Soil and Environmental Analysis: Modern Instrumental Techniques - Chapter 1 ppt

Soil and Environmental
Analysis
Modern Instrumental Techniques
Third
Edition
edited
by
Keith
A.
Smith
The University of Edinburgh
Edinburgh, Scotland
Malcolm
S.
Cresser
The University of York
York, England
MARCEL



MARCEL DEKKER,
INC.
DEKKER
TM
Copyright n 2004 by Marcel Dekker, Inc. All Rights Reserved.
The previous edition was Soil Analysis: Modern Instrumental Techniques, Second
Edition, K. A. Smith, ed. (Marcel Dekker, Inc., 1991).
Although great care has been taken to provide accurate and current information,
neither the author(s) nor the publisher, nor anyone else associated with this
publication, shall be liable for any loss, damage, or liability directly or indirectly

caused or alleged to be caused by this book. The material contained herein is not
intended to provide specific advice or recommendations for any specific situation.
Trademark notice: Product or corporate names may be trademarks or registered
trademarks and are used only for identification and explanation without intent to
infringe. Also, mention of these names does not imply any specific endorsement of
the companies or products.
Library of Congress Cataloging-in-Publication Data
A catalog record for this book is available from the Library of Congress.
ISBN: 0-8247-0991-8
This book is printed on acid-free paper.
Headquarters
Marcel Dekker, Inc., 270 Madison Avenue, New York, NY 10016, U.S.A.
tel: 212-696-9000; fax: 212-685-4540
Distribution and Customer Service
Marcel Dekker, Inc., Cimarron Road, Monticello, New York 12701, U.S.A.
tel: 800-228-1160; fax: 845-796-1772
Eastern Hemisphere Distribution
Marcel Dekker AG, Hutgasse 4, Postfach 812, CH-4001 Basel, Switzerland
tel: 41-61-260-6300; fax: 41-61-260-6333
World Wide Web

The publisher offers discounts on this book when ordered in bulk quantities. For
more information, write to Special Sales/Professional Marketing at the headquarters
address above.
Copyright ß 2004 by Marcel Dekker, Inc. All Rights Reserved.
Neither this book nor any part may be reproduced or transmitted in any form or by
any means, electronic or mechanical, including photocopying, microfilming, and
recording, or by any information storage and retrieval system, without permission
in writing from the publisher.
Current printing (last digit):

10987654321
PRINTED IN THE UNITED STATES OF AMERICA
TM
Copyright n 2004 by Marcel Dekker, Inc. All Rights Reserved.
BOOKS IN SOILS, PLANTS, AND THE ENVIRONMENT
Agricultural Engineering
Animal Science
Crops
Irrigation and Hydrology
Micro biology
Plants
Soils
Editorial Board
Robert M. Peart, University of Florida, Gainesville
Harold Hafs, Rutgers University, New :Brunswick,
New Jersey
Moharnmad Pessarakli, University of Axizona,
Tucson
Donald R. Nielsen, University of Califcmia, Davis
Jan
Dirk
van Elsas, Research Institute
for
Plant
Protection, Wageningen, The Netherlands
L.
David Kuykendall, U.S. Department of
Agriculture,
Beltsville, Maryland
Kenneth

B.
Marcum, Texas
A&M
University, El
Paso, Texas
Jean-Marc Bollag, Pennsylvania State University,
University Park, Pennsylvania
Tsuyoshi Miyazalu, University of Tokyo
Soil Biochemistry, Volume I,
edited by
A.
D. McLaren and G. H. Peterson
Soil Biochemistry, Volume
2,
edited by
A.
D. McLaren and J. SkujinS
Soil Biochemistry, Volume
3,
edited by
E.
A.
Paul and
A.
D.
Mcl-aren
Soil Biochemistry, Volume
4,
edited by
E.

A.
Paul and
A.
D.
Mcl-aren
Soil Biochemistry, Volume
5,
edited by
E.
A.
Paul and J.
N.
Ladld
Soil Biochemistry, Volume
6,
edited by Jean-Marc Bollag and
G.
Stotzky
Soil Biochemistry, Volume
7,
edited by
G.
Stotzky and Jean-Ma~rc Bollag
Soil Biochemistry, Volume
8,
edited by Jean-Marc Bollag and G. Stotzky
Soil Biochemistry, Volume
9,
edited by
G.

Stotzky and Jean-Ma1.c Bollag
Soil Biochemistry, Volume 10,
edited by Jean-Marc Bollag and
(2.
Stotzky
Organic Chemicals in the Soil Environment, Volumes
1
and
2,
edited by
C.
A.
I. Goring and J.
W.
Hamaker
Humic Substances in the Environment,
M. Schnitzer and S.
U.
Khan
Microbial Life in the Soil: An Introduction,
T. Hattori
Principles of Soil Chemistry,
Kim
H.
Tan
Soil Analysis: Instrumental Techniques and Related Procedufies,
edited by
Keith
A.
Smith

Soil Reclamation Processes: Microbiological Analyses and Applications,
edited by Robert
L.
Tate Ill and Donald
A.
Klein
Symbiotic Nitrogen Fixation Technology,
edited by Gerald
H.
Elkan
MARCEL
DEKKER,
INC.
270
Madison
Avenue, New
York,
New
York
10016
TM
Copyright n 2004 by Marcel Dekker, Inc. All Rights Reserved.
TM
Copyright n 2004 by Marcel Dekker, Inc. All Rights Reserved.
Soil-Water Interactions: Mechanisms and Applications, Shingo lwata and
Toshio Tabuchi with Benno P. Warkentin
Soil Analysis: Modern Instrumental Techniques, Second Edition, edited by
Keith A. Smith
Soil Analysis: Physical Methods, edited by Keith A. Smith and Chris
E.

Mullins
Growth and Mineral Nutrition of Field Crops, N. K. Fageria,
V.
C. Baligar, and
Charles Allan Jones
Semiarid Lands and Deserts: Soil Resource and Reclamation, edited by J.
SkujinS
Plant Roots: The Hidden Half, edited by Yoav Waisel, Amram Eshel, and Uzi
Kafkafi
Plant Biochemical Regulators, edited by Harold W. Gausman
Maximizing Crop Yields, N. K.
Fageria
Transgenic Plants: Fundamentals and Applications, edited by Andrew
H
iatt
Soil Microbial Ecology: Applications in Agricultural and Environmental
Management, edited by F. Blaine Metting, Jr.
Principles of Soil Chemistry: Second Edition, Kim
H.
Tan
Water Flow in Soils, edited by Tsuyoshi Miyazaki
Handbook of Plant and Crop Stress, edited by Mohammad Pessarakli
Genetic Improvement of Field Crops, edited by Gustavo A. Slafer
Agricultural Field Experiments: Design and Analysis, Roger G. Petersen
Environmental Soil Science, Kim
H.
Tan
Mechanisms of Plant Growth and Improved Productivity: Modern Ap-
proaches, edited by Amarjit S. Basra
Selenium in the Environment, edited by W. T. Frankenberger, Jr., and Sally

Benson
Plant-Environment Interactions, edited by Robert
E.
W il kinson
Handbook of Plant and Crop Physiology, edited by Mohammad Pessarakli
Handbook of Phytoalexin Metabolism and Action, edited by M. Daniel and R.
P. Purkayastha
Soil- Water Interactions: Mechanisms and Applications, Second Edition, Re-
vised and Expanded, Shingo Iwata, Toshio Tabuchi, and Benno P.
Warkentin
Stored-Grain Ecosystems, edited by Digvir S.
Jayas, Noel
D.
G.
White, and
William E. Muir
Agrochemicals from Natural Products, edited by C. R. A. Godfrey
Seed Development and Germination, edited by Jaime
Kigel and Gad Galili
Nitrogen Fertilization in the Environment, edited by Peter Edward Bacon
Phytohormones in Soils: Microbial Production and Function, William T.
Frankenberger, Jr., and Muhammad Arshad
Handbook of Weed Management Systems, edited by Albert
E.
Smith
Soil Sampling, Preparation, and Analysis, Kim H
.
Tan
Soil Erosion, Conservation, and Rehabilitation, edited by Menachem Agassi
Plant Roots: The Hidden Half, Second Edition, Revised and Expanded,

edited by Yoav Waisel, Amram Eshel, and Uzi Kafkafi
Photoassimilate Distribution in Plants and Crops: Source-Sink Relation-
ships, edited by Eli Zamski and Arthur A. Schaffer
MARCEL
DEKKER,
INC.
270
Madison Avenue, New York, New York
10016
TM
Copyright n 2004 by Marcel Dekker, Inc. All Rights Reserved.
TM
Copyright n 2004 by Marcel Dekker, Inc. All Rights Reserved.
Mass Spectrometry of Soils,
edited by Thomas W. Boutton and Shinichi
Yamasaki
Handbook of Photosynthesis,
edited by Mohammad Pessarakli
Chemical and Isotopic Groundwater Hydrology: The Applied Approach,
Second Edition, Revised and Expanded,
Emanuel Mazor
Fauna in Soil Ecosystems: Recycling Processes, Nutrient Fluxes, and Agri-
cultural Production,
edited by Gero Benckiser
Soil and Plant Analysis in Sustainable Agriculture and Environment,
edited
by Teresa Hood and J. Benton Jones, Jr.
Seeds Handbook: Biology, Production, Processing, and Storage,
B. B.
Desai, P. M. Kotecha, and D. K. Salunkhe

Modern Soil Microbiology,
edited by J. D. van Elsas, J. T. Trevors, and E. M.
H.
Wellington
Growth and Mineral Nutrition of Field Crops: Second Edition,
N.
K. Fageria,
V. C. Baligar, and Charles Allan Jones
Fungal Pathogenesis in Plants and Crops: Molecular Biology and Host
Defense Mechanisms,
P. Vidhyasekaran
Plant Pathogen Detection and Disease Diagnosis,
P. Narayanasamy
Agricultural Systems Modeling and Simulation,
edited by Robert M. Peart
and R. Bruce Curry
Agricultural Biotechnology,
edited by Arie Altman
Plant-Microbe Interactions and Biological Control,
edited by Greg J. Boland
and L. David Kuykendall
Handbook of Soil Conditioners: Substances That Enhance the Physical
Properties of Soil,
edited by Arthur Wallace and Richard
E.
Terry
Environmental Chemistry of Selenium,
edited by William T. Fran ken berger,
Jr., and Richard A. Engberg
Principles of Soil Chemistry: Third Edition, Revised and Expanded,

Kim
H.
Tan
Sulfur in the Environment,
edited by Douglas
G.
Maynard
Soil-Machine Interactions: A Finite Element Perspective,
edited by Jie Shen
and Radhey Lal Kushwaha
Mycotoxins in Agriculture and Food Safety,
edited by Kaushal K. Sinha and
Deepak Bhatnagar
Plant Amino Acids: Biochemistry and Biotechnology,
edited by Bijay K. Singh
Handbook of Functional Plant Ecology,
edited by Francisco I. Pugnaire and
Fernando Valladares
Handbook of Plant and Crop Stress: Second Edition, Revised and Ex-
panded,
edited by Mohammad Pessarakli
Plant Responses to Environmental Stresses: From Phytohormones to Ge-
nome Reorganization,
edited by
H.
R. Lerner
Handbook of Pest Management,
edited by John R. Ruberson
Environmental Soil Science: Second Edition, Revised and Expanded,
Kim

H.
Tan
Microbial Endophytes,
edited by Charles W. Bacon and James
F.
White, Jr.
Plant-Environment Interactions: Second Edition,
edited by Robert E. W
il-
kinson
Microbial Pest Control,
Sushi1 K. Khetan
MARCEL
DEKKER,
INC.
270
Madison Avenue, New York, New York
10016
TM
Copyright n 2004 by Marcel Dekker, Inc. All Rights Reserved.
TM
Copyright n 2004 by Marcel Dekker, Inc. All Rights Reserved.
Soil and Environmental Analysis: Physical Methods, Second Edition, Re-
vised and Expanded,
edited by Keith A. Smith and Chris E. Mullins
The Rhizosphere: Biochemistry and Organic Substances at the Soil-Plant
Interface,
Roberto Pinton, Zeno Varanini, and Paolo Nannipieri
Woody Plants and Woody Plant Management: Ecology, Safety, and Envi-
ronmental Impact,

Rodney W. Bovey
Metals in the Environment: Analysis by Biodiversity,
M. N. V. Prasad
Plant Pathogen Detection and Disease Diagnosis: Second Edition, Revised
and Expanded,
P. Narayanasamy
Handbook of Plant and Crop Physiology: Second Edition,
Revised and
Expanded,
edited by Mohammad Pessarakli
Environmental Chemistry of Arsenic,
edited by William T. Frankenberger, Jr.
Enzymes in the Environment: Activity, Ecology, and Applications,
edited by
Richard G. Burns and Richard P. Dick
Plant Roots: The Hidden Half, Third Edition, Revised and Expanded,
edited
by Yoav Waisel, Amram Eshel, and Uzi Kafkafi
Handbook of Plant Growth: pH as the Master Variable,
edited by Zdenko
Rengel
Biological Control of Crop Diseases,
edited by Samuel S. Gnanamanickam
Pesticides in Agriculture and the Environment,
edited by Willis B. Wheeler
Mathematical Models of Crop Growth and Yield,
Allen R. Overman and
Richard V. Scholtz Ill
Plant Biotechnology and Transgenic Plants,
edited by Kirsi-Marja Oksrnan-

Caldentey and Wolfgang H. Barz
Handbook of Posthatvest Technology: Cereals, Fruits, Vegetables, Tea, and
Spices,
edited by Amalendu Chakraverty, Arun S. Mujumdar,
G.
S.
Vijaya Raghavan, and Hosahalli S. Ramaswamy
Handbook of Soil Acidity,
edited by Zdenko Rengel
Humic Matter in Soil and the Environment: Principles and Controversies,
Kim
H. Tan
Molecular Host Resistance to Pests,
S. Sadasivam and B. Thayumanavan
Soil and Environmental Analysis: Modern Instrumental Techniques, Third
Edition,
edited by Keith A. Smith and Malcolm S. Cresser
Chemical and Isotopic Groundwater Hydrology: Third Edition,
Ema~nuel
Mazor
Additional Volumes in Preparation
Agricultural Systems Management: Optimizing Efficiency and Performance,
Robert M. Peart and Dean W. David Shoup
Seeds Handbook: Biology, Production, Processing, and Storage, Second
Edition, Revised and Expanded,
Babasaheb B. Desai
Physiology and Biotechnology Integration for Plant Breeding,
edited by
Henry T. Nguyen and Abraham Blum
MARCEL

DEKKER,
INC.
270
Madison Avenue, New York, New York
10016
TM
Copyright n 2004 by Marcel Dekker, Inc. All Rights Reserved.
TM
Copyright n 2004 by Marcel Dekker, Inc. All Rights Reserved.
Preface
This third edition retains the range of analytical techniques and the structure
of individual chapters of the second edition. However, there are some
significant changes, with the introduction of new topics and some deletions,
to take into account the changing priorities in environmental analysis.
The chapters on atomic absorption and flame emission spectrometry,
inductively coupled plasma spectrometry, continuous-flow and flow-
injection analysis, ion chromatography, combustion analyzers for carbon,
nitrogen and sulfur, an d x-ray fluorescence spectrometry have new or
additional authors and their applications sections cover a wider range of
environmental materials than before. This latter feature is one that generally
applies to this edition, so that although the most important application is
still the analysis of soils, much more attention is given to other materials.
Thus, for example, the chapter on CNS analyzers has been extended to
cover the important topic of the measurement of dissolved C and N in
waters, which involves the use of different instruments from those used for
soil analysis.
The coverage of ion-selective electrodes in the second edition has been
included in an extended chapter on electroanalytical methods. Similarly, the
previous chapter on isotope-ratio mass spectrometry has been extended to
cover isotopes of hydrogen, carbon, and sulfur, as well as those of nitrogen,

because of the importance of isotopic studies in current research into
environmental and biospheric processes involving these elements. The
previous chapters on nuclear and radiochemical analys is and instrumental
neutron activation analysis have been replaced by a new chapter on the
measurement of radioisotopes and ionizing radiation. The chapter on the
measurement of gases in the soil atmosphere has been replaced by two
chapters dealing with the measurement of gas fluxes between the land
surface and the atmosphere, reflecting the current concentration of research
effort on global warming. This expansion into a new interdisciplinary field,
taken together with other changes already mentioned, means that the
scientific coverage of the book extends to most of the techniques involved in
TM
Copyright n 2004 by Marcel Dekker, Inc. All Rights Reserved.
studies of the biogeochemical cycles of carbon, nitrogen, and sulfur, in the
context of improving our understanding of global change.
Contamination of soils, waters, and sediments with heavy metals
continues to present problems of a more localized nature, and several
chapters provide appropriate analytical techniques for investigating them.
The concern over organic pollutants has changed over the last few years
from a strong focus on persistent pesticides to worries about other
categories of organic compounds, such as PCBs and aromatic hydrocarbons.
The book reflects this change in that the chapter on pesticide analysis
has been replaced by one concentrating on these other contaminants.
This third edition is aimed at researchers working in soil science,
environmental chemistry, or ecological science, as well as scientists
operating analytical service laboratories with substantial throughputs of
soil, water, and other environmental samples. It provides information that
should help in method selection by those who need to undertake a new
determination as their projects develop. It will also guide those who are
considering replacing outdated or worn out equipment used for a particular

routine analytical task, either with a later model with new features (and
probably more ‘‘bells and whistles’’), or alternatively, with a new method of
instrumental analysis. In regard to selection of a new method, the book
should help in evaluating the techniques available, so that the optimal
choice, in terms of speed, cost, or sensitivity, may be selected. It will also be
useful to teachers and students of postgraduate courses in soil chemistry,
environmental chemistry, and soil and environmental analysis.
We wish to thank the contributors for their efforts, Mary Lightbody
for preparing the index, and Ann Pulido at Marcel Dekker, Inc., for
managing the editing process. We acknowledge the tolerance of colleagues,
families, and students who may have found us somewhat distracted from
other tasks from time to time, as this volume passed through its various
stages.
Keith A. Smith
Malcolm S. Cresser
TM
Copyright n 2004 by Marcel Dekker, Inc. All Rights Reserved.
Contents
Preface
Contributors
1.AtomicAbsorptionandFlameEmissionSpectrometry
Malcolm S. Cresser
2.InductivelyCoupledPlasmaSpectrometry
Stephen J. Hill, Andrew Fisher, and Mark Cave
3. Electroanalytical Methods in Environmental Chemical
Analysis
Iain L. Marr
4.Continuous-Flow,Flow-Injection,andDiscreteAnalysis
Anthony C. Edwards, Malcolm S. Cresser, Keith A. Smith,
and Albert Scott

5.IonChromatography
M. Ali Tabatabai, Nicholas T. Basta, and
Shreekant V. Karmarkar
6. Automated Instruments for the Determination of Total
Carbon,Hydrogen,Nitrogen,Sulfur,andOxygen
Keith A. Smith and M. Ali Tabatabai
7.X-RayFluorescenceAnalysis
Philip J. Potts
8.MeasurementofRadioisotopesandIonizingRadiation
Olivia J. Marsden and Francis R. Livens
9.StableIsotopeAnalysisandApplications
Charles M. Scrimgeour and David Robinson
TM
Copyright n 2004 by Marcel Dekker, Inc. All Rights Reserved.
10. Measurement of Trace Gases, I: Gas Analysis, Chamber
Methods,andRelatedProcedures
Keith A. Smith and Franz Conen
11. Measurement of Trace Gases, II: Micrometeor ological
MethodsatthePlot-to-LandscapeScale
John B. Moncrieff
12.AnalysisofOrganicPollutantsinEnvironmentalSamples
Julian J. C. Dawson, Helena Maciel, Graeme I. Paton, and
Kirk T. Semple
TM
Copyright n 2004 by Marcel Dekker, Inc. All Rights Reserved.
Contributors
Nicholas T. Basta Department of Plant and Soil Sciences, Oklahoma State
University, Stillwater, Oklahoma, U.S.A.
Mark Cave British Geological Survey, Nottingham, England
Franz Conen School of GeoSciences, The University of Edinburgh,

Edinburgh, Scotland
Malcolm S. Cresser Environment Department, The University of York,
York, England
Julian J. C. Dawson Department of Plant and Soil Science, The University
of Aberdeen, Aberdeen, Scotland
Anthony C. Edwards The Macaulay Institute, Aberdeen, Scotland
Andrew Fisher Advanced Environmental Diagnostics, The University of
Plymouth, Plymouth, England
Stephen J. Hill Advanced Environmental Diagnostics, The University of
Plymouth, Plymouth, England
Shreekant V. Karmarkar Lachat Instruments, Milwaukee, Wisconsin,
U.S.A.
Francis R. Livens Department of Chemistry, The University of
Manchester, Manchester, England
Helena Maciel Department of Plant and Soil Science, The University of
Aberdeen, Aberdeen, Scotland
Iain L. Marr Department of Chemistry, The University of Aberdeen,
Aberdeen, Scotland
Olivia J. Marsden Department of Chemistry, The University of
Manchester, Manchester, England
TM
Copyright n 2004 by Marcel Dekker, Inc. All Rights Reserved.
John B. Moncrieff School of GeoSciences, The University of Edinburgh,
Edinburgh, Scotland
Graeme I. Paton Department of Plant and Soil Science, The University of
Aberdeen, Aberdeen, Scotland
Philip J. Potts Department of Earth Sciences, The Open University,
Milton Keynes, England
David Robinson Department of Plant and Soil Science, The University of
Aberdeen, Aberdeen, Scotland

Albert Scott Scottish Agricultural College, Edinburgh, Scotland
Charles M. Scrimgeour The Scottish Crop Research Institut e, Dundee,
Scotland
Kirk T. Semple Environmental Science Division, Lancaster University,
Lancaster, England
Keith A. Smith School of GeoSciences, The University of Edinburgh,
Edinburgh, Scotland
M. Ali Tabatabai Department of Agronomy, Iowa State University,
Ames, Iowa, U.S.A.
TM
Copyright n 2004 by Marcel Dekker, Inc. All Rights Reserved.
1
Atomic Absorption and Flame
Emission Spectrometry
Malcolm S. Cresser
The University of York, York, England
I. INTRODUCTION
A. Interaction of Light with Atoms
This chapter is concerned with atomic absorption spectrometry (AAS),
which is a very widely used technique for the determination of over 20
elements in soils, plants, waters, and other environmental materials. It also
briefly covers flame emission spectrometry (FES), which is also widely used,
but for the determination of a smaller number of elements. For some
elements at very high concentrations, absorption of visible light by atoms
can be readily observed. Our sun’s spectrum, for example, shows several
dark absorption lines where the continuum emitted from the high-
temperature solar surface is selectively absorbed by free atoms of elements
such as sodium in the solar atmosphere. These dark lines, the Fraunhofer
lines, are perhaps the oldest and best-known example of atomic absorption.
Any particular electronic transition in an atom requires photons with

an appropriate amount of energy to induce a transition from a lower
discrete (quantized) energy state to a higher quantized energy state. If the
photons have insufficient energy (i.e., if the wavelength of the light is too
long), the transition cannot occur. Nor can a transition occur if the
wavelength is too short, because there is no mechanism by which the excess
energy can be absorbed. Atomic absorption spectra theref ore consist of
isolated, very narrow bands, or lines, with one line for each possible
electronic transition. This is why the atomic absorption bands of sodium in
the sun’s atmosphere are sharp lines.
TM
Copyright n 2004 by Marcel Dekker, Inc. All Rights Reserved.
Atoms may also be excited thermally in a body of hot gas such as
a flame or plasma. The thermally excited atoms may be subsequently
deactivated by losing their electronic excitation energy via conversion to
light energy, which is emitted in all directions. Measurement of the intensity
of the light emitted constitutes the basis of FES. Over a moderate range of
atom concentration, the intensity of the emitted light is proportional to the
number of excited atoms present in the flame.
B. Quantitative AAS and FES—What Do We Need to
Measure?
The wavelengths at which atomic absorption spectral lines occur are
characteristics of the particular atoms that are giving rise to the lines and
thus may be used for qualitative identification of the absorbing element(s).
For quantitative analysis, we need to measure some property that varies,
preferably linearly, with the concentration of the elements of interest. What
parameter should we measure if we wish to exploit atomic absorption
quantitatively?
Consider a narrow, monochromatic beam of photons (I
0
) passing

through a cloud of n atoms. If some photons are absorbed by the atoms in
the cloud, the number of transmitted photons (I
t
) will be less than I
0
. Thus
I
t
¼xI
0
where 1 > x > 0. Now suppose that the concentration of atoms in the cloud
is doubled. If the probability of an y particular photon being absorbed is
independent of the number of photons and depends only upon the number
of atoms intercepting the light beam, then I
0
– I
t
photons will still be
absorbed by the first n atoms. For the second n atoms, once again a fraction
x of these xI
0
photons will be absorbed. Thus when n is increased to 2n, I
t
is
given by x
2
I
0
. Similarly, for 3n,4n . . . atoms, I
t

would have the values x
3
I
0
,
x
4
I
0
. . . Thus the relationship between I
t
, I
0
, and the atom concentration, c,
is of the form
I
t
¼ x
c
I
0
log I
t
¼ c log x þlog I
0
log I
t
À log I
0
¼ kc

logðI
t
=I
0
Þ is proportional to c
Since I
t
< I
0
, I
t
/I
0
< 1, and log(I
t
/I
0
) < 0, if we define a parameter A, the
absorbance, as being equal to –log(I
t
/I
0
), then the parameter A will always
2 Cresser
TM
Copyright n 2004 by Marcel Dekker, Inc. All Rights Reserved.
be positive and proportional to concentration. Thus absorbance is the
parameter that should be measured if linear calibration plots are required
in AAS. Note that if 90% of the light is absorbed, I
t

/I
0
is equal to 10/100
or 0.1, and A is equal to 1. Similarly 99% absorption corresponds to an
absorbance of 2, 99.9% to an absorbance of 3, and so on. Precis ely
measuring values of absorbance much greater than 2 will clearly be
technically difficult.
In FES the parameter measured is simply the intensity of the light
emitted by thermally excited atoms. As stated earlier, this will increase
linearly with determinant element concentration. However, at high element
concentrations, some of the emitted light will be reabsorbed. This decreases
the light signal at the detector, so a calibration graph of emission intensity
versus concentration will curve toward the concentration axis.
By the early 1950s, the concept of absorbance and the nature of atomic
absorption spectra had been understood for many decades by spectro-
physicists. However, the concept had not been applied quantitatively at that
time, because of the limitation imposed by the narrowness of atomic
absorption lines. Monochromators then commonly available could provide
a ‘‘window’’ to isolate bands of the UV or visible spectrum about 0.1 nm
wide, but atomic absorption occurred over a much narrower spectral
interval, typically < 0.005 nm. Even if quite strong atomic absorbance
occurred when light passed through a cell containing free atoms of an
element, there would be no change in 95% of the light passing through the
monochromator window. Therefore I
t
would be only very slightly smaller
than I
0
, the ratio I
t

/I
0
would be close to 1, and absorbance would be close
to zero. In other words, sensitivity would always be very poor in AAS
when absorption of light from a continuum source was measured.
Walsh (1955) made a major breakthrough when he realized that
practical AAS instrumentation could be built around light sources that
emitted atomic spectral lines at the same wave lengths as those at which
atomic absorption occurred. By selecting appropriate sources, the emission
line widths could be even narrower than the absorption line widths. Thus
the potential sensitivity problem discussed above was solved at a stroke, and
the concept of the modern atomic absorption spectrometer was born.
The technique of FES predated AAS, because the instrumental
requirements of FES were conceptually simpler. In FES a monochromator
is used to isolate the light emitted by the element of interest from light
emitted by all other elements present in the sample. At the same time the
isolation of a narrow wavelength interval by the monochromator increases
the ratio of the intensity of the light emitted by the element of interest to the
intensity of the background light emitted by the flame. This improves the
detection capability of the technique.
Atomic Absorption and Flame Emission Spectrometry 3
TM
Copyright n 2004 by Marcel Dekker, Inc. All Rights Reserved.
C. Potential Selectivity of AAS
Walsh’s conceptual breakthrough was of enormous significance. Not only
had he suggested a potentially very sensitive analytical method to determine
many elements in the periodic table but he had suggested also a method
that, at least theoretically, should lead to virtually specific analysis. The very
sharpness of the lines in atomic absorption spectra that hitherto had held
back progress suddenly became the method’s most powerful asset. Because

the probability of spectral overlap of the absorption line of one element
with the emission line of another was extremely small, atomic spectral
interferences should be, and indeed are, extremely rare in AAS. In this
respect AAS is vastly superior to FES, where the selectivity depends upon
the complexity of the spectra of all elements present in the samples being
analyzed. Thus FES is much more prone to spectral interferences.
II. INSTRUMENTATION FOR AAS AND FES
We are now in a position to consider the essential components of a typical
atomic absorption spectr ometer, as represented in Fig. 1. In this figure, a
flame is used to convert the determinant species into a cloud of atoms, which
absorb light from a hollow cathode lamp. The most sensitive absorption
wavelength is isolated by a monochromator.
A. Hollow Cathode Lamps
As explained in Sec. I.B, to avoid the need for a very high resolution
monochromator to isolate a very narrow (< ca. 0.005 nm) band of light from
Figure 1 Schematic representation of the main components of a typical atomic
absorption spectrometer.
4 Cresser
TM
Copyright n 2004 by Marcel Dekker, Inc. All Rights Reserved.
a continuum spectrum prior to absorbance measurement, lamps are used
that emit sharp lines at the same wavelengths as those at which absorption
occurs. The vast majority of applications use single-element line sources,
and virtually all of these are hollow cathode lamps.
Figure 2 shows a typical hollow cathode lamp. The hollow cathode is
constructed from the element of interest or one of its alloys. The lamp is
filled with an inert gas, generally neon or argon, at low pressure. A high-
voltage, low-current discharge is struck between the cathode and an anode.
The latter, which commonly is made from tungsten, generally is a small
cylinder or sometimes a small flag-shaped electrode. Sheets of insulator,

often mica, confine the discharge to the central cathode region to obtain
good stability at high intensity. The end window of the lamp is often quartz
or optical silica, to transmit UV light. Ordi nary glasses absorb increasingly
strongly below about 320 nm.
For physical stability, traditionally hollow cathode lamps have an
octal (eight-pin) base that attaches to an eight-hole socket. In simple
instruments only two pins make any electrical connection, however,
although in some more complex instruments additional pins may be
connected to electrical components that serve to provide automated lamp
identification. The lamp base itself invariably has a protruding plastic lip to
make sure that the lamp can be fitted only in the correct position unless
excessive brute force is applied (Fig. 2). The lamps are expensive to replace
and therefore always should be handled gently. Always remember to budget
for a range of lamps when purchasing an instrument for the first time, as
they add significantly to the total package cost.
It is important to align correctly the lamp along the optical axis
through the center of the flame or electrothermal atomizer to the
monochromator entrance slit. Most AAS instruments can accommodate
lamps from diverse manufacturers (check before changing manufacturers!),
and such lamps often differ in size. The lamps fit inside some sort of
Figure 2 The essential components of a typical hollow cathode lamp.
Atomic Absorption and Flame Emission Spectrometry 5
TM
Copyright n 2004 by Marcel Dekker, Inc. All Rights Reserved.
supporting cradle, so that their position is fully adjustable both horizontally
and vertically. Optimal alignment generally is best found by maximizing
the signal from the detector while the atomizer is off. It is tempting to use
the image of the glowing cathode (usually red from the neon filler gas) at the
entrance slit to align the lamp, but this is not advisable for final fine
adjustment. This is because the focal point for UV light may be displaced by

a few mm from that for red light if the instrument optics use lenses.
Most determinations by AAS are completed using single-element
hollow cathode lamps, in spite of the high cost of having an additional lamp
for each element to be determined. Single element lamps often give superior
signal-to-noise ratios to those with multielement lamps and thus result in
slightly better detection limits and improved precision. An exception in the
author’s experience is the calcium/magnesium dual-element lamp, which
usually provides directly comparable performance to the corresponding
single-element lamps. Multielement lamps containing up to six or more
elements are commercially available but are not to be recommended
generally if optimal performance is required.
B. Alternative Line Sources
Some very volatile elements such as arsenic and selenium have their main
AAS wavelengths below 200 nm, at wavelengths where absorption by air
becomes significant. The hollow cathode lamps for these elements invariably
exhibit low intensity and poor stability. The search for more intense sources
for such elements resulted in the development of microwave-powered
electrodeless discharge lamps (EDLs) as line sources at the end of the 1960s.
For volatile elements, these lamps were generally much more intense than
the corresponding hollow cathode lamps, sometimes by two to three orders
of magnitude. They were sometimes notoriously unstable, however,
requiring too much operator skill to find favor for routine use in AAS.
Subsequently, however, radio-frequency-powered (r.f.) electrodeless
discharge lamps became available commercially for many elements, includ-
ing As, Bi, Cd, Cs, Ge, Hg, K, P, Pb, Rb, Sb, Se, Te, Tl, Sn, Ti, and Zn.
Although dimmer than the earlier microwave EDLs, r.f. EDLs exhibit far
superior stability and are still generally substantially brighter than the
corresponding hollow c athode lamps. In the author’s experience, they are
well worth considering if arsenic or selenium is to be determined routinely,
although considerable expense is incurred initially because the lamps require

a separate r.f. power supply.
In flame atomic absorption spectrometry, the device used to detect the
light signal will ‘‘see’’ the light emitted by the hollow cathode lamp but also
the light emitted by the flame at the wavelength being used. Thus the
6 Cresser
TM
Copyright n 2004 by Marcel Dekker, Inc. All Rights Reserved.
detector would ‘‘see’’ too much light, and absorbance would not be correctly
measured. However, if the lamp power supply is modulated so that the lamp
effectively flashes on and off at a frequency of, say, 180 Hz, and the
instrument is designe d so that it responds only to modulated light at this
frequency, it is possible to discriminate between the modulated light emitted
by the hollow cathode lamp and the unmodulated light emitted by the flame.
Thus the true absorbance can be measured, even if the atomizer is emitting
quite intensely. Therefore the power supplies to hollow cathode lamps in
atomic absorption spectrometers invariably are modulated. The required
lamp signal is isolated by synchronous demodulation, which further
improves the signal-to-noise ratio.
C. The Flame as an Atomizer
In AAS and FES, the determinant species in a solid or solution sample
must be converted into free atoms, because the techniques respectively
involve the absorption of light by, or emission of light from, free atoms. This
is most commonly achieved by dissolving the sample and spraying the
resulting solution into a flame hot enough to convert the determinant to free
atoms.
The function of the flame is threefold, in practice. It must dry,
vaporize, and then atomize the sample in a reproducible manner with
respect to both space and time. Unlike gravimetric or titrimetric analysis,
AAS and FES are secondary methods of analysis. Concentrations of
determinants are found by comparing the absorbance or emission values

obtained for samples with those obtained for standards of known
concentrations of the elements of interest. In both techniques it is therefore
vital that samples and standards are always atomized with the same
efficiency to produce an atomic vapor cloud with highly reproducible
geometry. In FES, the extent of thermal excitation must also be similar for
samples and standar ds. If samples and standards behave differently, errors
must inevitably result.
1. The Air–Acetylene Flame
Air–propane and air–butane flames were used to atomize samples in the
earliest days of flame AAS, as they had a reputation for being simple and
safe enough for routine use. Unfortunately it was found soon that such
flames were unsatisfactory for atomizing many thermally stable chemical
compounds. Sometimes, therefore, samples and standards were not
atomized to the same extent, so erroneous results were obtaine d. The
most commonly used flame at the present time is the air–acetylene flame.
Atomic Absorption and Flame Emission Spectrometry 7
TM
Copyright n 2004 by Marcel Dekker, Inc. All Rights Reserved.
This is still safe, relatively inexpensive, and hot enough at ca. 2200

Cto
atomize molecules of most, though not all, common elements. It is not hot
enough to break the element–oxygen bonds of some elements such as
aluminum and silicon, the so-called refractory oxide-forming elements. Such
determinants require a hotter flame. Another limitation of the air–acetylene
flame is that atomization efficiency of some elements may be influenced by
matrix elements and ions. For example, phosphate or aluminum suppress
the atomic absorption signals of calcium in this flame.
Producing a stable flame on a burner head requires a gas mixture for
which the upward flow velocity just exceeds the downward burning velocity.

If the burning velocity becomes greater, there is a danger that the flame will
burn back through the burner slot, resulting in a potenti ally dangerous
explosion. This process is known as a flashback. Pre-mixed oxygen–
acetylene flames, although substantially hotter than air–acetylene flames,
are never used routinely because the burning velocity is too great, and the
risk of flashback is too high.
If, on the other hand, the flow velocity exceeds the burning velocity by
too much, the flame ‘‘lifts off ’’ from the burner head. Chemistry students
may have experienced this phenomenon when trying to ignite the flame of a
Bunsen burner with the air hole fully open. The flame takes the form of an
unstable fire ball a cm or two above the burner port for a few seconds, and
then often is extinguished. In the case of the burner heads used in AAS, at a
given total flow of fuel plus oxidant, the flow velocity is regulated by the
dimensions of the burner slot; the narrower and/or shorter the slot, the
faster the gas flow velocity.
2. The Nitrous Oxide–Acetylene Flame
For years it was thought that no flame appreciably hotter than the air–
acetylene flame and also safe for routine use would be found, until John Willis
(1965) investigated the use of the premixed nitrous oxide–acetylene flame
(sometimes also known nowadays as the dinitrogen oxide–acetylene flame).
This flame had a temperature of around 3000

C, and it provided a fuel-rich
environment that was chemically very reducing. Thus it proved to be suitable
for breaking refractory metal–oxi de bonds. Its burning velocity exceeded that
of the air–acetylene flame, so that a smaller burner slot was necessary but
proved to be much less than that of oxygen–acetylene flames. Even so, the
early days of its use were marred occasionally by very noisy flashbacks.
3. Safe Use of Acetylene Flames
The spray chamber containing the fuel–oxidant gas mixture in modern AAS

instruments is always now fitted with a blowout membrane or safety bung. If
8 Cresser
TM
Copyright n 2004 by Marcel Dekker, Inc. All Rights Reserved.
the explosive fuel–oxidant mixture inside the spray chamber is accidentally
ignited via a flashback, the very rapid pressure buildup blows out the bung or
ruptures the membrane, immediately releasing the pressure and minimizing
the risk of damage to the mixing chamber, other instrumental components,
and most importantly the operator. The drain that takes away surplus
solution from the sample (see Fig. 1) also functions in this way to some
extent, by emptying. It is imperative after a flashback to replace the bung or
membrane, and to refill the drain, before attempting to relight the flame.
In the majority of modern instruments, especially the more expensive
ones, a flashback automatically trips off the fuel supply to minimize the risk
of fire. The trip switch must be reset before the flame can be relit. More
sophisticated instruments incorporate a range of additional sensors to
improve safety further. For example, instruments may be equipped with
devices to detect the presence of the correct burner head prior to allowing
the ope rator to light nitrous oxide–supported flames; gas pressure sensors
may be fitted in the fuel and oxidant lines to ensure adequate operating
pressures; flame detectors may be fitted that shut off the fuel automatically if
the flame does not appear to be alight. It is useful to be aware of these
devices, as they may cause delays in first lighting flames after gas cylinders
have been replaced, or if, for example, air has been flushed inadvertently
through the acetylene line.
It is impor tant to check at intervals that all fuel and oxidant lines and
their associated connectors within the instrument are in good condition,
because as far as the author is aware, virtually no instrument automatically
detects slow fuel leaks which could cause a buildup of an explosive gas
mixture within an instrument casing. Obviously piping and connectors

external to the instrument must also be checked to be in sound condition,
but hidden tubing is more likely to be overlooked.
Acetylene should not be used routinely at pressures above 10 p.s.i.
(70 kPa), because detonation becomes possible. Nor must it ever be allowed
to come into contact with copper piping or fittings, because of the risk of
formation of explosive copper acetylide. Acetylene cylinders contain the gas
in solution in acetone on a porous ceramic support, so the cylinders always
should be stored and handled in an upright position to avoid the risk of
liquid acetone entering fuel lines. If, however, this does happen, it is best
to get advice from the manufacturer of the instrument on the procedure
to follow. Once the operating pressure of the cylinder falls to about 80 p.s.i.
(0.6 MPa), it should be replaced, to prevent the passage of excessive acetone
vapor into the flame.
The fumes from acetylene flames may be toxic, so an efficient
extraction hood is required over the flame. The nitrous oxide–acetylene
flame is hotter and taller than the air–acetylene flame, and manufacturers’
Atomic Absorption and Flame Emission Spectrometry 9
TM
Copyright n 2004 by Marcel Dekker, Inc. All Rights Reserved.
advice should be sought on suitable extraction systems for the exhaust
gases. It is important to use an appropriate rate of extraction and hood
geometry.
4. Observations on Burner Heads
When flame AAS was first introduced, manufacturers opted for long-path
flames (100–120 mm), because they wanted the technique to be sensitive and
recognized that longer cells gave bigger signals in solution spectro-
photometry. In flame AAS, the situation is rather different. Using a
longer flame does not put more sample into the optical path. However,
increasing the flame cross-sectional area increases the residence time of
atoms in the hollow cathode lamp beam. The burner heads are designed to

be mechanically robust and safe. However, flat-topped burner heads initially
used rapidly got too hot to touch and were prone to clogging. Even ten
minutes after the flame has been extinguished after a period of extended use,
it is still possible to get a painful burn by touching the head of such a burner.
Boling (1966) described a novel burner head with three parallel slots
rather than the normal single slot (Fig. 3A). His idea was to produce a
flame-shielded flame that might have a higher central flame temperature by
Figure 3 Examples of burner head designs. A, the Boling Triple Slot burner; B, a
water-cooled burner head with a triangular cross section; C, a flat-topped burner
head; D, a typical modern burner head design. In each case the arrows indicate the
pattern of air entrainment.
10 Cresser
TM
Copyright n 2004 by Marcel Dekker, Inc. All Rights Reserved.
minimization of cooling with entrained air. The design was operationally
successful because sensitivities for elements such as chromium, which are
not readily atomized, were improved. Almost 20 years later, Cresser (1993)
was experimenting with water-cooling of burner heads to reduce clogging
problems for solutions with high concentrations of dissolved solids
frequently encountered in environmental analysis. With cooling water
along either side of the slot in the head shown in Fig. 3B, the burner head
became so cold that condensation in the slot rapidly extinguished the flame.
Even with the cooling water disconnected, the head remained so cool that it
was possible to touch the side of the head while the flame was alight without
any risk of a burn. The explanation appears to be much smoother air
entrainment when the burner head has a triangular cross section (Fig. 3B)
compared to the very turbulent entrainment and associated heating effect
with a flat-topped head (Fig. 3C). The head shown in Fig. 3B gave similar
sensitivity enhancements and reduced chemical interferences to the Boling
burner, suggesting that the benefits of the latter were possibly attributable to

its cross section rather than the triple slot per se. Nowadays the majority
of AAS instrument manufacturers use a head somewhere between a
flat-topped head and a full triangular cross section, as shown in Fig. 3D.
5. The Flame as an Emission Source
It is to be expected that the hotter nitrous oxide–acetylene flame would
excite more atoms and therefore provide greater emission intensity and
better sensitivity than the cooler air–acetylene flame, and this is indeed the
case. It is also capable of exciting refractory oxide-forming elements such as
aluminum. However, elements that are not refractory, and that emit at
wavelengths above about 350 nm, are excited to an appreciable extent in
air–acetylene and can be measured by FES. Longer emission wavelengths
are associated with lower excitation potentials (so atoms are more readily
excited at a given flame temperature), and sensitivity at long wavelengths is
therefore excellent. Thus elements such as sodium and potassium, which
emit orange and red light respectively, can be determined with much better
sensitivity by FES than by AAS, even in an air–acetylene flame.
Traditionally in FES, small, circular burner heads were the norm
(Cresser, 1994). With the development of AAS, however, it was soon
realized that excellent detection limits by FES could be obtained using a
long-path burner head in an AA spectrometer. The only modification
required was to make the de tector respond to a signal that, unlike the
hollow cathode lamp signal, was not modulated. This was often achieved by
modulating the light signal emitted from the flame with a mechanical
chopper. Now most AA spectrometers can be used in the emission mode.
Atomic Absorption and Flame Emission Spectrometry 11
TM
Copyright n 2004 by Marcel Dekker, Inc. All Rights Reserved.
The nitrous oxide–acetylene flame can give excellent detection limits
by FES, but it is not that widely used in practice for emission measurements.
Many analysts prefer AAS because of the greater freedom from risk of

spectral interference with the absorption technique. It is useful, however, to
be aware of the potential of FES if, for example, a hollow cathode lamp is
unavailable for a particular determination that is required in a hurry.
D. Electrothermal Atomization
It will be seen in Sec. II.E that a major constraint in flame AAS is the limited
efficiency of transport of sample from a beaker of solution to the flame. In
the late 1960s, this prompted a num ber of distinguished researchers such as
L’vov, West, Massmann, and others to look for alternative atomization
systems that did not depend upon generation and transport of aerosol to
flames (Potts, 1987; Slavin, 1991; Lajunen, 1992). Early systems used small
discrete portions of sample solution (ca. 50 mL) injected by hand onto the
center of a resistively heated graphite rod or into a graphite tube heated
using an arc system. Graphite was chemically inert (provided it was
sheathed in nitrogen or argon while being heated) and capable of
withstanding high temperatures. Moreover the graphite surface provided
reducing conditions well suited to the reduction of metal oxides to the free
elements. What ultimately evolved commercially was the graphite furnace
electrothermal atomization (ETA) system that is still widely in use today,
based upon a resistively heated graphite tube furnace (illustrated
schematically in Fig. 4). In the early days, samples were injected by hand,
a process requiring great manual dexterity. Fortunately, nowadays, sample
injection is invariably performed under computer control using a robotic
arm/autosampler system, and the precision attainable is excellent provided
the system has been properly optimized.
From the early days, ETA-AAS was plagued by interferences. This
was because several processes had to occur sequentially to achieve
atomization. The atomizer temperature was raised in stages, so that the
Figure 4 Exploded view of the main parts of a typical graphite furnace atomizer.
When assembled, the sample is injected via the two holes shown. For simplicity, the
water-cooled electrical connectors, lenses, and gas sheathing systems are not shown.

12 Cresser
TM
Copyright n 2004 by Marcel Dekker, Inc. All Rights Reserved.
sample could first be dried by evaporating the solvent and then ashed at a
few hundred degrees Celsius to remove any organic matrix, and finally
atomized at a high temperature. A fourth and even hotter stage was then
usually used to make sure every trace of determinant was removed from the
atomizer. The problem was that the matrix had to be eliminated without
any premature loss of determinant, and any chemical reactions between
determinant and matrix components during heating could not be allowed
to modify atomization efficiency. To make matters worse, atomic
recombinations had to be prevented if they made samples and standards
behave differently. As a consequence, although precision improved with
automation, and some impressively low detection limits were eventually
achieved (see, e.g., Slavin, 1991), attaining the necessary accuracy proved to
be a real challenge.
From the outset, the light from the hollow cathode lamp was focused
to a narrow beam passing along the furnace tube, parallel to the wall and
passing just above the sample. The problem with this configuration was that
the ends of the tube were, of necessity, much cooler than the tube center.
Perkin-Elmer circumvented this problem by designing a transversely
mounted tube, which was heated from the sides in such a way that the
whole of the furnace containing the sample was at a more or less uniform
temperature, as discussed by Lajunen (1992). This design minimized the
condensation of determinant and matrix components at the furnace ends,
but the tubes became rather expensive consumables because of the greater
design complexity. This has led to some interesting studies of anything that
might adversely affect the useful life of these tubes, such as corrosive acids
in sample matrices (Rohr et al., 1999).
Three major developments were needed for ETA to become a reliable

routine tool once automated sample introduction methodology had been
developed. The first was the development of pyrolyic coatings on the inner
graphite surface, which minimized penetration of sample solution into the
porous graphite. Second came the development of matrix modifiers, which
were chosen to delay atomization of determinant elements until high furnace
temperatures had been attained, making atomic recombinations less likely
to be a problem. Third came the ingenious idea from Boris L’vov of placing
the sample not directly on the furnace tube wall but on a small boat or
platform placed inside the tube (Fig. 5), and heating it primarily by
conduction through the edges of the platform resting on the tube inner
walls. This meant that by the time the sample was eventually atomized, the
tube walls and the gas inside the tube were already at a much higher
temperature than the volatilizing and atomizing determinant element. These
developments dramatically reduced interference effects (Lajunen, 1992).
They are discussed further in Sec. III.E.
Atomic Absorption and Flame Emission Spectrometry 13
TM
Copyright n 2004 by Marcel Dekker, Inc. All Rights Reserved.