times, the average of all measurements is an
important number and is called the "mean." It is calculated by
adding together the numerical values of all measurements and
dividing this sum by the number of measurements.
2. Median— For this same series of identical measurements on a sample,
the "middle" value is sometimes important and is called the median.
Thus the median is always one of the actual measurements, while
the mean may not be. However, if the total number of measurements
is an even number, there is no single "middle" value. In this case,
the median is the average of two "middle" values.
3. Mode — The value that occurs most frequently in the series is called
the "mode." For a large number of identical measurements, the
median and mode should be the same.
4. Deoiatwn — How much each measurement differs from the mean

is an important number and is called the deviation. A deviation is
associated with each measurement, and if a given deviation is large
compared to others in a series of identical measurements, this may
signal a potentially rejectable measurement which will be tested by
the statistical methods. Mathematically, the deviation is calculated
as follows:
d = |m - e | (13.1)
in which d is the deviation, m is the mean, and e represents the

individual experimental measurement. [The bars ( I I ) refer to
"absolute value," which means the value of d is calculated without
regard to sign; i.e., it is always a positive value.]
5. Relative Deviation — Perhaps of more practical significance is the
"relative deviation." This quantity relates the deviation to the value
of the mean. If the mean is a relatively large number, a deviation that
appears to be large may not be out of line because the value from
which it shows deviation is large also. Thus a relative deviation is
more useful. Mathematically, it is defined as follows:
in which dR is the relative deviation. dR can also be expressed as a

percent, or part per thousand, etc.:
relative % deviation = dR x 100 (13.3)
relative ppt deviation = dR x 1000 (13.4)
in which ppt represents "parts per thousand."

6. Average Deviation— An overall picture of the quality of data (in terms
of precision) is the "average deviation." It is the average of all
deviations for all measurements:
(13.5)
n
in which da is the average deviation and n is the number of devia

tions (corresponding to the number of measurements) considered.
7. Standard Deviation — Of more popular use for demonstrating data
quality is the "standard deviation," which is similar to the average
deviation:
(13.6)
The term (n - 1 ) is referred to as the number of "degrees of freedom,"

and s represents the standard deviation. The significance of both da
and s is that the smaller they are numerically, the more precise the
data, and thus presumably the more accurate the data. The standard
deviation is used in most statistical methods in one way or another
in evaluating reliability, that is, in establishing the level of confidence
for arriving at the correct answer in the analysis.
8. Relative Standard Deviation — One final deviation parameter is the
relative standard deviation, sR. It is analogous to dR in the sense that
dR is obtained by dividing d by the mean and multiplying by 100
or 1000. In this case s is divided by the mean and then multiplied
by 100 or 1000:
s
S,R (13.7)
m
and
relative % standard deviation = sR x 100 (13.8)
and
relative ppt standard deviation = sR x 1000 (13.9)
This parameter relates the standard deviation to the value of the

mean and represents a practical and popular expression of data
quality.
13.4 DISTRIBUTION OF RANDOM ERRORS
A graphical picture of the distribution of the results of an experiment,

and thus a picture of the distribution of the random errors, is the "Normal
Distribution Curve." It is a plot of the frequency of occurrence of a result
on the y-axis vs the numerical value of the results on the x-axis. The
normal (also called "Gaussian") distribution curve is bell-shaped, with
the average of all results being the most frequently observed at the center
of the bell shape and an equal drop-off in the frequency of occurrence
in both directions away from the mean (see Figure 13.1). It is a picture
of the precision (and thus presumably accuracy — see Chapter 1) of a
given data set. The more points there are bunched around the mean and
the sharper the drop-off away from the mean, the smaller the standard
deviation, the more precise the data, and the more confidence one can
have in any one result being correct. Typically, any measurement that
is within predetermined confidence limits, and thus within the so-called
confidence "interval," is deemed to be of sufficient accuracy to be reported
or to be included in the calculation of the mean, which is then reported.
Any measurement outside these confidence limits are of insufficient
accuracy and rejected.
13.5 STUDENT'S t
A more formal method is a calculation taking into account the mean,

the standard deviation, the number of measurements, and a "probability
factor" (f). This latter parameter was the result of the work of W.S. Gossett,
who published it under the pen name "Student" in 1908. It is often called
"Student's t." The calculation establishes the confidence limits:
analysis result = m ± 1211 (13.10)

n
and thus an analysis result, r, can be expressed as r ± c, where c is the

confidence limit or interval.
The values of t depend on the number of measurements, n, or, more
- — Values of measurements — -
FIGURE 13.1 A normal distribution curve showing confidence limits a particular
number of standard deviations from the mean, such as 1 or 2.
Table 13.1 t Values for Calculating Confidence Limits

n -1 t (at 90% level) t (at 95% level) t (at 99% level)
1 6.3 12.7 63.7

2 2.9 4.3 9.9
3 2.35 3.2 5.8
4 2.13 2.78 4.6
5 2.02 2.57 4.03
6 1.94 2.45 3.71
7 1.90 2.37 3.50
8 1.86 2.31 3.36
9 1.83 2.26 3.25
10 1.81 2.23 3.17
00 1.64 1.96 2.58
levels and for a number of different degrees of freedom. The smaller the
number of measurements, the less confidence the analyst has in his/her
analysis result, as indicated by the larger values for t in Table 13.1. The
larger the value of t, the larger the confidence interval. The more mea
surements that are made the better. From a practical point of view, it is
the analyst's decision as to how much time is spent making measure
ments weighed against a desired confidence in the results. An analyst can
reject results that push the interval past a desired value.
13.6 STATISTICS IN QUALITY CONTROL
In addition to providing the basis for rejection of "bad" laboratory data,

statistics can also be used as a basis for detecting unacceptable batches
of raw materials or problems in a quality control laboratory. If a given
analysis yields results that are quite different from what is expected, the
"rejection" indicated by statistics results in the rejection of the batch or
sample from being used or shipped, or whatever.
Confidence limits that are exceeded can also sometimes be indicated
in an instrument reading that exceeds a statistically predetermined limit,
rather than a final answer. An example is the GC analysis of cellophane
extracts for formaldehyde. Injection of the extracts one after the other into
a GC may give a series of formaldehyde peaks as shown in Figure 13.2.
Sample number 5 in this chromatogram shows a peak that exceeds the
preset confidence limit. Under these circumstances, the batch of cello
phane from which this sample was acquired is discarded.
In addition, "quality control charts" are sometimes kept in a lab in
order to detect problems or trends in analysis procedures or material
analyzed. The idea is that over a period of time, certain standard deviation
values or confidence limit values are established which create a range of
acceptable analysis results. A measurement falling outside the range or
a trend away from the mean would then indicate a problem — possibly
a determinate error, or a series of bad raw material batches, etc. An
example of a quality control chart is shown in Figure 13.3.
13.7 ASSISTANCE
For a detailed discussion of these and other concepts relating to quality

assurance in an analytical laboratory, you are referred to a book written
by John Keenan Taylor.’ Dr. Taylor's book is written to provide assistance
for beginning a quality assurance program in a laboratory and provides
the guidance needed to develop the reliability reputation that all
laboratories seek. The underlying goal sought by all analytical laboratories
is that all analytical results must be reliable, and there must be undisputable
evidence to prove it. Dr. Taylor's book is based on this premise.
Taylor, J.K., Quality Assurance of Chemical Measurements, Lewis Publishers, Inc., Chelsea,
MI, 1987.
TIME ------ ►
FIGURE 13.2 A hypothetical gas chromatogram of a series of cellophane extracts.

See text for a brief description.
FIGURE 13.3 A typical quality control chart. Notice the measurements outside the
range (defined by S) on day 4 and the trend away from the mean after
about day 7.
IN D E X
AA, see Atomic absorption in HPLC, 274

Absorbance, in U V /v is spectro Adsorption chromatography,
photometry, 117,122 2 1 1 -2 1 2 ,2 6 1 ,2 7 1 ,2 7 4 , see
Absorption also High performance
atomic, see Atomic absorption liquid chromatography
(AA) (HPLC)
of light, 108-110 Alkalinity, wet methods and, 83
Absorptivity Ammonium hydroxide
Beer's Law and, 126 density of, 55
molar, Beer's Law and, 127 molarity of, 50
Accuracy, precision vs, 3 -5 percent composition of, 55
Acetic acid for sample preparation, 39
density of, 55 Amperometric detectors, 28 0 -
molarity of, 50 282
percent composition of, 55 Amperometric titration, 31 7 -
Acetone, in extraction, 40-41 318
Acetonitrile, in extraction, 40 Amperometry, 303-318, see also
Acids Polarography;
densities of, 55 Voltammetry
equivalent weights of, 57-58 Ampholyte, isoelectric focusing
molarities of, 4 9 ,5 0 and, 223
percent composition of, 55 Analyte
for sample preparation, 36-39 defined, 6
Activity, Nernst Equation and, extraction of, see Extraction
290n Analytical balances, see Bal
Activity coefficient, 290n ances
Adjusted retention time Analytical chemistry, impor
in gas chromatography, 239, tance of, 1-3
249 Analytical colums, 233
341
342 Analytical Chem istry Refresher Manual
Analytical separations, 195-224 alternative atomizers for, 187-

chromatography in, 207-220, 191
see also Chromatography burners for, 169-171
distillation, 196-198 excitation of atoms in, 171-172
electrophoresis in, 220-224 Atomization
extraction, 198-207, see also atomic absorption and, 168—
Extraction 171
recrystallization, 196 vapor generation methods for,
"Aqua regia", for sample 190,194
preparation, 39 Atomizers, alternative, 187-191
Arc emission spectrography, Atoms, excitation of, 171-172
192-194 AutoAnalyzer, 319-325
Ascending chromatography, Automatic samplers, 319,321
214 Automatic titrators, 303,319,
Assay, defined, 6 320
Atomic absorption (AA), 167, Automation, 319-328
194, see also Atomic flow injection methods, 324,
spectroscopy 326
applications of, 1 6 ,1 8 3 -1 8 7 robotics, 326-328
atomization and, 168-171 segmented flow methods,
continuous spectrum, 172 320-325
detection limits of, 186 titration, 3 0 3 ,3 1 9 ,3 2 0
double-beam, 182,183 Auxiliary balances, 20-21
flame technique of, 178-187 Auxiliary electrode, 281,304
fuels and oxidants in, 168-169 Auxochrome, 126
graphite furnace, 176,187-190, Average deviation, defined, 334
194
instrumentation for, 179-183
interferences in, 184-185 Background current, 309-312
line spectrum, 172 Back titration, 76-77
matrix matching in, 185 Balances, 18-24
nebulizer in, 170 analytical, 22-24
safety of, maintenance and, auxiliary, 20-21
185-186 basic concept of, 18-20
sensitivity of, 186 instructions for use of, 22-24
Atomic emission spectrogra less accurate, 20-21
phy, 1 6 ,1 6 7 ,1 9 2 -1 9 4 multiple-beam, 20,21
Atomic fluorescence, 167,193, single-pan, 18-20
194 top-loading, 20-21
Atomic spectroscopy, 167-194, Baseline drift
see also Atomic absorption in gas chromatography, 258
(AA); Atomic emission in HPLC, 286
spectrography; Flame Bases
photometry; Inductively densities of, 55
coupled plasma (ICP) equivalent weights of, 57-58
Index 343
molarities of, 50 eter and, 120

percent composition of, 55 in flame atomic absorption,
for sample preparation, 39 181-182
Bathochromic shift, 126 Chromatogram
Beer-Lambert Law, see Beer's in gas chromatography, 23 9 -
Law 242
Beer's Law, 126-127 in HPLC, 274
deviations from, 128-129 Chromatograph, gas, 227,228
flame atomic absorption and, Chromatography, 207-220
179,183 adsorption, see Adsorption
Beer's Law Plot, 127,128 chromatography
Benzene, in extraction, 41 applications of, 16
Bipotentiometry, 80-83 ascending, 214
"Blank" solution, 118 bonded-phase, 210,261
Blow-out, pipet, 28 column methods of, 214
Bonded-phase chromatography configurations of, 214-220
(BPC), 210,261, see also descending, 214
High performance liquid gas, see Gas chromatography
chromatography (HPLC) (GC)
Buffer solutions, 6 5 -6 7 ,2 9 8 HPLC, see High performance
Burets, 31-34 liquid chromatography
Burners, atomic spectroscopy, (HPLC)
169-171 instrumental, 220
ion, see Ion-exchange chroma
tography
Calgamite, in water hardness ion-exchange, see Ion-ex
analysis, 79 change chromatography
Calibration, spectrophotometer, mobile phase of, 208
118-119 open-column, 218-220
Capacity factor paper, 215-218
in gas chromatography, 242 partition, see Partition chroma
in HPLC, 274 tography
Capillary columns, in gas planar methods of, 214
chromatography, 229,232, radial, 214
234 size-exclusion, see Size-
Carrier gas, 229,2 3 8 -2 3 9 exclusion chromatography
Cells, in electroanalytical stationary phase of, 208
methods, 287-288 thin-layer, 215-218
Charging current, 309-312 types of, 208-214
Chemical analysis, defined, 5 Chromophore, 126
Chemical shifts, in NMR Chromosorb, 234
spectroscopy, 159-160 Class A glassware, 2 5 ,2 8
Chloroform, in extraction, 41-42 Cleaning, glassware, 32-35
Chopper Cleaning solutions, safety and,
double-beam spectrophotom 35
Colorimeter, segmented flow residual, 309-312

and, 321,322 Cuvette, 116-117
Colorimetry, 112 Cyclic voltammetry, 316-318
Columns, see also specific type
gas chromatography, 230-236
temperature of, 236-238 Data
HPLC, 269-274 instrumental, 86-93
Combination pH electrode, reporting of, 331
298-299 Data acquisition, computers in,
Complexation, precipitation 98-99
and, equivalent weight Data analysis, 330
and, 60-61 Data display, 330
Complex ion, in water hardness Data handling, 329-332
analysis, 78-80 state of the art of, 331-332
Computers storage and, computers in, 9 9 -
in data acquisition, 98-99 100,331-332
in data manipulation and Data management, 331
storage, 9 9 -1 0 0 ,3 3 1 -3 3 2 Data stations, 330,331
in experiment control, 100-101 Debubbler, in segmented flow
in graph plotting, 100 methods, 322,324
Computing integrators, 331 Degassing, HPLC and, 265-266
in gas chromatography, 251, Deionized water, 197n, 280
252 Delves Cup, 191
Concentrated acids and bases, Demountable cell, 131
see Acids; Bases Density of concentrated acids
Concentration, instrument and bases, 55
readout and, 87-90 Descending chromatography,
Concentrators, 205 214
Conductivity detectors, 279-280 Detectors
Confidence interval, 336 flame atomic absorption, 182-
Continuous spectrum, atomic 183
absorption and, 172 gas chromatography, 242-249
Control sample, accuracy and, 4 HPLC, 275-283
Correlation chart, IR spectrom U V /v is spectrophotometry,
etry, 145,152 117-118
Correlation coefficient, 92-93 Determinate errors, 4 ,3 3 2 , see
Countercurrent distribution, also Error analysis
203-204 Deviation, defined, 334
Counter electrode, see Auxiliary Diethyl ether, in extraction, 41
electrode Differential pulse polarogra
Current phy, 312-314
background, 309-312 Diffraction grating, monochro
diffusion, 309-312 mator and, 115,116
Faradaic, 312 Diffuse reflectance, 135,139
non-Faradaic, 311,312 Diffusion current, 309-312
Index 345
Dilution, 4 8 -4 9 ,9 3 -9 4 in amperometric detector, 281

Dilution factor, 96-98 auxiliary, 281,304
Diode array detector, 276 combination pH, 298-299
Dispersing element, monochro dropping mercury, 306,307
mator, 115 hanging mercury drop, 315
Disposable pipets, 29 indicator, 281, 293,296-301
Distillation, 196-198,241 ion-selective, 288,299-301
Distribution coefficient, 202 pH, 296-298
Distribution ratio, 202-203 in polarography, 305-306
DME (dropping mercury reference, 281 ,2 9 1 -2 9 6 ,3 0 3 ,
electrode), 306 ,3 0 7 304
Double-beam atomic absorp saturated calomel reference,
tion, 182,183 291-293
Double-beam dispersive IR silver-silver chloride, 294-295
spectrometer, 139,141-142 standard hydrogen, 295-296
"Double-beam in space" in three-electrode system, 304
instrument, 139 in voltammetry, 315
"Double-beam in time" instru working, 304
ment, 139 Electrolyte, supporting, 303-305
Double-beam U V /v is spectro Electrolytic cell, 288,303, see also
photometer, 113,119-122 Polarography;
Dropping m ercury electrode Voltammetry
(DME), 3 0 6 ,3 0 7 Electrolytic conductivity
Duopettes, 30 detector, 247
Dynodes, photomultiplier tube, Electromagnetic radiation, 104
117 Electromagnetic spectrum, 104,
105,107
Electron capture detector
ECD (electron capture detector), (ECD), 245-246
245-246 Electronic analytical balances,
EDTA, see Ethylenediamine- 22-24, see also Balances
tetraacetic acid Electronic energy transitions,
Electroanalytical methods, 2 8 7 - 109
318, see also specific Electrophoresis, 220-224
methods Elution, in HPLC, 266-268
applications of, 16 Emission of light, 110-111
polarography, 303-315 Emission spectrography, see
potentiometry, 288-303 Atomic emission spectrog
voltammetry, 3 0 3 -3 0 5 ,3 1 5 - raphy
318 Emission spectroscopy, 174
Electrochemical detectors, 2 7 9 - End point, in titration, 72
282 Energy of light, 106,107
Electrodeless discharge lamp, Energy transitions
180-181 electronic, 109
Electrodes rotational, 109-110
vibrational, 109-110,129 Filtering, HPLC and, 264-265

Equipment, 17-46, see also Fingerprint region, IR spectra,
specific type 1 44-145,147
Equivalence point, in titration, Flame atomic absorption, 178-
71 187,194, see also Atomic
Equivalent, 5 7 -6 1 ,7 3 absorption (AA)
Equivalent weight, 57 alternative atomizers vs, 187-
acid/base, 57-58 191
precipitation / complexa tion, applications of, 183-187
60-61 instrumentation for, 179-183
redox, 59-60 Flame ionization detector (FID),
Eriochrome black T, in water 243-245
hardness analysis, 78-79 Flame photometer, 175-176,246
Error analysis, 4 ,3 3 2 -3 3 9 Flame photometric detector
assistance with, 338 (FPD), 246-247
quality control and, 338 Flame photometry, 1 6 7 ,1 7 2 -
random errors in, distribution 178,194
of, 335-337 applications of, 176-178
Student's t in, 336-337 instrumentation for, 175-176
terminology of, 333-335 Flame test, 173,174
Ethylenediamine-tetraacetic Flashback
acid (EDTA) avoidance of, 186
in complexation reaction, 60, premix burners and, 170-171
61 Flask, volumetric, see Volumet
in water hardness analysis, ric flask
78-80 Flow injection analysis (FIA),
Evaporator, rotary, 205 319, 324, 326
Excitation, in atomic spectros Fluorescence, 110-1 1 1 ,1 5 1 ,1 5 3
copy, 171-172 atomic, 1 6 7 ,193,194
Experiment control, computers Fluorescence detector, 277
in, 100-101 Fluorometry, 111, 151,153-155
Extinction coefficient applications of, 16
Beer's Law and, 126 Formality, molarity and, 54-56
molar, Beer's Law and, 127 Fourier Transform Infrared
Extraction, 198-207 Spectrometer (FTIR), 135,
liquid-liquid, 198-205 1 3 9 ,1 4 2 -1 4 4 ,2 8 2
liquid-solid, 205-207 FPD (flame photometric
sample preparation by, 39-42 detector), 246-247
Extraction theory, 200-203 Fractional distillation, 197-198
Fraction collector, in open-
column chromatography,
Faradaic current, 312 218-219
FIA (flow injection analysis), Frequency, of light, 104-105,
3 1 9 ,3 2 4 ,3 2 6 107
FID (flame ionization detector), Fronting, in gas chromatogra
243-245 phy, 257-258
Index 347
FTIR, see Fourier Transform Gas chromatography-infrared

Infrared Spectrometer spectrometry (GC-IR),
Fuels, atomic absorption and, 247-248
168-169 Gas chromatography-mass
Funnel, separatory, 3 9 ,1 9 9 -2 0 0 spectrometry (GC-MS),
Fusion, 42 165, 247-248
Gas-liquid chromatography
(GLC), 210-211,225, 226,
Galvanic cell, 287-28 8 ,3 0 7 , see see also Gas chromatogra
also Potentiometry phy (GC)
Gas chromatograph, 227,228 Gas-solid chromatography
Gas chromatography (GC), 207, (GSC), 225, see also Gas
2 0 8 ,2 1 4 ,2 2 0 ,2 2 5 -2 5 9 chromatography (GC)
altered retention times in, 258 Gaussian distribution, 336
applications of, 16 GC, see Gas chromatography
baseline drift in, 258 GC-IR, see Gas chromatogra
baseline perturbations in, 259 phy-infrared spectrometry
carrier gas flow rate in, 2 3 8 - GC-MS, see Gas chromatogra-
239 phy-mass spectrometry
chromatogram in, 239-242 Gel electrophoresis, 221-223
columns in, 230-236 Gel-filtration chromatography
temperature of, 236-238 (GFC), see Size-exclusion
detectors in, 242-249 chromatography
selectivity of, 242 Gel-permeation chromatogra
sensitivity of, 242 phy (GPC), see Size-
diminished peak size in, 257 exclusion chromatography
HPLC vs, 262-263 Glass electrode, 296-298
instruments for, design of, Glassware, 24-36, see also
226-228 specific type
qualitative analysis with, 2 4 9 - burets, 31-34
250 Class A, 2 5 ,2 8
quantitative analysis with, cleaning of, 32-35
250-256 pipets, 27-30
internal standard method in, pipetting devices, 30-32
256 storage of, 35-36
peak size measurement in, volumetric flask, 24-27
250-252 GLC, see Gas-liquid chromatog
response factor method in, raphy
253 ,2 5 5 -2 5 6 GPC (gel-permeation chroma
sample injection in, 228-230 tography), see Size-
stationary phase of, 234-236 exclusion chromatography
troubleshooting, 257-259 Gradient elution, 266-268
unexpected peaks appearing Gradient programmer, in
in, 259 HPLC, 266-268
unsymmetrical peak shapes Graphite furnace atomic
in, 257-258
absorption, 176,187-190, 264-266

194 gas chromatography vs, 2 6 2 -
Graphs, plotting of, computers 263
in, 100 ion-exchange, 271-272
Gravimetric analysis, 1 6 ,6 9 -7 0 normal-phase, 270
Gravimetric factor, 6 3 ,6 4 qualitative analysis with, 283
GSC, see Gas-solid chromatog quantitative analysis with,
raphy 283-284
Guard column, in HPLC, 264 recorder zero vs detector zero
in, 285
reverse-phase, 270-271
Half-cells, in electroanalytical sample and mobile phase
methods, 287 pretreatment in, 264-266
Half-reactions, 287 sample injection in, 268-269
Half-wave potential, 309-310 size-exclusion, 272
Half-width method, in gas solvent delivery in, 266-268
chromatography, 251 system leaks in, 284-285
Hall detector, 247 troubleshooting, 284-286
Hanging mercury drop elec unusually high pressure in,
trode (HME), 315 284
Height equivalent to a theoreti unusually low pressure in, 284
cal plate (HETP), 198 High performance liquid
in gas chromatography, 2 3 9 - chromatography-mass
242 spectrometer (HPLC-MS),
in HPLC, 274 165
Henderson-Hasselbalch equa HME (hanging mercury drop
tion, 66 electrode), 315
n-Hexane, in extraction, 40 Hollow cathode lamp, 179-181
High performance liquid HPLC, see High performance
chromatography (HPLC), liquid chromatography
2 0 7 ,2 2 0 ,2 6 1 -2 8 6 Hydrobromic acid
adsorption, 271,274 density of, 55
air bubbles in, 285 molarity of, 50
applications of, 16 percent composition of, 55
baseline drift in, 286 Hydrochloric acid
basic concepts of, 261-262 density of, 55
chromatogram in, 274 molarity of, 50
column "channeling" in, 285 percent composition of, 55
column selection for, 269-274 for sample preparation, 36-37
decreased retention time in, Hydrofluoric acid
285-286 density of, 55
detectors in, 275-283 molarity of, 50
filtering and degassing in, percent composition of, 55
Index 349
for sample preparation, 38 quantitative analysis with,

Hydrogen electrode, standard, 1 4 4 ,1 4 7 ,1 5 1 ,1 5 2
295-296 solid sampling with, 133-139
Hyperchromic effect, 126 diffuse reflectance in, 135,
Hypsochromic effect, 126 139
KBr pellet in, 134-135,138
Nujol mull in, 135,140
ICP, see Inductively coupled Injection port, in gas chroma
plasma tography, 228
Indeterminate (random) errors, Instrumental chromatography,
4 ,3 3 2 -3 3 3 , see also Error 220
analysis Instrumental methods, 85-101
distribution of, 335-337 calculations in, sample pre
Indicator, in titration, 71, see also treatment effect on, 96-98
specific indicators computers in, 98-101
Indicator electrodes, 296-301 data and readout in, 86-93
in amperometric detector, 281 concentration and, 87-90
combination pH, 298-299 correlation coefficient and,
ion-selective, 288,299-301 92-93
pH, 296-298 least squares method and,
SCE and, 293 90-92
Inductively coupled plasma recorders and, 86-87
(ICP), 1 6 7 ,1 7 4 ,1 7 6 ,1 9 1 - for quantitative analysis, 9 3 -
192,194 95
Infrared (IR) light, 106 Instruments, 17-46, see also
Infrared (IR) spectra specific instrument;
fingerprint region of, 144-145, specific method
147 Integration, peak splitting and,
peak ID region of, 145,147 in NMR spectroscopy,
Infrared (IR) spectrometry, 129- 160-162
151, see also Gas chroma Interferences
tography-infrared spec in atomic absorption, 184-185
trometry (GC-IR); Liquid in U V /vis spectrophotometry,
chromatography-infrared 128-129
spectrometry Interferogram, 143
applications of, 16 Interferometer, 142-143
correlation chart for, 145,152 Internal standard method, 94,
FTIR, 1 3 5 ,1 3 9 ,1 4 2 -1 4 4 256
instruments for, design of, Ion chromatography, see Ion-
139-144 exchange chromatography
liquid sampling with, 131-133, Ion-exchange chromatography,
136 212-213,261, see also High
qualitative analysis with, 144- performance liquid
150 chromatography (HPLC)
columns for, 271-272 wave theory of, 103-104

conductivity detectors in, 2 7 9 - Line spectrum
280 atomic absorption and, 172
Ion-selective electrode, 288, flame photometry and, 173
299-301 Liquid chromatography (LC),
IR light, see Infrared entries 208, see also Fligh perfor
Isocratic elution, 266-267 mance liquid chromatog
Isoelectric focusing, 223-224 raphy (HPLC)
Isoelectric point, 224 Liquid chromatography-
infrared spectrometry, 282
Liquid chromatography-mass
Karl Fischer titration, 7 4 ,8 0 -8 3 spectrometry, 282-283
KBr pellet technique, 134-135, Liquid-liquid extraction, 198-
138 205, see also Extraction
Kjeldahl method, 38, 75-77 concentrators and, 205
Kuderna-Danish evaporative countercurrent distribution
concentrator, 205 and, 203-204
separatory funnel in, 199-200
Laboratory safety, see Safety theory of, 200-203
Laboratory work Liquid-liquid partition chroma
planning of, 12-16 tography, 209,261, see also
tools for, 17-46, see also High performance liquid
specific tools chromatography (HPLC)
Lambda pipet, 29-30 Liquid sampling, IR spectrom
LC, see Liquid chromatography etry in, 131-133,136
Least squares method, 90-92 Liquid-solid extraction, 2 0 5 -
Ligand, defined, 78, see also 207, see also Extraction
specific ligands Loop injector, in HPLC, 268-269
Light
absorption of, 108-110
dual nature of, 103-104 Magnetic resonance, see
emission of, 110-111 Nuclear magnetic reso
energy of, 106,107 nance (NMR) spectros
frequency of, 1 04-105,107 copy
monochromatic, 112 Magnetic sector mass spectrom
nature of, parameters and, eter, 163,164
103-107 Maintenance, atomic absorption
particle theory of, 104 a n d ,186
polychromatic, 112 Manganese, in steel, spectro
source of, U V /vis spectropho photometric determina
tometers and, 113-114 tion of, 13-15
speed of, 104-105 Manifold, segmented flow and,
wavelength of, 104,107 321
energy and, 106 Mass spectra, 164-165
wave number of, 105-107 Mass spectrometers, 163,164
Index 351
HPLC-MS, 165 103-107

Mass spectrometry, 162-165, NMR, 155-162, see also
282, see also Gas chroma- Nuclear magnetic reso
tography-mass spectrom nance (NMR) spectros
etry (GC-MS); Liquid copy
chromatography-mass techniques and instruments
spectrometry used in, 111-112
instruments for, design of, Monochromatic light, 112
163-164 Monochromator
mass spectra and, 164-165 flame atomic absorption, 182-
Material Safety Data Sheet 183
(MSDS), 45 flame photometry, 175-176
Matrix matching, 9 5,185 U V /vis spectrophotometry,
McReynolds Constants, 236 114-116
Mean, defined, 333 MSDS (Material Safety Data
Measurement, 6 -9 Sheet), 45
Measuring pipets, 28-30 Multielement hollow cathode
Median, defined, 333 lamp, 180
Membrane potential, pH Multiple-beam balances, 20,21
electrode and, 296 Murexide, in water hardness
Methanol, in extraction, 41 analysis, 79
Methylene chloride, in extrac
tion, 40
Mode, defined, 333 Neat liquids, IR spectrometry
Mohr pipet, 28-30 and, 131
Molar absorptivity, Beer's Law Nebulizer, in atomic absorp
and, 127 tion, 170
Molar extinction coefficient, Nernst Equation, 288-291
Beer's Law and, 127 Nitric acid
Molarity, 54-56 density of, 55
of acids, 4 9 ,5 0 molarity of, 50
of bases, 50 percent composition of, 55
Molecular absorption spectrum, for sample preparation, 38
113 Nitrogen/phosphorus detector
double-beam spectrophotom (NPD), 246
eter and, 120 NMR spectroscopy, see Nuclear
qualitative analysis and, 123 magnetic resonance
Molecular spectroscopy, 103- spectroscopy
165, see also Fluorometry; Non-Faradaic current, 311,312
Spectrometry; Spectropho Normal distribution curve, 3 3 5 -
tometry 337
light and Normality, 56-62
absorption and emission of, equivalent and, 57-61
108-111 solution preparation and, 6 1 -
nature and parameters of, 62
Normal-phase high perfor chromatography, 250-252,

mance liquid chromatog 257
raphy, 270 Peak splitting, integration and,
NPD (nitrogen/phosphorus in NMR spectroscopy,
detector), 246 160-162
Nuclear magnetic resonance Percent
(NMR) spectroscopy, 155- solution preparation and, 4 9 -
162 54
chemical shifts in, 159-160 volum e/volum e, 50-51
peak splitting and integration w eight/volum e, 52-54
in, 160-162 w eight/w eight, 51-52
traditional instrument for, Percent composition of concen
158-159 trated acids and bases, 55
Nujol mull, 135,140 Percent transmittance, in U V /
vis spectrophotometry,
117,119,122
Open-column chromatography, Perchloric acid
218-220 density of, 55
Open tubular columns, in gas molarity of, 50
chromatography, 233 percent composition of, 55
Ostwald-Folin pipet, 30 for sample preparation, 38
Oxidants, atomic absorption Peristaltic proportioning pump,
and, 168-169 321
pH, see also Acids; Bases; Buffer
solutions
Packed columns, in gas chroma measurement of, 288
tography, 232-234 pH electrode in, 296
Paper chromatography, 215-218 Phase elution methods, 266-268
Paper electrophoresis, 221-222 pH electrode, 296-299
Particle theory of light, 104 Phosphoric acid
Partition chromatography, 2 0 8 - density of, 55
2 11,274 molarity of, 50
columns for, 270-271 percent composition of, 55
liquid-liquid, 261, see also High Photocathode, photomultiplier
performance liquid tube, 117
chromatography (HPLC) Photoionization detector (PID),
Partition coefficient, 202 248
Parts per billion (ppb), 62-65 Photometer, flame, 175-176,246
Parts per million (ppm), 62-65 Photometry, flame, see Flame
Pathlength, Beer's Law and, 126 photometry
Peak height method, in gas Photomultiplier tube, 1 1 2 ,1 1 7 -
chromatography, 250-251 118
Peak ID region, IR spectra, 145, PID (photoionization detector),
147 248
Peak size measurement, in gas Pipets, 27-30
Index 353
blow-out, 28 Precision, accuracy vs, 3-5

disposable, 29 Premix burner, 169-171
duopette, 30 Preparative columns, in gas
lambda, 29-30 chromatography, 233
measuring, 28-30 Primary lines
Mohr, 28-30 flame atomic absorption and,
Ostwald-Folin, 30 179,185
serological, 28-30 monochromator, 175
transfer, see Volumetric pipet Primary standard, 72-74
volumetric, see Volumetric Prism, monochromator and,
pipet 115,116
Pipetting devices, 30-32 Proportioning pump, peristal
Planck's Constant, 106 tic, 321
Planning, sampling and, 10-16 Proton Magnetic Resonance, see
Plasma source, ICP and, 191 Nuclear magnetic reso
Polarograph, 308 nance (NMR) spectros
Polarographic maxima, 312 copy
Polarography, 303-315 Pulse polarography, 312-314
classical, 306,309-313 Pumps
differential pulse, 312-314 proportioning, peristaltic, 321
modern techniques of, 312-314 in solvent delivery, in PIPLC,
pulse, 312-314 266
quantitative analysis with,
314-315
sampled dc, 312,313 Quadrupole mass spectrometer,
Polychromatic light, 112 163,164
Potential, 288, see also Qualitative analysis
Potentiometry; specific defined, 5 -6
type gas chromatography in, 2 4 9 -
Potentiometric titrations, 288, 250
302-303 HPLC in, 283
Potentiometry, 288-303 IR spectrometry in, 144-150
indicator electrodes in, 2 9 6 - planning of, 12-13
301, see also Indicator U V /vis spectrophotometry in,
electrodes 122-126,128-129
Nernst Equation and, 288-291 Quality assurance, for reagents,
reference electrodes in, 2 9 1 - 42-44
296, see also Reference Quality assurance program,
electrodes establishment of, assis
Potentiostat, 308 tance in, 338
ppb (parts per billion), 62-65 Quality control, statistics in, 338
ppm (parts per million), 62-65 Quality control charts, 338,339
Precipitation/complexation, Quantification, defined, 6
equivalent weight and, Quantitation, defined, 6
60-61 Quantitative analysis
defined, 6 Refractometry, 105n

gas chromatography in, 2 5 0 - Relative deviation, defined, 334
256 Relative retention, in gas
HPLC in, 283-284 chromatography, 249-250
instrumental methods for, 9 3 - Relative standard deviation,
95 defined, 335
IR spectrometry in, 144,147, Representative sample, 10-11
151,152 Residual current, 309-312
planning of, 13 Resolution, in chromatogram
polarography in, 314-315 in gas chromatography, 2 3 9 -
U V /v is spectrophotometry in, 241
126-129 in HPLC, 274
Quantitative transfer, defined, 6 Response factor method, in gas
chromatography, 253,
255-256
Radial chromatography, 214 Retention time
Radiation, electromagnetic, 104 in gas chromatography, 226,
Random errors, see Indetermi 239,240, 249-250
nate errors altered, 258
Readout in HPLC, 274,283
flame atomic absorption, 182- decreased, 285-286
183 Reverse-phase high-perfor
instrumental, 86-93 mance liquid chromatog
U V /v is spectrophotometry, raphy, 270-271
117-118 Rf factors, in chromatography,
Reagents, 3 6 -3 9 ,4 2 -4 4 217-218
Reciprocating piston pump, in Right angle configuration, in
HPLC, 266,267 fluorometry, 153
Recorders, 86-87 Robotics, 319,326-328
strip-chart, 86-87 Rotary evaporator, 205
x-y, 86-87 Rotational energy transitions,
Recrystallization, 196 109-110
Redox, equivalent weight, 5 9 -
60
Reference electrodes, 291-296, Safety, 45-46
303 atomic absorption and, 185-
in amperometric detector, 281 186
SCE, 291-293 cleaning solutions and, 35
silver-silver chloride, 294-295 Salt bridge, in saturated calomel
standard hydrogen, 295-296 reference electrode, 292
in three-electrode system, 304 Sample
Reflectance, diffuse, 135,139 control, accuracy and, 4
Refractive index, 105n, 277-278 obtainment of, 10-11
Refractive index detector, 2 7 7 - pretreatment of
279 effect on calculations of, 9 6 -
Refractometer, 278 98
Index 355
HPLC and, 264-266 Separatory funnel, 3 9 ,1 99-200

representative, 10-11 Serial dilution method, 93-94
Sample "clean-up", open- Serological pipet, 28-30
column chromatography Significant figures, rules for, 7 -
and, 219-220 9
Sample compartment, in U V / Silver-silver chloride electrode,
vis spectrophotometer, 294-295
116-117 Single-beam spectrophotom
Sampled dc polarography, 312, eter, 113,118-119
313 Single-pan balance, 18-20
Sample handling, 12 Size-exclusion chromatography,
Sample injection 2 1 3 -2 1 4 ,2 6 1 ,2 7 2 -2 7 3 , see
in gas chromatography, 2 2 8 - also High performance
230 liquid chromatography
in HPLC, 268-269 (HPLC)
Sample preparation Slit, monochromator, 115
bases in, 39 Snyder column, 205
concentrated acids in, 36-39 Sodium hydroxide, for sample
by extraction, 39-42 preparation, 39
by fusion, 42 Solid, analyte extraction from,
methods for, 39-42 205-207
reagents for, 3 6 -3 9 ,4 2 -4 4 Solid sampling, IR spectrometry
water in, 36 in, 133-139
Sample preservation, 12 Solubility rules, 37
Samplers, automatic, 319,321 Solution preparation, 47-67
Sampling buffer, 65-67
liquid, IR spectrometry in, dilution in, 48-49
1 3 1 -133,136 molarity and formality in, 5 4 -
planning and, 10-16 56
solid, IR spectrometry in, 133- normality in, 56-62
139 parts per million/billion and,
Saturated calomel reference 62-65
electrode (SCE), 291-293 percent in, 49-54
Sealed cell, 131 Solutions
Sealed demountable cell, 131, "blank", 118
132 standard, 72
Secondary lines series of, 93-94
atomic absorption and, 185 Solvent extraction, see Liquid-
monochromator, 175 liquid extraction
Segmented flow methods, 3 2 0 - Solvents, see also Sample
325 preparation; specific
Selectivity solvents
in gas chromatography, 242 delivery of, in HPLC, 266-268
in HPLC, 274 strength of, phase elution and,
Separations, see Analytical 268
separations Soxhlet extractor, 4 0 ,2 0 6 -2 0 7
Spark emission spectrography, 72-74

192-194 Standard reduction potentials,
Spectrography, atomic emis 2 89,290
sion, see Atomic emission Standard solutions, 7 2 ,9 3 -9 4
spectrography Statistics
Spectrometer, 112 errors and, 4, see also Error
FTIR, 135,139 analysis
IR, double-beam dispersive, in quality control, 338,339
139,141-142 Steel, manganese in, spectro-
mass photometric determina
magnetic sector, 163,164 tion of, 13-15
quadrupole, 163,164 Storage, glassware, 35-36
Spectrometry, 112, see also Strip-chart recorder, 86-87
Spectrophotometry Stripping voltammetry, 316, 317
IR, 129-151, see also Infrared Student's t, 336-337
(IR) spectrometry Sulfuric acid
mass, 162-165 density of, 55
Spectrophotometer, 112 molarity of, 50
double-beam, 113,119-122 percent composition of, 55
single-beam, 113 for sample preparation, 37-38
calibration of, 118-119 Supporting electrolyte, 303-305
Spectrophotometry, 112, see also Suppressors, conductivity
Spectrometry; U V /vis detectors and, 280
spectrophotometry Systematic errors, see Determi
applications of, 16 nate errors
in determination of manga
nese in steel, 13-15
Spectroscopy, 112 Tailing, in gas chromatography,
atomic, 167-194, see also 257-258
Atomic spectroscopy TCD (thermal conductivity
emission, 174 detector), 243
molecular, 103-165, see also TC imprint
Molecular spectroscopy; on lambda pipet, 29-30
specific type on volumetric flask, 25
Speed of light, 104-105 TD imprint
Standard, primary, 72-74 on pipet, 2 7 ,3 0
Standard additions method, 9 4 - on volumetric flask, 25
9 5 ,1 8 5 Technicon AutoAnalyzer, see
Standard curve, 127,128 AutoAnalyzer
Standard deviation, defined, Temperature programming, in
335 gas chromatography, 237
Standard hydrogen electrode, Terminology, 5 -6
295-296 Tesla coil, in ICP, 191
Standardization of solutions, Theoretical plates, 198
Index 357
in chromatogram 118,122
in gas chromatography, 2 39- Triangulation method, in gas
241 chromatography, 251
in HPLC, 274 t values, 336-337
Thermal conductivity detector
(TCD), 243
Thin-layer chromatography Unknowns, titration of, 74-75
(TLC), 215-218 UV absorption detector, for
Three-electrode system, 304 HPLC, 275-276
Titer, 74 UV light, 106
Titrand, 71 UV / vis spectrophotometry
Titrant, 71 applications of, 16
Titration, 70-72, see also Titri general description of, 112-113
metric analysis instruments for
amperometric, 317-318 calibration of, 118-119
automatic, 320, see also Titra design of, 113-122
tors, automatic detector/readout, 117-118
back, 76-77 light source and, 113-114
Karl Fischer, 7 4 ,8 0 -8 3 monochromator, 114-115
potentiometric, 288,302-303 sample compartment (cu
of unknowns, 74-75 vette), 116-117
Titrators, automatic, 303,319, interferences in, 128-129
320 qualitative analysis with, 122-
Titrimetric analysis, 70-75, see 126
also Titration interferences and deviations
applications of, 16 in, 128-129
Kjeldahl method in, see quantitative analysis with,
Kjeldahl method 122,126-128
real-world, examples of, 75-83 interferences and deviations
standardization in, 72-74 in, 128-129
unknowns in, 74-75
of water hardness, 74, 76-77
TLC (thin-layer chromatogra Vapor generation methods, for
phy), 215-218 atomization, 190,194
Toluene, in extraction, 41 Vapor pressure, gas chromatog
Top-loading balances, 20-21 raphy and, 226
Total consumption burner, 169- Vibrational energy transitions,
170 109-110,129
Transfer, quantitative, defined, Visible (vis) light, 106, see also
6 U V /vis spectrophotom-
Transfer pipet, see Volumetric etry
pipet Voltammetry, 303-305,315-318
Transmittance, in U V /vis amperometric titration in,
spectrophotometry, 117, 317-318
358 A nalytical Chem istry Refresher Manual
cyclic, 316-318
stripping, 316,317
Volumetric analysis, see Titri
metric analysis
Volumetric flask, 24-27,4 8
Volumetric pipet, 27-30
Volum e/volum e percent, 50-51
W ater
deionized, see Deionized water
distillation of, 197
for sample preparation, 36
W ater hardness, titrimetric
N O TES

Analytical Chemistry Refresher Manual (Kenkel, John)

Uploaded by

Document Informationclick to expand document information

Document Informationclick to expand document information

Copyright:

Available Formats

Analytical Chemistry Refresher Manual (Kenkel, John)

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Analytical Chemistry Refresher Manual (Kenkel, John)

Uploaded by

Copyright:

Available Formats

.

CRC Press is an imprint of the

©1992 by Taylor & Francis Group, LLC

First issued in paperback 2019

No claim to original U.S. Government works

ISBN-13: 978-0-367-45037-3 (pbk)

Visit the Taylor & Francis Web site at

and the CRC Press Web site at

Library of Congress Cataloging-in-Publication Data

Library of Congress Card Number 91-38547

This book is dedicated to my mother, Carmen Kenkel, who, follow­

John Kenkel has been on the chemistry faculty at Southeast Com­

Professor Kenkel, born and raised on a small farm in western Iowa,

His first book, "Analytical Chemistry For Technicians", published

He has won several awards for teaching excellence. In 1985, he was

This work is intended to be, as the title implies, a refresher in the

The Analytical Chemistry Refresher Manual would not have hap­

1.1 IMPORTANCE OF ANALYTICAL CHEMISTRY

We begin this refresher on analytical chemistry with the following

While Mr. Heylin's statement precedes an editorial on the critical

state of chemistry as a profession in today's world, it could just as

FIGURE 1.1 Illustration of numerous interactions that an analytical chemist can

1.2 PRECISION VS ACCURACY

Analytical chemistry is the science dealing with the determination

F IG U R E 1.2 Illustration of precision and accuracy. (Reprinted from Kenkel, J.,

The basic terminology associated with analytical chemistry and

Chemical Analysis This is the determination of the chemical

the need or desire to determine quantity of

1.4 FUNDAMENTALS OF MEASUREMENT

In the analytical chemistry laboratory, many measurements are

1. Any nonzero digit is significant. Example, 1.27 (three significant

To cover point number 2, the following rules apply:

1. The correct answer to a multiplication or division calculation

4.271 + 6.96 = 11.231 = 11.23

II.231 is the calculator answer, and 11.23 is the correctly rounded

(7 .2 7 - 4 .8 ) x 56.27 = 138.9869 = 1.4 x 102

138.9869 is the calculator answer, and 1.4 x 102 is the correctly

1.247 m x 100 cm /m = 124.7 cm (four significant figures)

6. In cases in which the logarithm of a number needs to be

This would appear to be an increase in the number of significant

1.5 SAMPLING AND PLANNING

1.5.1 Obtaining the Sample

A laboratory analysis is almost always meant to give a result which

1.5.2 Handling the Sample

1.5.3 Planning the Lab Work

Once the sampling and sample preservation schemes have been

analysis, as long as the analyte remains above the detection limit of

1.5.3a Exam ple A nalysis — The Spectrophotom etric D eterm ina­

Step 1. Calculate the volumes of 1000 ppm Mn required to prepare

1.5.3b Discussion o f Exam ple A nalysis

1. The measurement of the volumes of 1000 ppm Mn placed in the

affected if it is not quantitative. The 100-mL volumes must be

Another important aspect of "planning" is deciding on an approach

Table 1.1 A Summary of the Applications of the Major Analytical Techniques

Technique General Application

Gravimetric analysis Analytes separated by physical means and

Without some introduction to the general basic tools of an

2.2.1 The Basic Concept of the Balance

This book is dedicated to my mother, Carmen Kenkel, who, follow

John Kenkel has been on the chemistry faculty at Southeast Com

The Analytical Chemistry Refresher Manual would not have hap

1.5.3a Exam ple A nalysis — The Spectrophotom etric D eterm ina

The accurate measurement of volumes of solutions is a very impor

(Figure 2.7). Such a pipet is used to transfer unusually viscous solu

agents. It will be necessary to check the N a

(Chapter 4) is not planned, then its concentration must be known accu