100% found this document useful (1 vote)

48 views150 pages

Instant Notes in Analytical Chemistry(sec 1)

analytical notes

Uploaded by

Usman Saeed

We take content rights seriously. If you suspect this is your content, claim it here.

Available Formats

Download as PDF, TXT or read online on Scribd

100% found this document useful (1 vote)

48 views150 pages

Instant Notes in Analytical Chemistry(sec 1)

analytical notes

Uploaded by

Usman Saeed

We take content rights seriously. If you suspect this is your content, claim it here.

Available Formats

Download as PDF, TXT or read online on Scribd

You are on page 1/ 150

11

0111
Analytical Chemistry

0111

0
11
ii Section K – Lipid metabolism

The INSTANT NOTES series

Series editor
B.D. Hames
School of Biochemistry and Molecular Biology, University of Leeds, Leeds, UK

Animal Biology
Biochemistry 2nd edition
Chemistry for Biologists
Developmental Biology
Ecology 2nd edition
Genetics
Immunology
Microbiology
Molecular Biology 2nd edition
Neuroscience
Plant Biology
Psychology

Forthcoming titles
Bioinformatics

The INSTANT NOTES Chemistry series

Consulting editor: Howard Stanbury

Analytical Chemistry
Inorganic Chemistry
Medicinal Chemistry
Organic Chemistry
Physical Chemistry
11

Analytical Chemistry
0111

D. Kealey
School of Biological and Chemical Sciences
Birkbeck College, University of London, UK
0111
and
Department of Chemistry
University of Surrey, Guildford, UK

and

0111
P. J. Haines
Oakland Analytical Services,
Farnham, UK

0111

0
11
© BIOS Scientiﬁc Publishers Limited, 2002

First published 2002 (ISBN 1 85996 189 4)

This edition published in the Taylor & Francis e-Library, 2005.

“To purchase your own copy of this or any of Taylor & Francis or Routledge’s
collection of thousands of eBooks please go to www.eBookstore.tandf.co.uk.”

All rights reserved. No part of this book may be reproduced or transmitted, in any form or
by any means, without permission.

A CIP catalogue record for this book is available from the British Library.

ISBN 0-203-64544-8 Master e-book ISBN

ISBN 0-203-68109-6 (Adobe eReader Format)

ISBN 1 85996 189 4 (Print Edition)

BIOS Scientiﬁc Publishers Ltd

9 Newtec Place, Magdalen Road, Oxford OX4 1RE, UK
Tel. +44 (0)1865 726286. Fax +44 (0)1865 246823
World Wide Web home page: http://www.bios.co.uk/

Distributed exclusively in the United States, its dependent territories, Canada, Mexico, Central
and South America, and the Caribbean by Springer-Verlag New York Inc, 175 Fifth Avenue,
New York, USA, by arrangement with BIOS Scientiﬁc Publishers, Ltd, 9 Newtec Place,
Magdalen Road, Oxford OX4 1RE, UK
C ONTENTS
11

Abbreviations vii
Preface ix

Section A – The nature and scope of analytical chemistry 1

0111 A1 Analytical chemistry, its functions and applications 1
A2 Analytical problems and procedures 3
A3 Analytical techniques and methods 5
A4 Sampling and sample handling 10
A5 Calibration and standards 15
A6 Quality in analytical laboratories 18

Section B − Assessment of data 21

B1 Errors in analytical measurements 21
B2 Assessment of accuracy and precision 26
0111
B3 Signiﬁcance testing 34
B4 Calibration and linear regression 41
B5 Quality control and chemometrics 49

Section C − Analytical reactions in solution 55

C1 Solution equilibria 55
C2 Electrochemical reactions 61
C3 Potentiometry 66
C4 pH and its control 74
0111 C5 Titrimetry I: acid–base titrations 80
C6 Complexation, solubility and redox equilibria 85
C7 Titrimetry II: complexation, precipitation and redox
titrations 90
C8 Gravimetry 95
C9 Voltammetry and amperometry 98
C10 Conductimetry 104

Section D − Separation techniques 109

D1 Solvent and solid-phase extraction 109
0111 D2 Principles of chromatography 119
D3 Thin-layer chromatography 131
D4 Gas chromatography: principles and instrumentation 137
D5 Gas chromatography: procedures and applications 149
D6 High-performance liquid chromatography: principles
and instrumentation 155
D7 High-performance liquid chromatography: modes,
procedures and applications 166
D8 Electrophoresis and electrochromatography: principles
and instrumentation 174
0 D9 Electrophoresis and electrochromatography: modes,
11 procedures and applications 182
vi Contents

Section E − Spectrometric techniques 189

E1 Electromagnetic radiation and energy levels 189
E2 Atomic and molecular spectrometry 195
E3 Spectrometric instrumentation 201
E4 Flame atomic emission spectrometry 206
E5 Inductively coupled plasma spectrometry 209
E6 X-ray emission spectrometry 214
E7 Atomic absorption and atomic ﬂuorescence spectrometry 218
E8 Ultraviolet and visible molecular spectrometry:
principles and instrumentation 223
E9 Ultraviolet and visible molecular spectrometry:
applications 228
E10 Infrared and Raman spectrometry: principles and
instrumentation 233
E11 Infrared and Raman spectrometry: applications 242
E12 Nuclear magnetic resonance spectrometry: principles
and instrumentation 248
E13 Nuclear magnetic resonance spectrometry: interpretation
of proton and carbon-13 spectra 261
E14 Mass spectrometry 270

Section F − Combined techniques 283

F1 Advantages of combined techniques 283
F2 Sample identiﬁcation using multiple spectrometric
techniques data 285
F3 Gas chromatography–mass spectrometry 294
F4 Gas chromatography–infrared spectrometry 298
F5 Liquid chromatography–mass spectrometry 302

Section G − Thermal methods 305

G1 Thermogravimetry 305
G2 Differential thermal analysis and differential scanning
calorimetry 311
G3 Thermomechanical analysis 316
G4 Evolved gas analysis 320

Section H – Sensors, automation and computing 323

H1 Chemical sensors and biosensors 323
H2 Automated procedures 328
H3 Computer control and data collection 331
H4 Data enhancement and databases 333

AAS atomic absorption spectrometry ICP-MS ICP-mass spectrometry

ADC analog-to-digital converter IEC ion-exchange chromatography
AFS atomic fluorescence spectrometry ISE ion-selective electrode
ANOVA analysis of variance LVDT linear variable differential
ATR attenuated total reflectance transformer
0111 BPC bonded-phase chromatography MEKC micellar electrokinetic
CC chiral chromatography chromatography
CGE capillary gel electrophoresis MIR multiple internal reflectance
CI confidence interval MS mass spectrometry
CIEF capillary isoelectric focusing NIR near infrared
CL confidence limits NMR nuclear-magnetic resonance
CPU central processing unit NPD nitrogen-phosphorus detector
CRM certified reference material PAH polycyclic aromatic hydrocarbons
CZE capillary zone electrophoresis PC paper chromatography
DAC digital-to-analog converter PCA principal component analysis
0111 DAD diode array detector PCR principal component regression
DMA dynamic mechanical analysis PDMS polydimethylsiloxane
DME dropping mercury electrode PLS partial least squares
DSC differential scanning calorimetry QA quality assurance
DTA differential thermal analysis QC quality control
DTG derivative thermogravimetry RAM random access memory
DVM digital voltmeter RF radiofrequency
ECD electron-capture detector RI refractive index
EDAX energy dispersive analysis ROM read only memory
of X-rays RMM relative molecular mass
0111 EDTA ethylenediaminetetraacetic acid SCE saturated calomel electrode
EGA evolved gas analysis SDS sodium dodecyl sulfate
FA factor analysis SDS-PAGE SDS-polyacrylamide gel
FAES flame atomic emission electrophoresis
spectometry SE solvent extraction
FFT fast Fourier transform SEC size-exclusion chromatography
FID flame ionization detector SHE standard hydrogen electrode
or free induction decay SIM selected ion monitoring
GC gas chromatography SPE solid phase extraction
GLC gas liquid chromatography SPME solid phase microextraction
0111 GSC gas solid chromatography SRM standard reference material
HATR horizontal attenuated total TCD thermal conductivity detector
reflectance TG thermogravimetry
HPLC high-performance liquid TIC total ion current
chromatography TISAB total ionic strength adjustment
IC ion chromatography buffer
ICP inductively coupled plasma TLC thin-layer chromatography
ICP-AES ICP-atomic emission spectrometry TMA thermomechanical analysis
ICP-OES ICP-optical emission spectrometry

0
11
P REFACE

Analytical chemists and others in many disciplines frequently ask questions such as: What is this
substance?; How concentrated is this solution?; What is the structure of this molecule? The answers to
these and many other similar questions are provided by the techniques and methods of analytical
chemistry. They are common to a wide range of activities, and the demand for analytical data of a
chemical nature is steadily growing. Geologists, biologists, environmental and materials scientists,
physicists, pharmacists, clinicians and engineers may all find it necessary to use or rely on some of the
techniques of analysis described in this book.
If we look back some forty or fifty years, chemical analysis concentrated on perhaps three main areas:
qualitative testing, quantitative determinations, particularly by ‘classical’ techniques such as titrimetry
and gravimetry, and structural analysis by procedures requiring laborious and time-consuming calcu-
lations. The analytical chemist of today has an armoury of instrumental techniques, automated systems
and computers which enable analytical measurements to be made more easily, more quickly and more
accurately.
However, pitfalls still exist for the unwary! Unless the analytical chemist has a thorough understand-
ing of the principles, practice and limitations of each technique he/she employs, results may be inaccu-
rate, ambiguous, misleading or invalid. From many years of stressing the importance of following
appropriate analytical procedures to a large number of students of widely differing abilities, backgrounds
and degrees of enthusiasm, the authors have compiled an up-to-date, unified approach to the study of
analytical chemistry and its applications. Surveys of the day-to-day operations of many industrial and
other analytical laboratories in the UK, Europe and the USA have shown which techniques are the most
widely used, and which are of such limited application that extensive coverage at this level would be
inappropriate. The text therefore includes analytical techniques commonly used by most analytical
laboratories at this time. It is intended both to complement those on inorganic, organic and physical
chemistry in the Instant Notes series, and to offer to students in chemistry and other disciplines some guid-
ance on the use of analytical techniques where they are relevant to their work. We have not given extended
accounts of complex or more specialized analytical techniques, which might be studied beyond first- and
second-year courses. Nevertheless, the material should be useful as an overview of the subject for those
studying at a more advanced level or working in analytical laboratories, and for revision purposes.
The layout of the book has been determined by the series format and by the requirements of the
overall analytical process. Regardless of the discipline from which the need for chemical analysis arises,
common questions must be asked:
● How should a representative sample be obtained?
● What is to be determined and with what quantitative precision?
● What other components are present and will they interfere with the analytical measurements?
● How much material is available for analysis, and how many samples are to be analyzed?
● What instrumentation is to be used?
● How reliable is the data generated?
These and related questions are considered in Sections A and B.
Most of the subsequent sections provide notes on the principles, instrumentation and applications of
both individual and groups of techniques. Where suitable supplementary texts exist, reference is made
to them, and some suggestions on consulting the primary literature are made.
We have assumed a background roughly equivalent to UK A-level chemistry or a US general
chemistry course. Some simplification of mathematical treatments has been made; for example, in the
sections on statistics, and on the theoretical basis of the various techniques. However, the texts listed
under Further Reading give more comprehensive accounts and further examples of applications.
x Preface

We should like to thank all who have contributed to the development of this text, especially the many
instrument manufacturers who generously provided examples and illustrations, and in particular Perkin
Elmer Ltd. (UK) and Sherwood Scientiﬁc Ltd. (UK). We would like also to thank our colleagues who
allowed us to consult them freely and, not least, the many generations of our students who found
questions and problems where we had thought there were none!
DK
PJH
Section A – The nature and scope of analytical chemistry

A1 A NALYTICAL CHEMISTRY, ITS

FUNCTIONS AND
APPLICATIONS

Key Notes
Deﬁnition Analytical chemistry is a scientiﬁc discipline used to study the chemical
composition, structure and behavior of matter.

Purpose The purpose of chemical analysis is to gather and interpret chemical

information that will be of value to society in a wide range of contexts.

Scope and Quality control in manufacturing industries, the monitoring of clinical

applications and environmental samples, the assaying of geological specimens, and
the support of fundamental and applied research are the principal
applications.

Deﬁnition Analytical chemistry involves the application of a range of techniques and

methodologies to obtain and assess qualitative, quantitative and structural
information on the nature of matter.
● Qualitative analysis is the identiﬁcation of elements, species and/or
compounds present in a sample.
● Quantitative analysis is the determination of the absolute or relative amounts
of elements, species or compounds present in a sample.
● Structural analysis is the determination of the spatial arrangement of atoms in
an element or molecule or the identiﬁcation of characteristic groups of atoms
(functional groups).
● An element, species or compound that is the subject of analysis is known as an
analyte.
● The remainder of the material or sample of which the analyte(s) form(s) a part
is known as the matrix.

Purpose The gathering and interpretation of qualitative, quantitative and structural infor-
mation is essential to many aspects of human endeavor, both terrestrial and
extra-terrestrial. The maintenance of, and improvement in, the quality of life
throughout the world, and the management of resources rely heavily on
the information provided by chemical analysis. Manufacturing industries use
analytical data to monitor the quality of raw materials, intermediates and
2 Section A – The nature and scope of analytical chemistry

ﬁnished products. Progress and research in many areas is dependent on estab-

lishing the chemical composition of man-made or natural materials, and the
monitoring of toxic substances in the environment is of ever increasing impor-
tance. Studies of biological and other complex systems are supported by the
collection of large amounts of analytical data.

Scope and Analytical data are required in a wide range of disciplines and situations that
applications include not just chemistry and most other sciences, from biology to zoology, but
the arts, such as painting and sculpture, and archaeology. Space exploration and
clinical diagnosis are two quite disparate areas in which analytical data is vital.
Important areas of application include the following.
● Quality control (QC). In many manufacturing industries, the chemical
composition of raw materials, intermediates and finished products needs to
be monitored to ensure satisfactory quality and consistency. Virtually all
consumer products from automobiles to clothing, pharmaceuticals and food-
stuffs, electrical goods, sports equipment and horticultural products rely, in
part, on chemical analysis. The food, pharmaceutical and water industries in
particular have stringent requirements backed by legislation for major compo-
nents and permitted levels of impurities or contaminants. The electronics
industry needs analyses at ultra-trace levels (parts per billion) in relation to the
manufacture of semi-conductor materials. Automated, computer-controlled
procedures for process-stream analysis are employed in some industries.
● Monitoring and control of pollutants. The presence of toxic heavy metals
(e.g., lead, cadmium and mercury), organic chemicals (e.g., polychlorinated
biphenyls and detergents) and vehicle exhaust gases (oxides of carbon,
nitrogen and sulfur, and hydrocarbons) in the environment are health hazards
that need to be monitored by sensitive and accurate methods of analysis, and
remedial action taken. Major sources of pollution are gaseous, solid and liquid
wastes that are discharged or dumped from industrial sites, and vehicle
exhaust gases.
● Clinical and biological studies. The levels of important nutrients, including
trace metals (e.g., sodium, potassium, calcium and zinc), naturally produced
chemicals, such as cholesterol, sugars and urea, and administered drugs in the
body fluids of patients undergoing hospital treatment require monitoring.
Speed of analysis is often a crucial factor and automated procedures have been
designed for such analyses.
● Geological assays. The commercial value of ores and minerals is determined
by the levels of particular metals, which must be accurately established.
Highly accurate and reliable analytical procedures must be used for this
purpose, and referee laboratories are sometimes employed where disputes
arise.
● Fundamental and applied research. The chemical composition and structure
of materials used in or developed during research programs in numerous
disciplines can be of significance. Where new drugs or materials with potential
commercial value are synthesized, a complete chemical characterization may
be required involving considerable analytical work. Combinatorial chemistry
is an approach used in pharmaceutical research that generates very large
numbers of new compounds requiring confirmation of identity and structure.
Section A – The nature and scope of analytical chemistry

A2 A NALYTICAL PROBLEMS
AND PROCEDURES

Key Notes

Analytical problems Selecting or developing and validating appropriate methods of analysis

to provide reliable data in a variety of contexts are the principal problems
faced by analytical chemists.

Analytical Any chemical analysis can be broken down into a number of stages that
procedures include a consideration of the purpose of the analysis, the quality of the
results required and the individual steps in the overall analytical
procedure.

Analytical The stages or steps in an overall analytical procedure can be summarized as

procedures follows.
● Deﬁnition of the problem. Analytical information and level of accuracy
required. Costs, timing, availability of laboratory instruments and facilities.
● Choice of technique and method. Selection of the best technique for the
required analysis, such as chromatography, infrared spectrometry, titrimetry,
thermogravimetry. Selection of the method (i.e. the detailed stepwise instruc-
tions using the selected technique).
● Sampling. Selection of a small sample of the material to be analyzed. Where
this is heterogeneous, special procedures need to be used to ensure that a
genuinely representative sample is obtained (Topic A4).
4 Section A – The nature and scope of analytical chemistry

● Sample pre-treatment or conditioning. Conversion of the sample into a form

suitable for detecting or measuring the level of the analyte(s) by the selected
technique and method. This may involve dissolving it, converting the
analyte(s) into a specific chemical form or separating the analyte(s) from other
components of the sample (the sample matrix) that could interfere with detec-
tion or quantitative measurements.
● Qualitative analysis. Tests on the sample under specified and controlled
conditions. Tests on reference materials for comparison. Interpretation of the
tests.
● Quantitative analysis. Preparation of standards containing known amounts
of the analyte(s) or of pure reagents to be reacted with the analyte(s).
Calibration of instruments to determine the responses to the standards under
controlled conditions. Measurement of the instrumental response for each
sample under the same conditions as for the standards. All measurements
may be replicated to improve the reliability of the data, but this has cost and
time implications. Calculation of results and statistical evaluation.
● Preparation of report or certificate of analysis. This should include a
summary of the analytical procedure, the results and their statistical assess-
ment, and details of any problems encountered at any stage during the
analysis.
● Review of the original problem. The results need to be discussed with regard
to their significance and their relevance in solving the original problem.
Sometimes repeat analyses or new analyses may be undertaken.
Section A – The nature and scope of analytical chemistry

A3 A NALYTICAL TECHNIQUES
AND METHODS

Key Notes

Analytical Chemical or physico-chemical processes that provide the basis for

techniques analytical measurements are described as techniques.

Analytical methods A method is a detailed set of instructions for a particular analysis using a
speciﬁed technique.

Method validation A process whereby an analytical method is checked for reliability in

terms of accuracy, reproducibility and robustness in relation to its
intended applications.

Analytical An analytical method consists of a detailed, stepwise list of instructions to be

methods followed in the qualitative, quantitative or structural analysis of a sample for one
or more analytes and using a speciﬁed technique. It will include a summary and
6 Section A – The nature and scope of analytical chemistry

Table 1. Analytical techniques and principal applications

Technique Property measured Principal areas of application
Gravimetry Weight of pure analyte or compound Quantitative for major or minor
of known stoichiometry components
Titrimetry Volume of standard reagent solution Quantitative for major or minor
reacting with the analyte components
Atomic and molecular Wavelength and intensity of Qualitative, quantitative or structural
spectrometry electromagnetic radiation emitted or for major down to trace level
absorbed by the analyte components
Mass spectrometry Mass of analyte or fragments of it Qualitative or structural for major
down to trace level components
isotope ratios
Chromatography and Various physico-chemical properties Qualitative and quantitative
electrophoresis of separated analytes separations of mixtures at major to
trace levels
Thermal analysis Chemical/physical changes in the Characterization of single or mixed
analyte when heated or cooled major/minor components
Electrochemical analysis Electrical properties of the analyte Qualitative and quantitative for major
in solution to trace level components
Radiochemical analysis Characteristic ionizing nuclear Qualitative and quantitative at major
radiation emitted by the analyte to trace levels

Table 2. Spectrometric techniques and principal applications

Technique Basis Principal applications
Plasma emission spectrometry Atomic emission after excitation in high Determination of metals and some
temperature gas plasma non-metals mainly at trace levels
Flame emission spectrometry Atomic emission after flame excitation Determination of alkali and alkaline
earth metals
Atomic absorption spectrometry Atomic absorption after atomization Determination of trace metals and
by flame or electrothermal means some non-metals
Atomic fluorescence Atomic fluorescence emission after Determination of mercury and
spectrometry flame excitation hydrides of non-metals at trace
levels
X-ray emission spectrometry Atomic or atomic fluorescence Determination of major and minor
emission after excitation by electrons elemental components of
or radiation metallurgical and geological samples
γ-spectrometry γ-ray emission after nuclear excitation Monitoring of radioactive elements in
environmental samples
Ultraviolet/visible spectrometry Electronic molecular absorption in Quantitative determination of
solution unsaturated organic compounds
Infrared spectrometry Vibrational molecular absorption Identification of organic compounds
Nuclear magnetic resonance Nuclear absorption (change of spin Identification and structural analysis
spectrometry states) of organic compounds
Mass spectrometry Ionization and fragmentation of Identification and structural analysis
molecules of organic compounds
A3 – Analytical techniques and methods 7

Table 3. Separation techniques and principal applications

Technique Basis Principal applications

冧
Thin-layer chromatography Qualitative analysis of mixtures
Differential rates of migration of
Gas chromatography Quantitative and qualitative
analytes through a stationary phase
determination of volatile compounds
by movement of a liquid or gaseous
High-performance liquid mobile phase Quantitative and qualitative
chromatography determination of nonvolatile
compounds
Electrophoresis Differential rates of migration of Quantitative and qualitative
analytes through a buffered medium determination of ionic compounds

lists of chemicals and reagents to be used, laboratory apparatus and glassware,

and appropriate instrumentation. The quality and sources of chemicals,
including solvents, and the required performance characteristics of instruments
will also be specified as will the procedure for obtaining a representative sample
of the material to be analyzed. This is of crucial importance in obtaining mean-
ingful results (Topic A4). The preparation or pre-treatment of the sample will be
followed by any necessary standardization of reagents and/or calibration of
instruments under specified conditions (Topic A5). Qualitative tests for the
analyte(s) or quantitative measurements under the same conditions as those used
for standards complete the practical part of the method. The remaining steps will
be concerned with data processing, computational methods for quantitative
analysis and the formatting of the analytical report. The statistical assessment of
quantitative data is vital in establishing the reliability and value of the data, and
the use of various statistical parameters and tests is widespread (Section B).
Many standard analytical methods have been published as papers in analyt-
ical journals and other scientific literature, and in textbook form. Collections by
trades associations representing, for example, the cosmetics, food, iron and steel,
pharmaceutical, polymer plastics and paint, and water industries are available.
Standards organizations and statutory authorities, instrument manufacturers’
applications notes, the Royal Society of Chemistry and the US Environmental
Protection Agency are also valuable sources of standard methods. Often, labora-
tories will develop their own in-house methods or adapt existing ones for
specific purposes. Method development forms a significant part of the work of
most analytical laboratories, and method validation and periodic revalidation is
a necessity.
Selection of the most appropriate analytical method should take into account
the following factors:
● the purpose of the analysis, the required time scale and any cost constraints;
● the level of analyte(s) expected and the detection limit required;
● the nature of the sample, the amount available and the necessary sample
preparation procedure;
● the accuracy required for a quantitative analysis;
● the availability of reference materials, standards, chemicals and solvents,
instrumentation and any special facilities;
● possible interference with the detection or quantitative measurement of
the analyte(s) and the possible need for sample clean-up to avoid matrix
interference;
8 Section A – The nature and scope of analytical chemistry

● the degree of selectivity available − methods may be selective for a small

number of analytes or speciﬁc for only one;
● quality control and safety factors.

Method validation Analytical methods must be shown to give reliable data, free from bias and suit-
able for the intended use. Most methods are multi-step procedures, and the
process of validation generally involves a stepwise approach in which optimized
experimental parameters are tested for robustness (ruggedness), that is sensi-
tivity to variations in the conditions, and sources of errors investigated.
A common approach is to start with the final measurement stage, using cali-
bration standards of known high purity for each analyte to establish the perfor-
mance characteristics of the detection system (i.e. specificity, range, quantitative
response (linearity), sensitivity, stability and reproducibility). Robustness in
terms of temperature, humidity and pressure variations would be included at
this stage, and a statistical assessment made of the reproducibility of repeated
identical measurements (replicates). The process is then extended backwards in
sequence through the preceding stages of the method, checking that the optimum
conditions and performance established for the final measurement on analyte
calibration standards remain valid throughout. Where this is not the case, new
conditions must be investigated by modification of the procedure and the process
repeated. A summary of this approach is shown in Figure 1 in the form of a flow
diagram. At each stage, the results are assessed using appropriate statistical tests
(Section B) and compared for consistency with those of the previous stage. Where
unacceptable variations arise, changes to the procedure are implemented and the
assessment process repeated. The performance and robustness of the overall
method are finally tested with field trials in one or more routine analytical
laboratories before the method is considered to be fully validated.
A3 – Analytical techniques and methods 9

Step 1 Performance characteristics of detector

for single analyte calibration standards

Step 2 Process repeated for mixed analyte

calibration standards

Step 3 Process repeated for analyte calibration

standards with possible interfering
substances and for reagent blanks

Step 4 Process repeated for analyte calibration

standards with anticipated matrix
components to evaluate matrix
interference

Step 5 Analysis of 'spiked' simulated sample

matrix. i.e. matrix with added known
amounts of analyte(s), to test recoveries

Step 6 Field trials in routine laboratory with

more junior personnel to test ruggedness

Fig. 1. Flow chart for method validation.

Section A – The nature and scope of analytical chemistry

A4 S AMPLING AND SAMPLE

HANDLING

Key Notes
Representative A representative sample is one that truly reﬂects the composition of the
sample material to be analyzed within the context of a deﬁned analytical
problem.

Sample storage Due to varying periods of time that may elapse between sample
collection and analysis, storage conditions must be such as to avoid
undesirable losses, contamination or other changes that could affect the
results of the analysis.

Sample Preliminary treatment of a sample is sometimes necessary before it is in a

pre-treatment suitable form for analysis by the chosen technique and method. This may
involve a separation or concentration of the analytes or the removal of
matrix components that would otherwise interfere with the analysis.

Sample preparation Samples generally need to be brought into a form suitable for
measurements to be made under controlled conditions. This may involve
dissolution, grinding, fabricating into a speciﬁc size and shape,
pelletizing or mounting in a sample holder.

Representative The importance of obtaining a representative sample for analysis cannot be

sample overemphasized. Without it, results may be meaningless or even grossly
misleading. Sampling is particularly crucial where a heterogeneous material is to
be analyzed. It is vital that the aims of the analysis are understood and an appro-
priate sampling procedure adopted. In some situations, a sampling plan or
strategy may need to be devised so as to optimize the value of the analytical
information collected. This is necessary particularly where environmental
samples of soil, water or the atmosphere are to be collected or a complex indus-
trial process is to be monitored. Legal requirements may also determine a
sampling strategy, particularly in the food and drug industries. A small sample
taken for analysis is described as a laboratory sample. Where duplicate analyses
or several different analyses are required, the laboratory sample will be divided
into sub-samples which should have identical compositions.
Homogeneous materials (e.g., single or mixed solvents or solutions and most
gases) generally present no particular sampling problem as the composition of
any small laboratory sample taken from a larger volume will be representative of
the bulk solution. Heterogeneous materials have to be homogenized prior to
obtaining a laboratory sample if an average or bulk composition is required.
Conversely, where analyte levels in different parts of the material are to be
A4 – Sampling and sample handling 11

measured, they may need to be physically separated before laboratory samples

are taken. This is known as selective sampling. Typical examples of hetero-
geneous materials where selective sampling may be necessary include:
● surface waters such as streams, rivers, reservoirs and seawater, where the
concentrations of trace metals or organic compounds in solution and in sedi-
ments or suspended particulate matter may each be of importance;
● materials stored in bulk, such as grain, edible oils, or industrial organic chem-
icals, where physical segregation (stratiﬁcation) or other effects may lead to
variations in chemical composition throughout the bulk;
● ores, minerals and alloys, where information about the distribution of a partic-
ular metal or compound is sought;
● laboratory, industrial or urban atmospheres where the concentrations of toxic
vapors and fumes may be localized or vary with time.
Obtaining a laboratory sample to establish an average analyte level in a highly
heterogeneous material can be a lengthy procedure. For example, sampling a
large shipment of an ore or mineral, where the economic cost needs to be
determined by a very accurate assay, is typically approached in the following
manner.
(i) Relatively large pieces are randomly selected from different parts of the
shipment.
(ii) The pieces are crushed, ground to coarse granules and thoroughly mixed.
(iii) A repeated coning and quartering process, with additional grinding to
reduce particle size, is used until a laboratory-sized sample is obtained.
This involves creating a conical heap of the material, dividing it into four
equal portions, discarding two diagonally opposite portions and forming a
new conical heap from the remaining two quarters. The process is then
repeated as necessary (Fig. 1).

2
2 2

1 3 1 3 1 3

4 4
4

Fig. 1. A diagrammatic representation of coning and quartering (quarters 1 and 3, or 2 and 4 are discarded each time).
12 Section A – The nature and scope of analytical chemistry

The distribution of toxic heavy metals or organic compounds in a land rede-

velopment site presents a different problem. Here, to economize on the number
of analyses, a grid is superimposed on the site dividing it up into approximately
one- to five-metre squares. From each of these, samples of soil will be taken at
several specified depths. A three-dimensional representation of the distribution
of each analyte over the whole site can then be produced, and any localized high
concentrations, or hot spots, can be investigated by taking further, more closely-
spaced, samples. Individual samples may need to be ground, coned and
quartered as part of the sampling strategy.
Repeated sampling over a period of time is a common requirement. Examples
include the continuous monitoring of a process stream in a manufacturing plant
and the frequent sampling of patients’ body fluids for changes in the levels of
drugs, metabolites, sugars or enzymes, etc., during hospital treatment. Studies of
seasonal variations in the levels of pesticide, herbicide and fertilizer residues in
soils and surface waters, or the continuous monitoring of drinking water supplies
are two further examples.
Having obtained a representative sample, it must be labeled and stored under
appropriate conditions. Sample identification through proper labeling, increas-
ingly done by using bar codes and optical readers under computer control, is an
essential feature of sample handling.

Sample storage Samples often have to be collected from places remote from the analytical labora-
tory and several days or weeks may elapse before they are received by the labo-
ratory and analyzed. Furthermore, the workload of many laboratories is such that
incoming samples are stored for a period of time prior to analysis. In both
instances, sample containers and storage conditions (e.g., temperature, humidity,
light levels and exposure to the atmosphere) must be controlled such that no
signiﬁcant changes occur that could affect the validity of the analytical data. The
following effects during storage should be considered:

● increases in temperature leading to the loss of volatile analytes, thermal or

biological degradation, or increased chemical reactivity;
● decreases in temperature that lead to the formation of deposits or the precipi-
tation of analytes with low solubilities;
● changes in humidity that affect the moisture content of hygroscopic solids and
liquids or induce hydrolysis reactions;
● UV radiation, particularly from direct sunlight, that induces photochemical
reactions, photodecomposition or polymerization;
● air-induced oxidation;
● physical separation of the sample into layers of different density or changes in
crystallinity.

In addition, containers may leak or allow contaminants to enter.

A particular problem associated with samples having very low (trace and
ultra-trace) levels of analytes in solution is the possibility of losses by adsorp-
tion onto the walls of the container or contamination by substances being
leached from the container by the sample solvent. Trace metals may be depleted
by adsorption or ion-exchange processes if stored in glass containers, whilst
sodium, potassium, boron and silicates can be leached from the glass into the
sample solution. Plastic containers should always be used for such samples.
A4 – Sampling and sample handling 13

Conversely, sample solutions containing organic solvents and other organic

liquids should be stored in glass containers because the base plastic or additives
such as plasticizers and antioxidants may be leached from the walls of plastic
containers.

Sample pre- Samples arriving in an analytical laboratory come in a very wide assortment of
treatment sizes, conditions and physical forms and can contain analytes from major
constituents down to ultra-trace levels. They can have a variable moisture content
and the matrix components of samples submitted for determinations of the same
analyte(s) may also vary widely. A preliminary, or pre-treatment, is often used to
condition them in readiness for the application of a speciﬁc method of analysis or
to pre-concentrate (enrich) analytes present at very low levels. Examples of pre-
treatments are:

● drying at 100°C to 120°C to eliminate the effect of a variable moisture content;

● weighing before and after drying enables the water content to be calculated or
it can be established by thermogravimetric analysis (Topic G1);
● separating the analytes into groups with common characteristics by dis-
tillation, ﬁltration, centrifugation, solvent or solid phase extraction (Topic
D1);
● removing or reducing the level of matrix components that are known to cause
interference with measurements of the analytes;
● concentrating the analytes if they are below the concentration range of the
analytical method to be used by evaporation, distillation, co-precipitation, ion
exchange, solvent or solid phase extraction or electrolysis.

Sample clean-up in relation to matrix interference and to protect special-

ized analytical equipment such as chromatographic columns and detection
systems from high levels of matrix components is widely practised using solid
phase extraction (SPE) cartridges (Topic D1). Substances such as lipids, fats,
proteins, pigments, polymeric and tarry substances are particularly detri-
mental.

Sample A laboratory sample generally needs to be prepared for analytical measurement

preparation by treatment with reagents that convert the analyte(s) into an appropriate chem-
ical form for the selected technique and method, although in some instances it is
examined directly as received or mounted in a sample holder for surface
analysis. If the material is readily soluble in aqueous or organic solvents, a simple
dissolution step may suffice. However, many samples need first to be decom-
posed to release the analyte(s) and facilitate specific reactions in solution. Sample
solutions may need to be diluted or concentrated by enrichment so that analytes
are in an optimum concentration range for the method. The stabilization of solu-
tions with respect to pH, ionic strength and solvent composition, and the removal
or masking of interfering matrix components not accounted for in any pre-treat-
ment may also be necessary. An internal standard for reference purposes in
quantitative analysis (Topic A5 and Section B) is sometimes added before adjust-
ment to the final prescribed volume. Some common methods of decomposition
and dissolution are given in Table 1.
14 Section A – The nature and scope of analytical chemistry

Table 1. Some methods for sample decomposition and dissolution

Method of attack Type of sample
Heated with concentrated mineral Geological, metallurgical
acids (HCl, HNO3, aqua regia) or
strong alkali, including microwave
digestion
Fusion with ﬂux (Na2O2, Na2CO3, Geological, refractory materials
LiBO2, KHSO4, KOH)
Heated with HF and H2SO4 or HClO4 Silicates where SiO2 is not the analyte
Acid leaching with HNO3 Soils and sediments
Dry oxidation by heating in a furnace Organic materials with inorganic analytes
or wet oxidation by boiling with
concentrated H2SO4 and HNO3 or HClO4
Section A – The nature and scope of analytical chemistry

A5 C ALIBRATION AND
STANDARDS

Key Notes

Calibration Calibration or standardization is the process of establishing the response

of a detection or measurement system to known amounts or
concentrations of an analyte under speciﬁed conditions, or the
comparison of a measured quantity with a reference value.

Chemical standard A chemical standard is a material or substance of very high purity

and/or known composition that is used to standardize a reagent or
calibrate an instrument.

Reference material A reference material is a material or substance, one or more properties of

which are sufﬁciently homogeneous and well established for it to be used
for the calibration of apparatus, the assessment of a measurement method
or for assigning values to materials.

Table 1. Some primary standards used in titrimetric analysis

Type of titration Primary standard
Acid-base Sodium carbonate, Na2CO3
Sodium tetraborate, Na2B4O7.10H2O
Potassium hydrogen phthalate, KH(C8H4O4)
Benzoic acid, C6H5COOH
Redox Potassium dichromate, K2Cr2O7
Potassium iodate, KIO3
Sodium oxalate, Na2C2O4
Precipitation (silver halide) Silver nitrate, AgNO3
Sodium chloride, NaCl
Complexometric (EDTA) Zinc, Zn
Magnesium, Mg
EDTA (disodium salt), C10H14N2O8Na2

Bureau of Analytical Standards. CRMs or SRMs are produced in various forms

and for different purposes and they may contain one or more certified compo-
nents, such as
● pure substances or solutions for calibration or identification;
● materials of known matrix composition to facilitate comparisons of analytical
data;
● materials with approximately known matrix composition and specified
components.
They have a number of principal uses, including
● validation of new methods of analysis;
● standardization/calibration of other reference materials;
● confirmation of the validity of standardized methods;
● support of quality control and quality assurance schemes.
Section A – The nature and scope of analytical chemistry

A6 Q UALITY IN ANALYTICAL
LABORATORIES

Key Notes

Quality control Quality control (QC) is the process of ensuring that the operational
techniques and activities used in an analytical laboratory provide results
suitable for the intended purpose.

Quality assurance Quality assurance (QA) is the combination of planned and systematic
actions necessary to provide adequate confidence that the process of
quality control satisfies specified requirements.

Accreditation This is a system whereby the quality control and quality assurance
system procedures adopted by a laboratory are evaluated by inspection and
accredited by an independent body.

● provision of full documentation and traceability of results to recognized

reference materials through recorded identification;
● maintenance and calibration of instrumentation to manufacturers’ specifica-
tions;
● management and control of laboratory chemicals and other materials including
checks on quality;
● adequate training of laboratory personnel to ensure understanding and
competence;
● external verification of results wherever possible;
● accreditation of the laboratory by an independent organization.

Quality assurance The overall management of an analytical laboratory should include the provision
of evidence and assurances that appropriate QC procedures for laboratory activ-
ities are being correctly implemented. Quality assurance (QA) is a managerial
responsibility that is designed to ensure that this is the case and to generate
confidence in the analytical results. Part of QA is to build confidence through the
laboratory participating in interlaboratory studies where several laboratories
analyze one or more identical homogeneous materials under specified condi-
tions. Proficiency testing is a particular type of study to assess the performance
of a laboratory or analyst relative to others, whilst method performance studies
and certification studies are undertaken to check a particular analytical method
or reference material respectively. The results of such studies and their statistical
assessment enable the performances of individual participating laboratories to be
demonstrated and any deficiencies in methodology and the training of personnel
to be addressed.

Accreditation Because of differences in the interpretation of the term quality, which can be
system defined as fitness for purpose, QC and QA systems adopted by analyical labora-
tories in different industries and fields of activity can vary widely. For this
reason, defined quality standards have been introduced by a number of organi-
zations throughout the world. Laboratories can design and implement their own
quality systems and apply to be inspected and accredited by the organization for
the standard most appropriate to their activity. A number of organizations that
offer accreditation suitable for analytical laboratories and their corresponding
quality standards are given in Table 1.

Table 1. Accreditation organizations and their quality standards

Name of accreditation organization Quality standard
Organization for Economic Co-operation Good Laboratory Practice (GLP)
and Development (OECD)
The International Organization for ISO 9000 series of quality standards
Standardization (ISO) ISO Guide 25 general requirements for
competence of calibration and testing
laboratories
European Committee for Standardization EN 29000 series
(CEN) EN 45000 series
British Standards Institution (BSI) BS 5750 quality standard
BS 7500 series
National Measurement Accreditation NAMAS
Service (NAMAS)
Section B – Assessment of data

B1 E RRORS IN ANALYTICAL
MEASUREMENTS

Key Notes

Measurement errors All measurement processes are subject to measurement errors that affect
numerical data and which arise from a variety of sources.

Absolute and An absolute error is the numerical difference between a measured value
relative errors and a true or accepted value. A relative error is the absolute error divided
by the true or accepted value.

Determinate errors Also known as systematic errors, or bias, these generally arise from
determinate or identiﬁable sources causing measured values to differ
from a true or accepted value.

Indeterminate errors Also known as random errors, these arise from a variety of uncontrolled
sources and cause small random variations in a measured quantity when
the measurement is repeated a number of times.

Accumulated errors Where several different measurements are combined to compute an

overall analytical result, the errors associated with each individual
measurement contribute to a total or accumulated error.

Determinate errors can be proportional to the size of sample taken for

analysis. If so, they will have the same effect on the magnitude of a result
regardless of the size of the sample, and their presence can thus be difﬁcult to
detect. For example, copper(II) can be determined by titration after reaction with
potassium iodide to release iodine according to the equation
2Cu2+ + 4I- Æ 2CuI + I2
However, the reaction is not speciﬁc to copper(II), and any iron(III) present in
the sample will react in the same way. Results for the determination of copper in
an alloy containing 20%, but which also contained 0.2% of iron are shown in
Figure 2 for a range of sample sizes. The same absolute error of +0.2% or relative
error of 1% (i.e. a positive bias) occurs regardless of sample size, due to the
presence of the iron. This type of error may go undetected unless the
constituents of the sample and the chemistry of the method are known.

21
Copper found (%)

Positive
bias
20
True value

19
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7
Sample size (g)
Fig. 2. Effect of a proportional error on the determination of copper by titration in the
presence of iron.

Constant determinate errors are independent of sample size, and therefore

become less signiﬁcant as the sample size is increased. For example, where a
visual indicator is employed in a volumetric procedure, a small amount of
titrant is required to change the color at the end-point, even in a blank solution
(i.e. when the solution contains none of the species to be determined). This
indicator blank (Topic C5) is the same regardless of the size of the titer when
the species being determined is present. The relative error, therefore, decreases
with the magnitude of the titer, as shown graphically in Figure 3. Thus, for an
indicator blank of 0.02 cm3, the relative error for a 1 cm3 titer is 2%, but this falls
to only 0.08% for a 25 cm3 titer.

Indeterminate Known also as random errors, these arise from random ﬂuctuations in
errors measured quantities, which always occur even under closely controlled condi-
tions. It is impossible to eliminate them entirely, but they can be minimized by
careful experimental design and control. Environmental factors such as temper-
ature, pressure and humidity, and electrical properties such as current, voltage
and resistance are all susceptible to small continuous and random variations
described as noise. These contribute to the overall indeterminate error in any
24 Section B – Assessment of data

2.5

Relative error (%)

1.5

0.5

0
0 10 20 30
Size of titer (cm3)

Fig. 3. Effect of a constant error on titers of differing magnitudes.

physical or physico-chemical measurement, but no one speciﬁc source can be

identiﬁed.
A series of measurements made under the same prescribed conditions and
represented graphically is known as a frequency distribution. The frequency of
occurrence of each experimental value is plotted as a function of the magnitude
of the error or deviation from the average or mean value. For analytical data,
the values are often distributed symmetrically about the mean value, the most
common being the normal error or Gaussian distribution curve. The curve
(Fig. 4) shows that
● small errors are more probable than large ones,
● positive and negative errors are equally probable, and
● the maximum of the curve corresponds to the mean value.
The normal error curve is the basis of a number of statistical tests that can be
applied to analytical data to assess the effects of indeterminate errors, to compare
values and to establish levels of conﬁdence in results (Topics B2 and B3).
Frequency of occurrence
of each deviation

– 0 +
Deviation from mean, µ

Fig. 4. The normal error or Gaussian distribution curve.

B1 – Errors in analytical measurements 25

Accumulated Errors are associated with every measurement made in an analytical procedure,
errors and these will be aggregated in the ﬁnal calculated result. The accumulation or
propagation of errors is treated similarly for both determinate (systematic) and
indeterminate (random) errors.
Determinate (systematic) errors can be either positive or negative, hence some
cancellation of errors is likely in computing an overall determinate error, and in
some instances this may be zero. The overall error is calculated using one of two
alternative expressions, that is
● where only a linear combination of individual measurements is required to
compute the result, the overall absolute determinate error, ET, is given by
ET = E1 + E2 + E3 + …….
E1 and E2 etc., being the absolute determinate errors in the individual
measurements taking sign into account
● where a multiplicative expression is required to compute the result, the
overall relative determinate error, ETR, is given by
ETR = E1R + E2R + E3R + …….
E1R and E2R etc., being the relative determinate errors in the individual measure-
ments taking sign into account.
The accumulated effect of indeterminate (random) errors is computed by
combining statistical parameters for each measurement (Topic B2).
Section B – Assessment of data

B2 A SSESSMENT OF ACCURACY
AND PRECISION

Key Notes
Accuracy and Accuracy is the closeness of an experimental measurement or result to
precision the true or accepted value. Precision is the closeness of agreement
between replicated measurements or results obtained under the same
prescribed conditions.

Standard deviation The standard deviation of a set of values is a statistic based on the normal
error (Gaussian) curve and used as a measure of precision.

Relative standard Relative standard deviation (coefﬁcient of variation) is the standard

deviation deviation expressed as a percentage of the measured value.

Pooled standard A standard deviation can be calculated for two or more sets of data by
deviation pooling the values to give a more reliable measure of precision.

Variance This is the square of the standard deviation, which is used in some
statistical tests.

Overall precision An estimate of the overall precision of an analytical procedure can be

made by combining the precisions of individual measurements.

Conﬁdence interval This is the range of values around an experimental result within which
the true or accepted value is expected to lie with a deﬁned level of
probability.

Fig. 1. Plots of titration data to distinguish accuracy and precision.

It should be noted that good precision does not necessarily produce good
accuracy (analyst A) and poor precision does not necessarily produce poor
accuracy (analyst B). However, conﬁdence in the analytical procedure and the
results is greater when good precision can be demonstrated (analyst D).
Accuracy is generally the more important characteristic of quantitative data to
be assessed, although consistency, as measured by precision, is of particular
concern in some circumstances. Trueness is a term associated with accuracy,
which describes the closeness of agreement between the average of a large
number of results and a true or accepted reference value. The degree of accuracy
required depends on the context of the analytical problem; results must be shown
to be ﬁt for the purpose for which they are intended. For example, one result may
be satisfactory if it is within 10% of a true or accepted value whilst it may be
necessary for another to be within 0.5%. By repeating an analysis a number of
times and computing an average value for the result, the level of accuracy will be
improved, provided that no systematic error (bias) has occurred. Accuracy
cannot be established with certainty where a true or accepted value is not known,
as is often the case. However, statistical tests indicating the accuracy of a result
with a given probability are widely used (vide infra).
Precision, which is a measure of the variability or dispersion within a set of
replicated values or results obtained under the same prescribed conditions, can
be assessed in several ways. The spread or range (i.e. the difference between the
highest and lowest value) is sometimes used, but the most popular method is to
estimate the standard deviation of the data (vide infra). The precision of results
obtained within one working session is known as repeatability or within-run
precision. The precision of results obtained over a series of working sessions is
known as reproducibility or between-runs precision. It is sometimes necessary
to separate the contributions made to the overall precision by within-run and
28 Section B – Assessment of data

between-runs variability. It may also be important to establish the precision of

individual steps in an analysis.

Standard This is the most widely used measure of precision and is a parameter of the
deviation normal error or Gaussian curve (Topic B1, Fig. 4). Figure 2 shows two curves for
the frequency distribution of two theoretical sets of data, each having an inﬁnite
number of values and known as a statistical population.
Frequency of occurrence

sd = s2
of each deviation

s1 > s2

sd = s1

– m +
Deviation from mean

Fig. 2. Normal error or Gaussian curves for the frequency distributions of two statistical
populations with differing spreads.

The maximum in each curve corresponds to the population mean, which for
these examples has the same value, m. However, the spread of values for the
two sets is quite different, and this is reflected in the half-widths of the two
curves at the points of inflection, which, by definition, is the population stan-
dard deviation, s. As s2 is much less than s1, the precision of the second set is
much better than that of the first. The abscissa scale can be calibrated in
absolute units or, more commonly, as positive and negative deviations from the
mean, m.
In general, the smaller the spread of values or deviations, the smaller the
value of s and hence the better the precision. In practice, the true values of m
and s can never be known because they relate to a population of infinite size.
However, an assumption is made that a small number of experimental values or
a statistical sample drawn from a statistical population _ is also distributed
normally or approximately so. The experimental mean, x, of a set of values x1,
x2, x3,…….xn is therefore considered to be an estimate of the true or population
mean, m, and the experimental standard deviation, s, is an estimate of the true
or population standard deviation, s.
A useful property of the normal error curve is that, regardless of the magni-
tude of m and s, the area under the curve within defined limits on either side of
m (usually expressed in multiples of ±s) is a constant proportion of the total
area. Expressed as a percentage of the total area, this indicates that a particular
percentage of the population will be found between those limits.
Thus, approximately 68% of the area, and therefore of the population, will be
B2 – Assessment of accuracy and precision 29

found within ±1s of the mean, approximately 95% will be found within ±2s and
approximately 99.7% within ±3s. More practically convenient levels, as shown
in Figure 3, are those corresponding to 90%, 95% and 99% of the population,
which are deﬁned by ±1.64s, ±1.96s and ±2.58s respectively. Many statistical
tests are based on these probability levels.

80% 90%
frequency (y/N)

frequency (y/N)
Relative

Relative
–1.29s +1.29s –1.64s +1.64s

–4s –3s –2s –1s 0 1s 2s 3s 4s –4s –3s –2s –1s 0 1s 2s 3s 4s

95% 99%
frequency (y/N)

frequency (y/N)
Relative

Relative

–1.96s +1.96s –2.58s +2.58s

–4s –3s –2s –1s 0 1s 2s 3s 4s –4s –3s –2s –1s 0 1s 2s 3s 4s

Fig. 3. Proportions of a population within deﬁned limits of the mean.

The value of the population standard deviation, s, is given by the formula

莦
冪
i=N
冱 (x − m) i
2

i=1
s= (1)
N

where xi represents any individual value in the population and N is the total
number of values, strictly infinite. The summation symbol, S, is used to show
that the numerator of the equation is the sum for i = 1 to i = N of the squares of
the deviations of the individual x values from the population mean, m. For very
large sets of data (e.g., when N >50), it may be justifiable to use this formula as
the difference between s and s will then be negligible. However, most analytical
data consists of sets of values of less than ten and often as small as three.
Therefore, a modified formula is used to calculate an estimated _ standard
deviation, s, to replace s, and using an experimental mean, x, to replace the
population mean, m:

莦
冪
i=N _
冱 (x − x) i
2

i=1
s= (2)
N−1
30 Section B – Assessment of data

Note that N in the denominator is replaced by N - 1, which is known as the

number of degrees_ of freedom and is deﬁned as the number of independent
deviations (xi − x) used to calculate s. For single sets of data, this is always one
less than the number in the set because when N - 1 deviations are known the last
i=N _
one can be deduced as, taking sign into account, 冱 (xi − x) must be zero (see
i=1
Example 1 below).
In summary, the calculation of an estimated standard deviation, s, for a small
number of values involves the following steps:

● calculation of an experimental mean;

● calculation of the deviations of individual xi values from the mean;
● squaring the deviations and summing them;
● dividing by the number of degrees of freedom, N - 1, and
● taking the square root of the result.

Note that if N were used in the denominator, the calculated value of s would be
an underestimate of s.
Estimated standard deviations are easily obtained using a calculator that
incorporates statistical function keys or with one of the many computer soft-
ware packages. It is, however, useful to be able to perform a stepwise arithmetic
calculation, and an example using the set of ﬁve replicate titers by analyst A
(Fig. 1) is shown below.

Example 1
xi /cm3 (xi − x) (xi − x)2
20.16 -0.04 1.6 × 10-3
20.22 +0.02 4 × 10-4
20.18 -0.02 4 × 10-4
20.20 0.00 0
20.24 +0.04 1.6 × 10-3
Â 101.00 4 ¥ 10-3
_
x 20.20

冪莦 4 莦
−3
4 × 10
s = 0.032 cm 3

Relative The relative standard deviation, RSD or sr, is also known as the coefﬁcient of
standard variation, CV. It is a measure of relative precision and is normally expressed as
deviation a percentage of the mean value or result
_
sr (s/x) × 100 (3)
It is an example of a relative error (Topic B1) and is particularly useful for
comparisons between sets of data of differing magnitude or units, and in calcu-
lating accumulated (propagated) errors. The RSD for the data in Example 1 is
given below.

Example 2
0.032
sr × 100 = 0.16%
20.20
B2 – Assessment of accuracy and precision 31

Pooled standard Where replicate samples are analyzed on a number of occasions under the same
deviation prescribed conditions, an improved estimate of the standard deviation can be
obtained by pooling the data from the individual sets. A general formula for the
pooled standard deviation, spooled, is given by the expression

莦莦莦莦
冪
i=N1 _ _i=N2 _ i=N3_ i=Nk
冱 (x − x ) + 冱 (x − x ) + 冱 (x − x ) + ... + 冱 (x − x )
i 1
2
i 2
2
i 3
2
i k
2

i=1 i=1 i=1 i=1

spooled =
i=k
(4)
冱N = k i
i=1

_ _
where N1, N2, N3…Nk are the numbers of results in each of the k sets, and x1, x2,
_ _
x3, . . . xk, are the means for each of the k sets.

Variance The square of the standard deviation, s2, or estimated standard deviation, s2, is
used in a number of statistical computations and tests, such as for calculating
accumulated (propagated) errors (Topic B1 and below) or when comparing the
precisions of two sets of data (Topic B3).

Overall precision Random errors accumulated within an analytical procedure contribute to the
overall precision. Where the calculated result is derived by the addition or
subtraction of the individual values, the overall precision can be found by
summing the variances of all the measurements so as to provide an estimate of
the overall standard deviation, i.e.

兹冢s
苶苶 2苶
soverall = 1 + s2 + s3 + . . .
2 2
冣
Example
In a titrimetric procedure, the buret must be read twice, and the error associated
with each reading must be taken into account in estimating the overall preci-
sion. If the reading error has an estimated standard deviation of 0.02 cm3, then
the overall estimated standard deviation of the titration is given by

soverall =
兹苶
冢0.02 苶
+ 0.02苶冣 0.028 cm
2 2 3

Note that this is less than twice the estimated standard deviation of a single
reading. The overall standard deviation of weighing by difference is estimated
in the same way.
If the calculated result is derived from a multiplicative expression, the overall
relative precision is found by summing the squares of the relative standard devia-
tions of all the measurements, i.e.

sr(overall) =
兹苶
冢s + s苶
+s 苶
+ . . .冣
2 2 2
r1 r2 r3

Conﬁdence The true or accepted mean of a set of experimental results is generally

interval unknown except where a certified reference material is being checked or
analyzed for calibration purposes.
_ In all other cases, an estimate of the accu-
racy of the experimental mean, x,_must be made. This can be done by defining a
range of values on either side of x within which the true mean, m, is expected to
lie with a defined level of probability. This range, which ideally should be as
narrow as possible, is based on the standard deviation and is known as the
32 Section B – Assessment of data

confidence interval, CI, and the upper and lower limits of the range as confi-
dence limits, CL. Confidence limits can be calculated using the standard devia-
tion, s, if it is known, or the estimated standard deviation, s, for the data. In
either case, a probability level must be defined, otherwise the test is of no
value.
When the standard deviation is already known from past history, the confi-
dence limits are given by the equation
_ zs
CL(m) = x ± (5)
兹N
苶
where z is a statistical factor related to the probability level required, usually
90%, 95% or 99%. The values of z for these levels are 1.64, 1.96 and 2.58, respec-
tively, and correspond to the multiples of the standard deviation shown in
Figure 3.
Where an estimated standard deviation is to be used, s is replaced by s,
which must first be calculated from the current data. The confidence limits are
then given by the equation
_ ts
CL(m) = x ± (6)
兹N苶
where z is replaced by an alternative statistical factor, t, also related to the prob-
ability level but in addition determined by the number of degrees of freedom
for the set of data, i.e. one less than the number of results. It should be noted
that (i) the confidence interval is inversely proportional to 兹N 苶, and (ii) the
higher the selected probability level, the greater the confidence interval becomes
as both z and t increase. A probability level of 100 percent is meaningless, as the
confidence limits would then have to be ±•.
The following examples demonstrate the calculation of confidence limits
using each of the two formulae.

Example 3
The chloride content of water samples has been determined a very large number
of times using a particular method, and the standard deviation found to be
7 ppm. Further analysis of a particular sample gave experimental values of
350 ppm for a single determination, for the mean of two replicates and for the
mean of four replicates. Using equation (5), and at the 95% probability level,
z = 1.96 and the conﬁdence limits are:
1.96 × 7
1 determinations CL(m) = 350 ± = 350 ± 14 ppm
兹1苶
1.96 × 7
2 determinations CL(m) = 350 ± = 350 ± 10 ppm
兹2苶
1.96 × 7
4 determinations CL(m) = 350 ± = 350 ± 7 ppm7
兹4苶

Example 4
The same chloride analysis as in Example 3, but using a new method for which
the standard deviation was not known, gave the following replicate results,
mean and estimated standard deviation:
B2 – Assessment of accuracy and precision 33

Chloride/ppm Mean Estimated standard deviation

346 351.67 ppm 6.66 ppm
359
350
Using equation (6), and at the 95% probability level, t = 4.3 for two degrees of
freedom, and the confidence limits are:
4.3 × 6.66
3 determinations CL(m) = 352 ± = 352 ± 17 ppm
兹3苶
The wider limits given by equation (6) when the standard deviation is estimated
with only three results reflects the much greater uncertainty associated with this
value, which in turn affects the confidence in the degree of accuracy. To demon-
strate good accuracy, the confidence interval, CI, should be as small as possible
and increasing the number of replicates will clearly achieve this. However, due
to the 兹N 苶 term in the denominator, to reduce the interval by, say, a factor of
two requires an increase in the number of replicates by a factor of four as shown
by Example 3. Unfortunately, the law of diminishing returns applies here, so if
the CI is to be halved again, the number of replicates must be increased from
four to sixteen. Similarly, in Example 4, the number of replicates would have to
be increased from three to twelve to halve the CI, which would represent an
unacceptable amount of time and money for most analytical laboratories.
Section B – Assessment of data

B3 S IGNIFICANCE TESTING

Key Notes

Signiﬁcance tests These are statistical tests used to compare individual values or sets of
values for signiﬁcant differences.

Outliers A measurement or result that appears to differ signiﬁcantly from others

in the same set of replicates is described as an outlier.

Q-test The Q-test is used to determine whether to reject or retain a suspected

outlier.

F-test The F-test enables the precisions of two sets of data to be compared using
their variances.

t-test The t-test is used to compare two experimental means, an experimental

mean with a known value or sets of pairs of experimental values.

Analysis of variance F-tests can be applied to several sets of data to assess and compare
different sources of variability.

Signiﬁcance Signiﬁcance tests involve a comparison between a calculated experimental

tests factor and a tabulated factor determined by the number of values in the set(s) of
experimental data and a selected probability level that the conclusion is correct.
They are used for several purposes, such as:
● to check individual values in a set of data for the presence of determinate
errors (bias);
● to compare the precision of two or more sets of data using their variances;
● to compare the means of two or more sets of data with one another or with
known values to establish levels of accuracy.
Tests are based on a null hypothesis - an assumption that there is no signifi-
cant difference between the values being compared. The hypothesis is accepted
if the calculated experimental factor is less than the corresponding tabulated
factor, otherwise it is rejected and there is said to be a significant difference
between the values at the selected probability level. The conclusion should
always be stated clearly and unambiguously.
Probability levels of 90%, 95% and 99% are generally considered appropriate
for most purposes, but it should be remembered that there are also corre-
sponding 10%, 5% or 1% probabilities, respectively, of the opposite conclusion
being valid. For example, if a test indicates that the null hypothesis is correct
and that there is no significant difference between two values at the 95% proba-
bility level, it also allows the possibility that there is a significant difference at
the 5% level.
B3 – Significance testing 35

Separate tabular values for some significance test factors have been compiled
for what are described as one-tailed and two-tailed tests. The exact purpose of
the comparison that is to be made determines which table to use.
● The one-tailed test is used EITHER to establish whether one experimental
value is significantly greater than the other OR the other way around.
● The two-tailed test is used to establish whether there is a significant differ-
ence between the two values being compared, whether one is higher or
lower than the other not being specified.
The two-tailed test is by far the most widely used. Examples are given below.

Outliers Inspection of a set of replicate measurements or results may reveal that one or
more is considerably higher or lower than the remainder and appears to be outside
the range expected from the inherent effects of indeterminate (random) errors
alone. Such values are termed outliers, or suspect values, because it is possible
that they may have a bias due to a determinate error. On occasions, the source of
error may already be known or it is discovered on investigation, and the outlier(s)
can be rejected without recourse to a statistical test. Frequently, however, this is
not the case, and a test of signiﬁcance such as the Q-test should be applied to a
suspect value to determine whether it should be rejected and therefore not included
in any further computations and statistical assessments of the data.

Q-test Also known as Dixon’s Q-test, this is one of several that have been devised to
test suspected outliers in a set of replicates. It involves the calculation of a ratio,
Qexptl, deﬁned as the absolute difference between a suspect value and the value
closest to it divided by the spread of all the values in the set:
Qexptl = suspect value − nearest value /(largest value − smallest value)
Qexptl is then compared with a tabulated value, Qtab, at a selected level of proba-
bility, usually 90% or 95%, for a set of n values (Table 1). If Qexptl is less than Qtab,
then the null hypothesis that there is no signiﬁcant difference between the
suspect value and the other values in the set is accepted, and the suspect value
is retained for further data processing. However, if Qexptl is greater than Qtab, then
the suspect value is regarded as an outlier and is rejected. A rejected value
should NOT be used in the remaining calculations.

Table 1. Critical values of Q at the 95% (P = 0.05) level for a two-tailed test
Sample size Critical value
4 0.831
5 0.717
6 0.621
7 0.570
8 0.524

Example 1
Four replicate values were obtained for the determination of a pesticide in river
water
0.403, 0.410, 0.401, 0.380 mg dm-3
36 Section B – Assessment of data

Inspection of the data suggests that 0.380 mg dm-3 is a possible outlier.

Qexptl = 0.380 - 0.401 /(0.410 - 0.380) = 0.021/0.03 = 0.70
Qtab = 0.83 for four values at the 95% probability level
As Qexptl is less than Qtab, 0.380 mg dm-3 is not an outlier at the 95% level and
should be retained.

Example 2
If, in Example 1, three additional values of 0.400, 0.413 and 0.411 mg dm-3 were
included, 0.380 mg dm-3 is still a possible outlier.
Qexptl = 0.380 - 0.400 /(0.413 - 0.380) = 0.020/0.033 = 0.61
Qtab = 0.57 for seven values at the 95% probability level
Now, as Qexptl is greater than Qtab, 0.380 mg dm-3 is an outlier at the 95% level and
should be rejected. Note that because the three additional values are all around
0.4 mg dm-3, the suspect value of 0.380 mg dm-3 appears even more anomalous.

F-test This test is used to compare the precisions of two sets of data which may origi-
nate from two analysts in the same laboratory, two different methods of analysis
for the same analyte or results from two different laboratories. A statistic, F, is
defined as the ratio of the population variances, s12/s22, or the sample variances,
s12/s22, of the two sets of data where the larger variance is always placed in the
numerator so that F ≥1.
If the null hypothesis is true, the variances are equal and the value of F will be
one or very close to it. As for the Q-test, an experimental value, Fexptl, is calcu-
lated and compared with a tabulated value, Ftab, at a defined probability level,
usually 90% or 95%, and for the number of degrees of freedom, N - 1, for each
set of data. If Fexptl is less than Ftab, then the null hypothesis that there is no
significant difference between the two variances and hence between the preci-
sion of the two sets of data, is accepted. However, if Fexptl is greater than Ftab,
there is a significant difference between the two variances and hence between
the precisions of the two sets of data.
Some values of Ftab at the 95% probability level are given in Table 2. The
columns in the table correspond to the numbers of degrees of freedom for the
numerator set of data, while the rows correspond to the number of degrees of
freedom for the denominator set. Two versions of the table are available,
depending on the exact purpose of the comparison to be made: a one-tailed F-
test will show whether the precision of one set of data is significantly better than
the other, while a two-tailed F-test will show whether the two precisions are
significantly different.

Table 2. Critical values of F at the 95% (P = 0.05) level for a two-tailed test
n1 5 7 9
n2
5 7.146 6.853 6.681
7 5.285 4.995 4.823
9 4.484 4.197 4.026

n1 = number of degrees of freedom of the numerator. n2 = number of degrees of freedom of the

denominator
B3 – Signiﬁcance testing 37

The application of a two-tailed F-test is demonstrated by the following

example.

Example 3
A proposed new method for the determination of sulfate in an industrial waste
efﬂuent is compared with an existing method, giving the following results:

Method Mean/g dm-3 No. of No. of degrees s/mg dm-3

replicates of freedom
Existing 72 8 7 3.38
New 72 8 7 1.50

Is there a signiﬁcant difference between the precisions of the two methods?

Fexptl
⁄
s2existing
s2
new
(3.38)2
2 5.08
(1.50)

The two-tailed tabular value for F with 7 degrees of freedom for both the
numerator and the denominator is
F7,7 = 5.00 at the 95% probability level
As Fexptl is greater than Ftab, the null hypothesis is rejected; the two methods are
giving signiﬁcantly different precisions.

t-test This test is used to compare the experimental means of two sets of data or to
compare the experimental mean of one set of data with a known or reference
value. A statistic, t, is deﬁned, depending on the circumstances, by one of three
alternative equations.
_ _
Comparison of two experimental means, xA and xB
_ _
冢冣
1
(xA − xB) NM ⁄2
t= × (1)
spooled N+M
where spooled is the pooled estimated standard deviation (Topic B2) for sets A and
B, and N and M are the numbers of values in sets A and B respectively. If N = M,
1

then the second term reduces to (N/2) ⁄ . A simpliﬁed version of equation (4),
2

Topic B2, can be used to calculate spooled as there are only two sets of data.

冦冤冥冤冥冧
1
⁄2
spooled = (N − 1)s2A + (M − 1)s2B / N + M − 2 (2)

In some circumstances, the use of equation (1) may not be appropriate for the
comparison of two experimental means. Examples of when this may be the case
are if
● the amount of sample is so restricted as to allow only one determination by
each of the two methods;
● the methods are to be compared for a series of samples containing different
levels of analyte rather than replicating the analysis at one level only;
● samples are to be analyzed over a long period of time when the same experi-
mental conditions cannot be guaranteed.
It may therefore be essential or convenient to pair the results (one from each
method) and use a paired t-test where t is deﬁned by
38 Section B – Assessment of data

_
x
t = d × N ⁄
1
2
(3)
sd
_
xd being the mean difference between paired values and sd the estimated
standard deviation of the differences.

Comparison of one experimental mean with a known value, m

_
(x − m)
t= ×N⁄
1
2
(4)
s
Using the appropriate equation, an experimental value, texptl, is calculated and
compared with a tabulated value, ttab, at a defined probability level, usually
between 90 and 99%, and for N - 1 degrees of freedom (equations (3) and (4)) or
(N + M - 2) degrees of freedom (equation (1)). If texptl is less than ttab, then the null
hypothesis that there is no significant difference between the two experimental
means or between the experimental mean and a known value is accepted, i.e.
there is no evidence of a bias. However, if texptl is greater than ttab, there is a
significant difference indicating a bias.
Both one-tailed and two-tailed t-tests can be used, depending on circum-
stances, but two-tailed are often preferred (Table 3). The application of all three
t-test equations is demonstrated by the following examples.

Table 3. Critical values of t at the 95% and 99% (P = 0.05 and 0.01) levels for a
two-tailed test
Number of degrees of freedom 95 percent level 99 percent level
2 4.30 9.92
5 2.57 4.03
10 2.23 3.10
18 2.10 2.88

Example 1
Two methods for the determination of polyaromatic hydrocarbons in soils were
compared by analyzing a standard with the following results:
No. of determinations by each method: 10
No. of degrees of freedom: 18
_
UV spectrophotometry: x_ = 28.00 mg kg-1 s = 0.30 mg kg-1
Fluorimetry: x = 26.25 mg kg-1 s = 0.23 mg kg-1
Do the mean results for the two methods differ signiﬁcantly?
Equation (2) is ﬁrst used to calculate a pooled standard deviation:

冦冤冥冤冥冧 = {(9 ¥ 0.3 + 9 ¥ 0.23 )/18}

1
⁄2
(N − 1)s2A + (M − 1)s2B / N + M − 2
1
2 2 ⁄2
spooled =

spooled = 0.267 mg kg−1

Then equation (1) is used to evaluate texptl
_ _
(xA − xB)
冢冣
1
NM ⁄ 2

texptl = × = {(28.0 - 26.25)/0.267} ¥ 5 ⁄ = 14.7

1
2

spooled N+M
For 18 degrees of freedom, the two-tailed value of ttab at the 95% probability
level is 2.10, and at the 99% level it is 2.88.
B3 – Signiﬁcance testing 39

As texptl is greater than ttab at both the 95 and 99% probability levels, there is a
signiﬁcant difference between the means of the two methods.

Example 2
A new high performance liquid chromatographic method for the determination
of pseudoephedrine in a pharmaceutical product at two different levels was
compared with an established method with the following results:
Pseudoephedrine per dose (mg)
Method 1 Method 2
59.9 58.6
59.3 58.3
60.4 60.5
30.7 29.4
30.2 30.4
30.1 28.9

Do the means of the two methods differ signiﬁcantly?

Because the two levels of pseudoephedrine differ considerably, equation (3)
for a paired t-test is used to calculate texptl. The differences between the pairs of
values are 1.3, 1.0, -0.1, 1.3, -0.2 and 1.2 mg per dose, and the estimated
standard deviation of the differences from their mean of 0.750 mg per dose is
0.706 mg per dose. Substitution of these values into the equation gives
_
xd
texptl = × N ⁄ = (0.750/0.706) ¥ 6 ⁄ = 2.60
1 1
2 2

sd
For 5 degrees of freedom, the two-tailed value of ttab at the 95% probability level
is 2.57. As texptl is greater than ttab, there is a signiﬁcant difference between the
means of the two methods. (Note: using equation (1) would give a texptl value of
0.08 and an incorrect conclusion.)

Example 3
A method for the determination of mercury by atomic absorption spectrometry
gave values of 400, 385 and 382 ppm for a standard known to contain 400 ppm.
Does the mean value differ signiﬁcantly from the true value, or is there any
evidence of systematic error (bias)?
_
x = 389 ppm s = 9.64 ppm m = 400 ppm
Using equation (4) to evaluate texptl
_
(x − m) (389 − 400)
texptl = × N ⁄ = × 3 ⁄ = 1.98
1 1
2 2

s 9.64
For 2 degrees of freedom, the two-tailed ttab value is 4.30 at the 95% probability
level. As texptl is less than the two-tailed value of ttab, the mean is not signiﬁcantly
different from the true value. There is, therefore, no evidence of a systematic
error, or bias.

Analysis of Analysis of variance, also known as ANOVA, is a statistical technique for inves-
variance tigating different sources of variability associated with a series of results. It
enables the effect of each source to be assessed separately and compared with
the other(s) using F-tests. Indeterminate or random errors affect all measure-
ments, but additional sources of variability may also arise. The additional
sources can be divided into two types:
40 Section B – Assessment of data

● additional random effects, described as random-effect factors;

● specific effects from determinate sources, described as controlled or fixed-
effect factors.
Where one additional effect may be present, a one-way ANOVA is used, whilst
for two additional effects, two-way ANOVA is appropriate. Both involve much
lengthier calculations than the simpler tests of significance, but facilities for
these are available with computer packages such as Microsoft Excel and
Minitab.
Typical examples of the use of ANOVA are:
● analysis of a heterogeneous material where variation in composition is an
additional random factor;
● analysis of samples by several laboratories, methods or analysts where the
laboratories, methods or analysts are additional fixed-effect factors;
● analysis of a material stored under different conditions to investigate
stability where the storage conditions provide an additional fixed-effect
factor.
Section B – Assessment of data

B4 C ALIBRATION AND LINEAR

REGRESSION

Key Notes
Calibration Calibration is the process of establishing a relation between a detection or
measurement system and known amounts or concentrations of an analyte
under speciﬁed conditions.

Correlation The coefﬁcient is used to assess the degree of linearity between two
coefﬁcient variables, e.g. an instrument response and an analyte mass or
concentration.

Linear regression Calculations to deﬁne the best straight line through a series of calibration
points represented graphically are described as linear regression.

Limit of detection The smallest mass or concentration of an analyte that can be measured
quantitatively at a deﬁned level of probability deﬁnes a limit of detection.

Standard addition This is a calibration procedure that avoids matrix interference by

measuring instrument response for an analyte in both the sample and a
sample to which known amounts of an analyte standard have been
added.

Internal This is a calibration procedure where the ratio of the instrument response
standardization for an analyte to that of an added standard is measured for a series of
analyte standards and samples.

Internal Internally normalized results give the relative composition of a mixture

normalization by expressing the instrument response for each analyte as a fraction or
percentage of the sum of the responses for all of the analytes.

Calibration Many quantitative analytical procedures rely on instrumental measurements

where a property of the analyte(s) is monitored by a suitable detection system.
The detector generates an electrical signal, the magnitude of which is deter-
mined by the mass or concentration of the analyte. Before using a particular
analytical procedure to analyze samples, it is ﬁrst necessary to establish the
detector responses to known amounts of the analyte (calibration standards)
over a selected mass or concentration range, a process known as calibration.
The relation between the two variables is often linear (directly proportional),
but there is generally an upper limit to the range of values beyond which a
curved or curvilinear relation is observed. In some instances, there may be no
direct linear relation at all, or a logarithmic or more complex mathematical
correlation may be found.
42 Section B – Assessment of data

Calibration data are generally used to construct a calibration graph, where

detector response is plotted on the ordinate axis (y-values) and mass or concen-
tration of the analyte on the abscissa axis (x-values) as shown in Figure 1.
The graphs are often linear, being defined by the equation
y = bx + a (1)
where b is the slope and a the intercept on the y-axis. In some cases, it is prefer-
able to plot a logarithmic function of the detector response or analyte concentra-
tion to obtain a linear calibration curve.
Unknown levels of the analyte are determined from the graph by interpola-
tion. Where a linear relation has been established, a calibration factor can be
used to convert detector response to mass or concentration of analyte when
analyzing samples.
Theoretically, the graph should pass through the origin, but frequently in
practice there is a small positive intercept due to traces of analyte in the reagent
blank or contributions to the detector signal by other components in the
standards. Calibration points also show a degree of scatter due to the effects of
experimental errors in preparing the standards, or noise in the measuring
circuitry. A line of best fit through the points, known as a regression line, is
therefore drawn or computed.
Calibration graphs may show curvature, particularly at higher mass or
concentration levels, but this does not invalidate their use if the data are repro-
ducible. However, it is advisable to prepare additional standards to define the
curve more closely, and the use of a factor to compute analyte levels in samples
is precluded.
Statistical methods are used to assess calibration data
● for linearity or otherwise;
● to calculate the parameters defining a calibration curve;
● to assess the effects of determinate and indeterminate errors on standards
and samples.

25
Detector response

20
Sample response
15

5 Interpolated sample
mass/concentration
0
0 2 4 6 8 10 12
Analyte mass/concentration
Fig. 1. A typical calibration graph.

Correlation The correlation coefﬁcient, r, indicates the degree of linearity between x and y
coefﬁcient and is given by the expression
B4 – Calibration and linear regression 43

i=N _ _
冱 {(x − x)(y − y)}
i i
i=1
r = (2)
冦冤冥冤冥冧
i=N _ i=N _ ⁄ 1

冱 (xi − x)2 冱 (yi − y)2

i=1 i=1

_ _
where x1y1; x2y2; x3y3;….xn,yn are the co-ordinates of the plotted points, x and y
are the means of the x and y values respectively, and Â indicates sums of terms
(see standard deviation equations (1), (2) and (4), Topic B2).
The range of possible values for r is -1 £ r £ +1. A value of unity indicates a
perfect linear correlation between x and y, all the points lying exactly on a
straight line, whilst a value of zero indicates no linear correlation. Values may
be positive or negative depending on the slope of the calibration graph. These
alternatives are illustrated in Figure 2 (a) to (c).
Most calibration graphs have a positive slope, and correlation coefﬁcients
frequently exceed 0.99. They are normally quoted to four decimal places. (Note
that graphs with a slight curvature may still have correlation coefﬁcients
exceeding about 0.98 (Fig. 2(d)), hence great care must be taken before
concluding that the data shows a linear relation. Visual inspection of the plotted
points is the only way of avoiding mistakes.)

Linear When inspection of the calibration data and the value of the correlation coefﬁ-
regression cient show that there is a linear relation between the detector response and the

(a) (b)
10 10
Detector response

Detector response

8 8

6 r = +1 6 r = –1

4 4

2 2

0 0
0 2 4 6 8 10 0 2 4 6 8 10
Analyte mass/concentration Analyte mass/concentration

Detector response

8 8
r = 0.9850
6 r=0 6

4 4

2 2

0 0
0 2 4 6 8 10 0 2 4 6 8 10
Analyte mass/concentration Analyte mass/concentration
Fig. 2. Examples of correlation coefﬁcients. (a) Perfect positive correlation; (b) perfect negative correlation; (c) no
correlation, and (d) curved correlation.
44 Section B – Assessment of data

mass or concentration of the analyte, it is necessary to draw a line of best ﬁt

through the plotted points before they can be used as a working curve. Although
this can be done by eye, a more accurate method is to employ linear regression.
It is invariably the case that, due to the effects of indeterminate errors on the data,
most of the points do not lie exactly on the line, as shown in Figure 1. Linear
regression enables a line of best fit through the points to be defined by calculating
values for the slope and y-axis intercept (b and a respectively in equation (1)), and
the method of least squares is commonly used for this purpose. An assumption
is made that only errors in the detector responses (y-values) are significant, any
errors in the values for the mass or concentration of the analyte being neglected.
The deviations in the y-direction of the individual plotted points from the
calculated regression line are known as y-residuals (Fig. 3) and the line repre-
sents the regression of y upon x. The method of least squares minimizes the sum
of the squares of the y-residuals by equating them to zero in defining equations
for the slope and intercept of the regression line.
For the slope, b
i=N _ _
冱 {(x − x)(y − y )}
i i
b=
i=1
i=N _
(3)
冱 (x − x)
i
2

i=1

For the y-axis intercept, a

_ _
a = y − b.x (4)
N.B.
_ _ As equation (4) is a re-arrangement of equation (1), it follows that the point
x, y, known as the centroid, must lie on the regression line.

1.2

1
Absorbance at 325 nm

0.8 Slope = 0.00878

y-residuals
0.6

0.4

0.2

Intercept = 0.0686
0
0 20 40 60 80 100 120
Concentration (mg cm–3)
Fig. 3. Calibration graph, regression line, slope and intercept values for the UV spectrophoto-
metric determination
_ _ of an active ingredient in a sun cream. ——— = regression line; 䡩 =
centroid, x, y; ----------- = conﬁdence limits lines at the 99 percent level; 䊴——䊳 = conﬁdence
limits for sample concentration of 30 mg cm-3. Inset: Illustration of y-residuals.
B4 – Calibration and linear regression 45

Example
A calibration graph was prepared as part of a validation procedure for a new
method to determine an active constituent of a sun cream by UV spectrophoto-
metry. The following data were obtained:

Analyte conc. (mg cm-3) 0 20 40 60 80 100 120

UV absorbance at 325 nm 0.095 0.227 0.409 0.573 0.786 0.955 1.123

The data is ﬁrst checked for linearity by calculation of the correlation coefﬁcient,
r, and visual inspection of a plotted curve. Some calculators and computer soft-
ware can perform the computation from the raw data, but it is instructive to
show the full working, for which tabulation is preferable.
_ _ _ _ _ _
xi yi (xi − x ) (xi − x )2 (yi − y) (yi − y )2 (xi − x )(yi − y)
0 0.095 −60 3600 −0.5004 0.2504 30.024
20 0.227 −40 1600 −0.3684 0.1357 14.736
40 0.409 −20 400 −0.1864 0.0347 3.728
60 0.573 0 0 −0.0224 0.0005 0
80 0.786 20 400 0.1906 0.0363 3.812
100 0.955 40 1600 0.3596 0.1293 14.384
120 1.123 60 3600 0.5276 0.2784 31.656

S 420 4.168 0 11200 0 0.8653 98.340

_ _
x = 60 y = 0.59543

Substitution of the totals in columns 4, 6 and 7 in equation (2) gives

r = 98.340/(11200 ¥ 0.8653)1/2 = 98.340/98.445 = 0.9989
Figure 3 and the correlation coefficient of 0.9989 show that there is a good linear
relation between the measured UV absorbance and the analyte concentration.
The slope and y-axis intercept of the regression line, given by equations (3)
and (4) respectively are
b = 98.340/11200 = 0.00878 a = 0.59543 - (0.00878 ¥ 60) = 0.0686
The y-axis intercept, slope and analyte masses or concentrations calculated
by interpolation from the regression line are all affected by errors. Additional
equations can be used to obtain the following statistics:
● estimated standard deviations for the slope and intercept;
● estimated standard deviations for analyte masses or concentrations deter-
mined from the calibration graph;
● confidence limits for analyte masses and concentrations at selected proba-
bility levels;
● limit of detection of the analyte (vide infra).
Confidence limits (Topic B2) over the entire range of the calibration graph at
selected probability levels, e.g. 95 or 99 percent, can be displayed (dashed
curves, Fig. 3). A horizontal line drawn through a given experimental point on
the regression line and intersecting the confidence limits lines on either side
gives the upper and lower limits for that particular mass or concentration. _ _
Figure 3 shows the 99% limits, the narrowest interval being at the centroid, x, y,
of the graph, and widening steadily towards each end.
Some calculators and computer packages have the ability to perform the
regression calculations described. Where there is a nonlinear relation between
46 Section B – Assessment of data

the detector response and the mass or concentration of the analyte more
complex curvilinear or logarithmic regression calculations are required.

Limit of For any analytical procedure, it is important to establish the smallest amount of
detection an analyte that can be detected and/or measured quantitatively. In statistical
terms, and for instrumental data, this is defined as the smallest amount of an
analyte giving a detector response significantly different from a blank or back-
ground response (i.e. the response from standards containing the same reagents
and having the same overall composition (matrix) as the samples, where this is
known, but containing no analyte). Detection limits are usually based on esti-
mates of the standard deviation of replicate measurements of prepared blanks.
A detection limit of two or_ three times the estimated standard deviation of the
blanks above their mean, xB, is often quoted, where as many blanks as possible
(at least 5 to 10) have been prepared and measured.
This is somewhat arbitrary, and it is perfectly acceptable to define alternatives
provided that the basis is clear and comparisons are made at the same proba-
bility level.

Standard Where components of a sample other than the analyte(s) (the matrix) interfere
addition with the instrument response for the analyte, the use of a calibration curve
based on standards of pure analyte may lead to erroneous results. Such matrix
interference effects can be largely if not entirely avoided by preparing calibra-
tion standards where known amounts of pure analyte are added to a series of
equal sized portions of the sample, a procedure known as spiking. In addition,
one portion of sample is not spiked with analyte. (Note: if spiking sample solu-
tions with analyte changes the volume significantly, volume corrections must be
applied.)
The effects of the matrix on measurements of the analyte in both the spiked
and unspiked samples should be identical. The instrument responses are then
used to construct a calibration graph where the x-axis values are the added
amounts of analyte and the response for the unspiked sample is at x = 0 (i.e., the
curve does NOT pass through the origin). The regression line is calculated and
extrapolated back to give a negative intercept on the x-axis at y = 0, which corre-
sponds to the amount of analyte in the sample (Fig. 4).
The less reliable procedure of extrapolation rather than interpolation is
outweighed by the advantage of eliminating or minimizing matrix interference.
The method of standard addition is widely used, particularly when the
composition of the sample matrix is variable or unknown so that the response of
a reagent/matrix blank would be unreliable. At least three and preferably more
spiked samples should be prepared, but if the amount of sample is limited, as
few as one sample must suffice. It is especially useful with such analytical tech-
niques as flame and plasma emission spectrometry and potentiometry (Topics
E4, E5 and C8).

Example
The calcium level in a clinical sample was determined by ﬂame emission
spectrometry using a standard addition method, which gave the following data:
Spiked calcium (ppm) 0 10 20 30 40 50
Emission intensity 0.257 0.314 0.364 0.413 0.468 0.528
at 423 nm
B4 – Calibration and linear regression 47

0.7

0.6 Slope = 0.005349

Emission intensity at 423 nm

0.5

0.4
Spiked samples

0.3
Unspiked
sample
0.2
Sample concentration
48 ppm
0.1
Intercept = 0.2569

0
–60 –50 –40 –30 –20 –10 0 10 20 30 40 50 60
Calcium (ppm)

Fig. 4. Standard addition calibration graph, regression line, slope and intercept values for the
ﬂame emission determination of calcium in a clinical sample.

Detailed calculations of the correlation coefﬁcient, r, and the slope and intercept
values have not been given, but should be set out as in the previous example if a
suitable calculator or computer program is not available.
The amount of calcium in the sample can be read from the extrapolated graph
or calculated from the slope, b, and the intercept, a
Calcium concentration in sample = a/b = 0.2569/0.005349 = 48 ppm

Internal For some analytical techniques, particularly chromatography, variations in

standardization experimental conditions can adversely affect the precision of the data. A
calibration procedure whereby a constant amount of a selected substance,
the internal standard is added to all samples and analyte standards alike
compensates for variations in sample size and other parameters. The ratio of
the detector response for the analyte in each standard to the corresponding
response for the added internal standard is plotted on the y-axis of a calibration
graph against the mass or concentration of the analyte on the x-axis. The corre-
lation coefﬁcient, slope and intercept values can be computed as shown previ-
ously, and response ratios for the analyte and added internal standard in the
samples can then be used to determine the amount of analyte in the samples by
interpolation on the graph. If only one or two analyte standards are prepared,
the amount of analyte in a sample can be calculated by simple proportion, i.e.
analyte in sample response ratio for sample
=
analyte in standard response ratio for standard

Internal For some purposes, only the relative amounts of the analytes in a multicomponent
normalization mixture are required. These are normalized to 100 or 1 by expressing each as a
48 Section B – Assessment of data

percentage or fraction of the total. Internal normalization is of particular value in

quantitative chromatography where several components of a sample can be
determined simultaneously, and absolute levels are not of interest. The relative
composition is calculated from the instrument response, peak area in the case of
a chromatographic analysis, for each component in the mixture using the formula
Ax
%xi =
i=n × 100
冱 i
i=1
A

where xi is one of n components and A is the measured area or response.

Example
Figure 5 is a chromatographic record (chromatogram) of the separation of a 5-
component mixture. The measured peak areas (using electronic integration with
a computing-integrator, computer and chromatography data processing software
or geometric construction such as triangulation, 1⁄2 ¥ base ¥ height) and percent-
ages by internal normalization, which must total 100 percent, are given in Table 1
(e.g., for component 1, relative percent = (167.8/466.94) ¥ 100 = 35.9 percent).

Table 1. Peak areas and percentage composition by internal normalization for a

5-component mixture
Component Measured peak area Relative percent
(arbitrary units)
1 167.8 35.9
2 31.63 6.8
3 108.5 23.2
4 80.63 17.3
5 78.38 16.8
Totals 466.94 100.0

1
Inject

4
5
2

0 3 6 9 12 15 18
Time (min)

Fig. 5. Chromatogram of a 5-component mixture.

Section B – Assessment of data

B5 Q UALITY CONTROL AND

CHEMOMETRICS

Key Notes
Control charts Graphical representations of quantitative data from an ongoing series of
measurements can be used to monitor the stability of a system for quality
control (QC) purposes.

Collaborative testing Schemes have been devised to compare results for the analysis of
homogeneous samples or standard materials from groups of analytical
laboratories to test speciﬁed methods of analysis.

Multivariate Advanced chemometric procedures can be applied to the statistical

statistics analysis of large amounts of data and numerous experimental variables
so as to extract the maximum amount of useful information.

Fig. 1. A Shewart averages control chart.

multiple of the estimated standard deviation of the method, usually 95% or

±1.96s/N1/2;
● action lines are positioned to correspond to a selected probability level
or multiple of the estimated standard deviation of the method usually 99.7% or
±3.01s/N1/2;
● the pairs of warning lines will move inwards towards the target line as the
number of results per plotted point, N, increases, so the value of N must be
ﬁxed before the chart is used;
● in the absence of determinate errors (bias), 95% of values should fall within
the upper and lower warning lines. The system is then considered to be
stable or under control;
● two or more consecutive values falling outside the warning lines but within
the action lines indicate a possible loss of control;
● two or more consecutive values falling outside the action lines indicate the
occurrence of one or more determinate errors; the system is then considered
to be out of control and remedial action should be taken;
● trends in plotted values are an indication of incipient problems, as are ten or
more consecutive values on one side of the target line.
The chart shows that the system is in control during the ﬁrst 6 samples
analyzed, but the upper warning level has been breached by result 7. However,
the next 8 results are within the warning levels, but results 17 to 20 indicate a
downward trend culminating with both lower limits being breached indicating
a loss of control due to one or more determinate errors. At this point, the causes
should be sought and remedial action taken.

Collaborative The principal purpose of collaborative testing is to assess the accuracy of

testing results from a group of laboratories performing one or more speciﬁc quantita-
tive determinations. Interlaboratory studies are also used to assess other perfor-
mance characteristics of a particular method or the competence of individual
analysts in relation to a speciﬁed technique, analyte or group of analytes. They
are frequently used by trade associations, standards organizations or agencies
B5 – Quality control and chemometrics 51

with stringent analytical requirements to develop, test and validate analytical

methods.
Proficiency testing schemes are designed to test the competence of individual
laboratories to perform specific analyses or a range of determinations of
analytes in various matrixes and possibly using alternative techniques.
Some typical examples of collaborative studies and proficiency testing are
● alcohol in beverages;
● metals or volatile organics in soils;
● hazardous airborne substances;
● trace metals or pesticides in domestic water supplies;
● drugs of abuse by chromatographic methods;
● adulterants in foodstuffs;
● additives in polymers and composites.
In statistical terms, collaborative trials are designed to reveal the presence of a
bias in the results arising from determinate errors that may occur in addition to
inherent indeterminate (random) errors. Statistics that are used in these schemes
include estimated standard deviation and coefficient of variation, confidence
limits (Topic B2) and tests of significance including ANOVA (Topic B3). Of
particular value is a z-score to indicate the accuracy of results from the partici-
pating laboratories. The z-value is a statistical factor based on a Gaussian or
normal distribution, which is included in the equation for confidence limits of
an experimental value in relation to a true or accepted value (Topic B2, equation
(5)). Rearrangement of this equation, for a single determination, xi, (N=1) gives

冨
xi − m
z=
s 冨
where m is the true or accepted mass or concentration of the analyte and s is a
value for the standard deviation for the method selected by the organizers of the
study (strictly an estimated value, s). A typical chart of z-scores for a collabora-
tive study involving 22 laboratories is shown in Figure 2.
Results that have z-scores of between +1.96 and -1.96 are considered to have
acceptable accuracy, as these values correspond to conﬁdence limits at the 95%
probability level (Topic B2, Fig. 3). In the example, only laboratories 12 and 18

1
z-score

–1

–2

–3
19 13 21 22 20 10 5 15 16 1 11 8 6 14 2 17 7 4 9 3 12 18
Laboratory number

Fig. 2. z-score chart for results from 22 laboratories.

52 Section B – Assessment of data

have results with unacceptable accuracy, whilst laboratories 5 and 15 have

results closest to the true or accepted value. In practice, the situation is often
worse than this, with some laboratories incurring worryingly large determinate
errors that need to be identified and rectified.
Although accuracy is of prime importance in collaborative studies and pro-
ficiency testing, precision should also be monitored. The level to be expected
varies considerably with the concentration of an analyte and type of sample. A
useful guide for different levels of an analyte as measured by the coefficient of
variation (Topic B2) is exemplified in Figure 3.

50
Coefficient of variation (%)

0
10% 1% 0.1% 0.01% 10 ppm 1 ppm 0.1 ppm 0.01 ppm 1 ppb
Analyte concentration (log scale)

Fig. 3. Guide to typical coefﬁcients of variation as a function of analyte concentration.

Note that values of less than 1% should be attainable for major components of
a sample, whilst at ppb (parts per billion) levels, over 50% is acceptable.

Multivariate Computerized and automated analytical instrumentation facilitates the collec-

statistics tion of large amounts of data and the simultaneous monitoring of numerous
experimental parameters. To maximize the useful information that can be
extracted, sophisticated multivariate chemometric techniques are employed.
The mathematical computations involve matrix algebra and vectors, and rely
on the availability of specialized computer software. Typical applications
include the characterization of substances from proﬁles based on spectral,
chromatographic and other data, and quantitative analysis based on multiple
simultaneous measurements. Two important applications of multivariate
statistics are pattern recognition and multivariate modeling.

Pattern recognition
Sets of measurements characterizing a sample, e.g. the position of prominent
infrared absorption bands (Topic H4), signiﬁcant mass spectral fragments,
levels of particular analytes, and selected physical properties, are described as
patterns. These can be used to classify substances or to identify unknowns by
pattern matching. Figure 4 shows the distribution of trace levels of copper and
manganese in twelve geological samples where three clusters are evident. Just
B5 – Quality control and chemometrics 53

Manganese (ppm) 30

0
0 5 10 15
Copper (ppm)
Fig. 4. Copper and manganese distribution in geological samples showing three clusters with
differing proportions of each metal.

as two parameters can be plotted as a single point with speciﬁed x and y

co-ordinates on a two-dimensional graph, the values of n parameters can be
represented by a point in n-dimensional space. Although n-co-ordinate
graphs cannot be visualized, they can be studied through appropriate
computer processing and manipulation. Where a number of substances have
similar sets of n co-ordinates, and therefore similar characteristics, they
produce closely-spaced groups of points described as clusters, the interpreta-
tion of this data being described as cluster analysis. Mathematical procedures
to detect clusters include principal component analysis (PCA) and factor
analysis (FA), which seek to simplify the data by projection from n dimen-
sions onto a line, plane or 3-D graph to reduce the number of dimensions
without losing information.
Cluster analysis can be used in many ways, e.g. to monitor clinical specimens
from hospital patients where, for example, the levels of pH, glucose, potassium,
calcium, phosphate and specific enzymes vary according to the absence, pres-
ence or severity of a particular disease. It can also be applied to the characteriza-
tion of glass fragments for forensic purposes through profiling and comparisons
of their trace metal contents, for identifying the source of a crude oil spillage on
the basis of the proportions of minor organic compounds and metals, and in the
classification of organic compounds with similar structural features to facilitate
the elucidation of unknown structures.

Multivariate modeling
Quantitative analysis for one or more analytes through the simultaneous
measurement of experimental parameters such as molecular UV or infrared
absorbance at multiple wavelengths can be achieved even where clearly deﬁned
spectral bands are not discernible. Standards of known composition are used to
compute and reﬁne quantitative calibration data assuming linear or nonlinear
models. Principal component regression (PCR) and partial least squares (PLS)
regression are two multivariate regression techniques developed from linear
regression (Topic B4) to optimize the data.
54 Section B – Assessment of data

Multivariate modeling is the basis of quantitative analysis by near infrared

spectrometry (Topic E11) to determine moisture or fats in cereals, meat and
other foodstuffs. It is particularly applicable to environmental analysis where
complex and inter-related variables affect the distributions of a wide range of
organic compounds and elements occurring naturally or as pollutants.
Section C – Analytical reactions in solution

C1 S OLUTION EQUILIBRIA

Key Notes
Solvents The major component of a solution is referred to as the solvent, and there
is a wide range of inorganic and organic solvents used in analytical
chemistry. Their properties determine their use.

Solubility When a substance called the solute is dissolved in a solvent to form a

solution, its behavior is often altered. Reactions in solution are faster than
in the solid state. The amount of substance that can dissolve in a given
amount of solvent at a stated temperature and pressure is called the
solubility and is determined by the nature of the materials and the laws
governing the solubility equilibrium.

Ions in solution Some substances form ions, which are species possessing a charge. These
behave in a distinct way in solution. They may attract molecules of
solvent, may associate together, and may react with other species to form
complexes or a precipitate.

The pX notation Since concentrations vary over a very wide range, they are often
represented by the logarithmic pX notation where pX = - log(X), where X
is the concentration or activity of an ion, or an equilibrium constant.

Equilibria in The laws of thermodynamics govern the behavior of all species in

solution solution. Every reaction depends upon the thermodynamic properties of
the species involved. Where those properties are changed by the solvent
by association, by reaction or temperature, the behavior will alter.
Physical and chemical equilibria in solution are most important.

Table 1. Properties of some solvents

Solvent Boiling point (∞C) Density, Dielectric
(g cm-3) constant, er
Water 100 1.00 78.6
Ammonia -34 0.68 22.0
Ethanol 78 0.79 24.3
n-hexane 69 0.66 1.88
Diethyl ether 34 0.71 4.33
Note: density at 25∞C or at BP; dielectric constant = relative permittivity

to the solvent. The solvent structure is also disrupted by the presence of species
different in size, shape and polarity from the solvent molecules.
Ideally, the behavior should depend on the concentration m (in molarity,
mole fraction or other units), but often this must be modiﬁed and the activity, a,
used:
a = m g = p/pn
where g is called the activity coefﬁcient. The vapor pressure of the solution is p,
and that in the standard state is pn. Activities are dimensionless.
Solvents, such as water, with high dielectric constants (or relative permittivi-
ties) reduce the force F between ions of charges z1e and z2e a distance r apart:
F = z1z2 e2/eoer r2
where eo is the permittivity of free space. Also, they will solvate ions more
strongly and thus assist ion formation and separation.
Hexane, diethyl ether and tetrachloromethane (CCl4) all have low dielectric
constants and are nonpolar. They are very poor at ionizing solutes. However,
they are very good solvents for nonpolar substances.

Solubility The equilibrium amount of solute which will dissolve in a given amount of
solvent at a given temperature and pressure is called the solubility. The solu-
bility may be quoted in any units of concentration, for example, mol m-3,
molarity, mole fraction, mass per unit volume or parts per million (ppm).
There is a general ‘rule of thumb’ that ‘like dissolves like’. For example, a
nonpolar hydrocarbon solvent such as hexane would be a very good solvent for
solid hydrocarbons such as dodecane or naphthalene. An ester would be a good
solvent for esters, and water or other polar solvents are appropriate for polar
and ionic compounds.
● Gases dissolve in solvents according to Henry’s Law, provided they do not
react with the solvent:
p B = xB K
where xB is the mole fraction of solute gas B which dissolves at a partial pres-
sure pB of B, and K is a constant at a given temperature. This is analytically
important for several reasons. For example, nitrogen is bubbled through
solutions to decrease the partial pressure of oxygen in electrochemical experi-
ments. Similarly, air is removed from liquid chromatography solvents by
C1 – Solution equilibria 57

passing helium through them, or by boiling them, since gas solubility

decreases as the temperature is increased.
● Liquids. When different liquids are mixed, many types of behavior may
occur. If the molecules in the liquids are of similar size, shape, polarity and
chemical nature they may mix in all proportions. For example, benzene and
methylbenzene (toluene) mix completely. In such ideal solutions, obeying
Raoult’s law, the activity coefﬁcient is close to 1:

a = p/pn = x

If the component molecules differ greatly in polarity, size or chemical nature

(e.g., water and tetrachloromethane) they may not mix at all. This is an
important condition for solvent extraction (Topic D1). The distribution of a
solute between a pair of immiscible liquids depends primarily on the
solubility of the solute in each liquid.
● Solids generally follow the ‘like dissolves like’ rule. Nonpolar, covalent
materials dissolve best in nonpolar solvents. Solid triglycerides such as
tristearin are extracted by diethyl ether, but are nearly insoluble in water.
Salts, such as sodium chloride are highly soluble in water, but virtually
insoluble in ether.

Ions in solution The behavior of ions in solution may be summarized as follows.

(i) Solids whose structure consists of ions held together by electrostatic forces
(e.g. NaCl) must be separated into discrete ions when they dissolve. These
ions often gain stability by solvation with molecules of the solvent. Such
solutions are described as strong electrolytes.
(ii) Some covalent molecules, such as ethanoic acid, may form ions by the
unequal breaking of a covalent bond, followed by stabilization by solvation.
This occurs only partially and these are called weak electrolytes.

H : OCOCH3 [ H+ + -OCOCH3

(iii) In some cases ions do not separate completely. In concentrated solutions,

oppositely charged ions may exist as ion-pairs, large ions of surfactants
may aggregate into micelles, which are used in capillary electrophoresis
(Topics D8 and D9), and the dissociation of covalent molecules may be only
partial.
(iv) At ‘inﬁnite dilution’, that is, as the concentration approaches zero, ions are
truly separate. However, even at quite low concentrations, the attractions
between ions of opposite charge will cause each ion to become surrounded
by an irregular cloud, or ionic atmosphere. As the solution becomes even
more concentrated, the ionic atmosphere becomes more compact around
each ion and alters its behavior greatly.
In very dilute solutions, the effects of the ionic atmosphere may be approxi-
mated by using the Debye-Hückel theory, which predicts that the mean ionic
activity coefﬁcient, g± is, for an electrolyte with positive ions with charge z+,
negative ions with charge z-, is given by:

log(g±) = -A (z+ . z-) √|(I)

where I is the ionic strength = 1⁄2 S (ci zi2) for all the ions in the solution.
For more concentrated ionic solutions, above 0.1 M, no general theory exists,
58 Section C – Analytical reactions in solution

but additional terms are often added to the Debye-Hückel equation to compen-
sate for the change in the activity.

The pX notation The concentration of species in solution may range from very small to large. For
example in a saturated aqueous solution of silver chloride, the concentration of
silver ions is about 10-5 M, while for concentrated hydrochloric acid the
concentration of hydrogen and chloride ions is about 10 M. For convenience,
a logarithmic scale is often used:
pX = -log (X)
where X is the concentration of the species, or a related quantity. Thus, for the
examples above, pAg = 5 in saturated aqueous silver chloride and pH = -1 in
concentrated HCl.
Since equilibrium constants are derived from activities or concentrations as
noted below, this notation is also used for them:
pK = -log (K)

Equilibria in Most reactions will eventually reach equilibrium. That is, the concentrations of
solution reactants and products change no further, since the rates of the forward and
reverse reactions are the same.
From the above arguments concerning solutions, and from the laws of
thermodynamics, any equilibrium in solution involving species D, F, U and V:
D+F[U+V
will have an equilibrium constant , KT, at a particular temperature T given by:
KT = (aU . aV)/(aD . aF)
where the activities are the values at equilibrium. It should be noted that KT
changes with temperature. The larger the equilibrium constant, the greater will
be the ratio of products to reactants at equilibrium.

There are many types of equilibria that occur in solution, but for the impor-
tant analytical conditions of ionic equilibria in aqueous solution, four examples
are very important.
(i) Acid and base dissociation. In aqueous solution, strong electrolytes (e.g.,
NaCl, HNO3, NaOH) exist in their ionic forms all the time. However, weak
electrolytes exhibit dissociation equilibria. For ethanoic acid, for example:
HOOCCH3 + H2O [ H3O+ + CH3COO-
Ka = (aH . aA)/(aHA . aW) = 1.75 ¥ 10-5
where HA, W, H and A represent each of the species in the above
equilibrium. In dilute solutions the activity of the water aW is close to 1.
For ammonia:
NH3 + H2O [ NH4+ + OH-
Kb = (aNH4+ . aOH-)/(aNH3 . aW) = 1.76 ¥ 10-5
Water behaves in a similar way:
2 H2O [ H3O+ + OH-
KW = (aH3O+ . aOH-) = 10-14
C1 – Solution equilibria 59

(ii) Complexation equilibria. The reaction between an acceptor metal ion M

and a ligand L to form a complex ML is characterized by an equilibrium
constant. This is discussed further in Topic C7, but a simple example will
sufﬁce here:
M(aq) + L(aq) [ ML(aq)
Kf = (aML)/(aM . aL)
For example, for the copper-EDTA complex at 25oC: Kf = 6.3 ¥ 1018
(iii) Solubility equilibria. If a compound is practically insoluble in water, this
is useful analytically because it provides a means of separating this
compound from others that are soluble. The technique of gravimetric
analysis has been developed to give very accurate analyses of materials by
weighing pure precipitates of insoluble compounds to give quantitative
measurements of their concentration. For the quantitative determination of
sulfate ions, SO42-, the solution may be treated with a solution of a soluble
barium salt such as barium chloride BaCl2, when the following reaction
occurs:
Ba2+ + SO42- [ BaSO4 (s)
Conversely, if solid barium sulfate is put into water:
BaSO4 (s) = Ba2+ + SO42-
The solubility product, Ksp, is an equilibrium constant for this reaction
Ksp = a(Ba2+) . a(SO42-) = 1.2 ¥ 10-10
bearing in mind that the pure, solid BaSO4 has a = 1. This means that a
solution of barium sulfate in pure water has a concentration of sulfate ions
of only 1.1 ¥ 10-5 M. The concentration of the barium ions is the same.
(iv) Redox equilibria. When a species gains electrons during a reaction, it
undergoes reduction and, conversely, when a species loses electrons it
undergoes oxidation. In the total reaction, these processes occur simultane-
ously, for example:
Ce4+ + Fe2+ = Ce3+ + Fe3+
The cerium is reduced from oxidation state 4 to 3, while the iron is oxidized
from 2 to 3. Any general ‘redox process’ may be written:
Ox1 + Red2 = Red1 + Ox2
The equilibrium constant of redox reactions is generally expressed in terms
of the appropriate electrode potentials (Topics C5, C8), but for the above
reaction:
K = (a(Ce3+) . a(Fe3+))/(a(Ce4+) . a(Fe2+)) = 2.2 ¥ 1012

Summary
For ionic equilibria in solution, which are widely used in analytical chemistry,
a large equilibrium constant for the reaction indicates that it will proceed
practically to completion. If the equilibrium constant is of the order of 1010,
then the ratio of products to reactants will be much greater than 1000 to 1. For
example:
60 Section C – Analytical reactions in solution

H+ + OH- = H2O K = 1014

C(aq) + A(aq) = CA(solid) K = 1010
M(aq) + L(aq) = ML(complex) K = 1010
Ox1 + Red2 = Red1 + Ox2 K = 1012
Therefore, these reactions may be used for quantitative measurements, for
example by volumetric or gravimetric techniques (Topics C5, C7 and C8).
It should be noted that, in calculations involving solution equilibria, certain
rules should always be considered.
● Electroneutrality. The concentrations of positive and negative charges must
be equal. Sometimes, ions that do not react are omitted from the equations,
although they must be present in the solution.
● Stoichiometry. The total amounts of all species containing an element must
be constant, since no element can be created or destroyed.
● Equilibria. All possible equilibria, including those involving the solvent,
must be taken into account.
Section C – Analytical reactions in solution

C2 E LECTROCHEMICAL
REACTIONS

Key Notes
Electrochemical Reactions involving charged species in solution are important in many
reactions analyses. Their chemistry and the laws governing them must be known.

Electrochemical cells The construction of an electrochemical cell using two electrodes allows
the study of the cell reaction, changes in concentration, and electrolysis.

Electrode potentials The potential difference between electrodes depends on the properties of
the electrodes and the concentrations in the solution.

Electrolysis If the cell reaction is driven by applying an external voltage supply, then
useful chemical reactions may occur.

Electrochemical In order to study an electrochemical reaction, the appropriate cell must be

cells constructed. It is impossible to measure the electrical properties using a single
contact. For example, connecting to one end of a resistor will not allow measure-
ment of its resistance. Connections must be made to two electrodes, and a cell must
be constructed. Electrical connection to the solution, whether to measure a cell emf
or to conduct electrolysis reactions, must be made through two electrodes.
Cells with two similar, inert electrodes placed in the same solution may be
used for measuring conductance, discussed in Topic C10. Cells where electro-
lytic reactions occur are used in voltammetry, which is discussed in Topic C9.
For potentiometric methods, discussed here and in other parts of this section,
two dissimilar electrodes are used to construct a cell whose emf depends on the
solutions and electrodes used.
Many different types of electrodes are available and the most important are
described in Topic C3. The simplest are made of a metal, such as zinc, copper, or
62 Section C – Analytical reactions in solution

platinum. Consider a cell set up with a zinc electrode dipping into a zinc sulfate
solution, and connected through a porous disc to a copper sulfate solution in
contact with a copper electrode, as shown in Figure 1.
The cell is conventionally written as:
ZnÔZn2+ SO42-ÔCu2+ SO42-ÔCu
The cell reaction will result in simultaneous reduction or addition of electrons to
a species in solution at the one electrode and oxidation, or removal of electrons
at the other.
Cu2+ + 2e- = Cu reduction (note: ‘reduction on the right’)
- 2+
Zn - 2e = Zn oxidation (note: ‘oxidation on the left’)
Adding these gives the cell reaction (excluding the sulfate anions).
Cu2+ + Zn = Cu + Zn2+
If the cell reaction is allowed to take place, by connecting the copper and zinc
electrodes electrically, then it acts as a galvanic cell or battery, and does work. If
the potential difference is measured with no current ﬂow, it is deﬁned as the
electromotive force, or emf, E, and this is equivalent to reversible conditions. If
an external voltage is applied to the cell it can cause electrolysis, driving the cell
reaction.
In order to prevent the zinc sulfate and the copper sulfate mixing, a porous
disc, or a conducting salt bridge of potassium sulfate in a tube, or a wick can be
used to connect the solutions. This helps to reduce the liquid junction
potential, which occurs when two solutions of unequal concentration, or
containing dissimilar ions are placed in contact. Because of the different rates of
diffusion of the ions, a liquid junction (or diffusion) potential, EL , is set up and
this affects the total cell emf. For example, between a solution of 0.1 M HCl and
a solution of 0.01 M HCl, EL is about 38 mV.
By using a third solution, which must not react, and in which the cation
and anion have similar mobilities and diffusion characteristics, to connect the
two electrode solutions the liquid junction potential may be minimized.
Concentrated potassium chloride, about 4 M, is often used, since K+ and Cl- ions
have almost the same mobility (see Topic C10). Potassium nitrate or ammonium
nitrate is also suitable if chloride ions react. Use of a salt bridge reduces the
effective EL to about 1 mV.
The laws of thermodynamics, which are described in detail in textbooks of
physical chemistry, provide the connection between the chemical reaction and
the electrochemical quantities that are measured in analysis.

Meter

Zinc electrode Copper electrode

Zinc sulfate solution Copper sulfate solution

Porous plate

Fig. 1. Daniell cell.

C2 – Electrochemical reactions 63

The free energy G depends on the activity (or approximately the concentra-
tion) of the reactants and products. For a general reaction:
Ox1 + Red2 = Red1 + Ox2
n
DG = DG + RT ln [(a(Ox2).a(Red1))/(a(Ox1).a(Red2))]
where the superscript, n, means that it applies to the standard state. For gases,
this is 1 atmosphere pressure; for liquids and solids it is less than 1 atmosphere
pressure for the pure material; and for solutions, it is an activity of 1.
The electromotive force measures the free energy change under reversible
conditions when n Faradays of electrical charge occur in the cell reaction and is
related to it by the equation:
DG = -nFE
For example, in the Daniell cell reaction, 2 moles of electrons are transferred,
and the cell emf is about 1.1 V. Therefore:
DG = -2 ¥ 96485 ¥ 1.1 = -212.3 kJ mol-1
Combining the above equations, for the Daniell cell:
E = En + (RT/2F) ln [(a(Zn2+).a(Cu))/(a(Zn).a(Cu2+))]
This is often referred to as the Nernst equation for the cell. (Note: the value of
En measures the cell emf and relates to the free energy change under standard
state conditions, while the value of E relates to the free energy change under
other, speciﬁed conditions. It is important to recognize that these equations
apply at any temperature, not just a standard temperature.)

Electrode In order to establish any scale it is necessary to have a reference point. Since
potentials measurement of emf is made using a cell, one electrode must be taken as a
reference electrode. The standard hydrogen electrode (SHE) has been chosen as
the primary reference.
A hydrogen electrode may be constructed using a platinum metal plate as
contact, and bubbling hydrogen gas through so that it makes contact with both
solution and platinum. When the gas is at 1 atmosphere pressure and the solu-
tion has an activity of hydrogen ions of 1, this is a standard hydrogen electrode
(SHE).
By convention, the standard hydrogen electrode is assigned an electrode
potential of 0.000 V and all other electrode potentials are compared to it.
A cell can be constructed with a copper electrode in copper sulfate and a
standard hydrogen electrode, as shown in Figure 2.

H2 supply Copper electrode

Pt electrode

Sintered disk
Solution with Copper sulfate solution
H+ (a=1)

Fig. 2. Copper-hydrogen electrochemical cell.

64 Section C – Analytical reactions in solution

The cell reaction is the reduction of copper ions by hydrogen:

Cu2+ + H2 = Cu + 2 H+
Writing the Nernst equation for this cell:
E = En - (RT/nF) ln [(a (Cu).a (H+)2)/(a (Cu2+).a (H2))]
The superscript, n, means the standard value, when the activities are all 1. Note
that the quantity (RT/F) has the value 0.02569 V at 25∞C. Here, n = 2, since two
electrons are transferred at each electrode. For a standard hydrogen electrode,
a (H+) = 1 and a (H2) = p(H2) = 1, so, we may write:
E = En - (RT/2F) ln [a (Cu)/(a (Cu2+)]
which is the equation for the copper electrode potential. It may also be written:
E = En + (RT/2F) ln [a (Cu2+)/(a (Cu)]
and for a pure copper electrode, a(Cu) = 1, so at 25∞C:
E = En + 0.01284 ln [a (Cu2+)]
This means that the electrode potential of a copper electrode depends logarith-
mically on the activity of the copper ions in solution. This is known as the
Nernst equation. Standard electrode potentials are listed in Table 1 of Topic C3.
Although the SHE is essential to deﬁne electrode potentials, it is not suitable
for many routine analytical uses, and the saturated calomel electrode (SCE) is
often used in practical cells. This is described in Topic C3.
The Nernst equation is very useful in analytical measurements, since it allows
the analyst to measure variations in concentration during reactions or experi-
ments. It is important to note that, strictly, electrochemical cells measure the
activity, although in many cases this may be related to the concentration.

Electrolysis If an electric current is passed through a cell consisting of two electrodes in

contact with an electrolyte solution, the current causes the ions to migrate
towards the electrode of opposite charge, where a reaction takes place. For
example, when acidiﬁed water is electrolyzed:
(i) the hydrogen ions are attracted to the negative electrode (cathode) and gain
electrons to form hydrogen gas:
2H+ + 2e- = H2
(ii) the hydroxyl ions are attracted to the positive electrode (anode) and lose
electrons to become water and oxygen gas:
2OH- - 2e- = H2O + 1⁄2 O2
The total reaction is therefore:
2H2O = 2H+ + 2OH- = H2O + H2 + 1⁄2 O2
In order to cause 1 mole of reaction involving the transfer of n electrons, then n
moles of electronic charge must be transferred or nF Coulombs cause 1 mole of
such a reaction, where F is the Faraday equal to 96458 C.
Electrolysis is the basis of the analytical techniques of polarography, voltam-
metry, amperometry and coulometry (see Topic C9). Electrolysis may also be
used for the deposition, production and puriﬁcation of materials. For example,
C2 – Electrochemical reactions 65

the copper electrolysis cell has two electrodes placed in an acidiﬁed aqueous
solution of a copper salt.

Example
If a current of 0.123 amps is passed through a copper electrolysis cell for one
hour, how much copper will deposit?
0.123 A = 0.123 C s-1
Since 1 hour = 3600 s, therefore, It = 442.8 C are passed.
1 mole of copper is 63.5 g, which requires 2F = 2 ¥ 96485 C, so 442.8 C deposit
63.5 ¥ 442.8/(2 ¥ 96485) g = 0.1457 g.
It is possible to deposit metals onto an electrode quantitatively, using the
technique of electrogravimetry, although this is not often used. Electrolysis may
also be used to generate reagents to react with analytes, and this is referred to as
coulometry.
Section C – Analytical reactions in solution

C3 P OTENTIOMETRY

Key Notes

Cells In order to make measurements of electrode potentials, or to study the

changes that take place in a solution reaction, an appropriate
electrochemical cell must be set up.

Indicator electrodes The indicator electrode makes electrical contact with the solution and acts
as a sensor, which responds to the activity of particular ions in solution
and acquires a potential dependent on the concentration of those ions.

Selectivity The ideal electrode should respond to a single ion, but this is not often
the case. The effectiveness of any indicator electrode is determined by its
selectivity.

Direct potentiometry The direct measurement of concentrations is possible using electrodes of

high selectivity and reproducibility. The measurement may also be used
to follow titrations.

Salt bridge Test solution

Fig. 1. Experimental arrangement with salt bridge junction and DVM.

Table 1. Potentiometric indicator electrodes

Class Description Example
Class 1 Metal/metal ion Ag/Ag+ (cation reversible)
Class 2 Metal/saturated metal salt/anion Ag/AgCl/Cl- (anion reversible
Redox Inert metal/redox couple Pt/Ce4+, Ce3+
Pt/H+, H2
Membrane Inner electrode/solution/ Glass electrode
ion selective membrane Fluoride electrode
ISFET Coated ﬁeld-effect transistor pH-sensitive
Gas-sensing electrodes pH-electrode + membrane For CO2, NH3

E(cell) = E(right) - E(left) = E(Cu2+, Cu) - E(Zn2+, Zn)

Dividing the emf equation into separate electrode equations:
E(Cu2+, Cu) = En(Cu2+, Cu) + (RT/2F) ln (a (Cu2+)/(a (Cu))
E( Zn2+, Zn) = En(Zn2+, Zn) + (RT/2F) ln (a (Zn2+)/(a (Zn))
If these equations are obeyed, then the electrodes show Nernstian response.

Class 1 electrodes
These are the simplest electrodes, but not necessarily the easiest to use. A metal
rod is immersed in a solution of ions of the same metal, for example silver with
a solution containing silver ions. With some ions it is important to prevent
hydrolysis or complexation taking place.
Ag+ (solution) + e- = Ag(solid)
E(Ag+/Ag) = En(Ag+, Ag) + RT/F ln (a (Ag+)/(a (Ag))
if a pure silver rod is used, a(Ag) = 1, so we may write:
E(Ag+/Ag) = En(Ag+, Ag) + RT/F ln (a (Ag+))
This is therefore an electrode reversible to silver ions. An example of the use
of this electrode is given later. En(Ag+, Ag) = 0.800 V at 25∞C is the standard
electrode potential of the silver electrode.

Class 2 electrodes
When an insoluble salt of a metal (see also Topic C8) is present, the concentra-
tion of the metal ion depends on the concentration of the anion and on the
68 Section C – Analytical reactions in solution

solubility product Ksp. Therefore, if we put insoluble silver chloride, AgCl, in

contact with a silver rod,
a (Ag+) = Ksp/(a (Cl-))
so that the concentration of silver ions is controlled by the silver chloride
solubility.
E(AgCl/Ag ) = En(AgCl/Ag) - RT/F ln (a(Cl-))
Since Ksp (AgCl) ~ 1.6 ¥ 10-10 at 25∞C,
En (AgCl/Ag) = 0.22 V
This electrode is reversible to chloride ions.
Another important electrode of this class is the calomel electrode, which is
discussed below.
In order to make accurate potentiometric measurements, a cell must be
constructed that is reproducible and reliable. The emf should depend chieﬂy on
the particular species in the sample that it is intended to measure, and if neces-
sary the system must be calibrated.
As noted in Topic C2, electrode potentials are conventionally referred to the
standard hydrogen electrode (SHE), for which the standard electrode potential,
En = 0.000V. Another electrode, for example the silver electrode, may be com-
bined with the SHE to form a cell. Table 2 gives a representative list of standard
electrode potentials at 25∞C.

Table 2. Standard reduction electrode potentials at 25∞C

Electrode reaction En/V
Li+ + e- = Li -3.04
K+ + e- = K -2.92
Ca2+ + 2e- = Ca -2.87
Na+ + e- = Na -2.71
Mg2+ + 2e- = Mg -2.37
Al3+ + 3e- = Al -1.66
Mn2+ + 2e- = Mn -1.18
Zn2+ + 2e- = Zn -0.76
Fe2+ + 2e- = Fe -0.44
Sn2+ + 2e- = Sn -0.14
Pb2+ + 2e- = Pb -0.13
H+ + e- = 1⁄2 H 0.0000 exactly
AgBr + e- = Ag + Br - +0.10
Sn4+ + 2e- = Sn2+ +0.15
AgCl + e- = Ag + Cl- +0.22
Cu2+ + 2e- = Cu +0.34
Hg2Cl2 + 2e- = 2Hg + 2Cl- +0.27
1
⁄2 I2 + e- = I- +0.54
Fe3+ + e- = Fe2+ +0.76
Ag+ + e- = Ag +0.80
IO3- +6H+ + 5e- = 1⁄2 I2 + 3H2O +1.19
1
⁄2 O2 +2H+ + 2e- = H2O +1.23
-
1
⁄2 Cr2O72 +7H+ + 3e- = Cr3+ + 7/2 H2O +1.33
1
⁄2 Cl2 + e- = Cl- +1.36
Ce4+ + e- = Ce3+ +1.44
MnO4- +8H+ + 5e- = Mn2+ + 4H2O +1.52
C3 – Potentiometry 69

The SHE is rather inconvenient to use, since it requires a supply of inﬂam-

mable hydrogen and has a tendency to change emf slightly as the bubbles of
hydrogen cover the metal.
Reference electrodes should have a constant potential, should not react with
the sample, and should have a very small liquid junction potential with the
sample solution. Two reference electrodes are commonly used.
The calomel reference electrode is shown in Figure 2(a). This is an electrode of
Class 2, with liquid mercury in contact with mercury(I) chloride, or calomel, in a
solution of potassium chloride of high, ﬁxed concentration.
The electrode reaction is
Hg2Cl2 (solid) + 2e- = 2Hg (liquid) + 2Cl- (aq)
The electrode potential is given by:
E(cal) = En(cal) - RT/2F ln (a(Cl-)2)
If a concentrated solution of KCl is used, either saturated or 3.5 M, then this elec-
trode has a constant potential at 25∞C. Changes in the solution outside the electrode
have a very small effect on the potential of this electrode, since the chloride con-
centration is high and is not altered signiﬁcantly by the external solution.
Additionally, the concentrated KCl acts as a salt bridge.
E(cal, sat) = 0.244 V (saturated calomel electrode, SCE)
E(cal, 3.5M) = 0.250 V (calomel electrode, 3.5M KCl)
The silver/silver chloride reference is similar, having a silver wire coated with
silver chloride and in contact with concentrated KCl solution.
E (AgCl, 3.5M) = 0.204 V

(a) (b) (c)

Contact Contact Contact

Fill hole

AgCl/Ag
Hg/Hg2Cl2 electrode
paste

KCI solution

Buffer solution

Porous plug Glass membrane Selective surface

Fig. 2. (a) Calomel reference electrode; (b) glass electrode; (c) solid-state electrode.
70 Section C – Analytical reactions in solution

If the sample solution might react with chloride ions, for example silver or
lead salts, then a double junction reference electrode may be used, with an
additional liquid junction of KNO3.

Redox electrodes
If an inert wire, usually platinum, is placed into a solution containing both
oxidized and reduced species, it will take up an equilibrium potential depen-
dent on the ratio of their concentrations:
Ox + ne- = Red
E(Ox, Red) = En(Ox, Red) + (RT/nF) ln (a(Ox)/a (Red))

Ion-selective electrodes (ISE)

All of the above electrodes are sensitive to particular ions, but have certain
disadvantages, such as loss of response due to poisoning, or a tendency to
mechanical or other failure.
A large variety of special electrodes have been developed to test for a very
wide range of ions. Many involve an insoluble membrane, capable of electrical
conduction, whose surface potential depends chiefly on the concentration of a
particular ion.
The glass electrode for measuring pH, shown in Figure 2(b) is an early
example. A thin membrane of a conducting glass separates the inner solution of
a silver–silver chloride electrode in a chloride-containing solution from the
sample solution outside. Both sides of the glass membrane become hydrated,
and the outer surface exchanges its cations with the hydrogen ions of the sample
solution. The potential of the glass electrode depends on the ratio of the activi-
ties of H+ ions inside and out, on the potential of the inner silver halide electrode
and on the asymmetry potential, characteristic of each particular electrode.
This gives a total electrode potential:
E(glass) = E*(glass) + RT/F ln (a(H+))
= E*(glass) - (2.303 RT/F) pH
The unknown constant E* must be determined by calibration with known
buffers, as described in Topic C4. Modern glass electrodes for pH measurement
often incorporate a reference electrode.
Although these electrodes work very well in the pH range 1-9, at high pH
they suffer from an alkaline error due to the effect of other cations. This is
discussed in the next section. Glass electrodes are also fragile and may suffer
from slow response times.
Crystalline membrane electrodes, constructed either as the glass electrode,
or with a direct contact as shown in Figure 2(c), have an outer crystal surface
which responds to particular ions. The fluoride electrode has a crystal of LaF3,
treated with Eu(II) to increase conductivity, which responds selectively to the
adsorption of free F- ion on its surface. The selectivity is very good, due to the
ability of the small F- on to fit to the LaF3 crystal lattice to the exclusion of larger
ions. However, OH- ions are also small and can interfere, and F- may also form
undissociated hydrofluoric acid. Therefore, it is necessary to use this electrode
in a buffer solution at about pH 6.
Interferences occur with metals such as Al3+, Fe3+ and Ca2+, which combine
with the fluoride. With a SCE reference electrode, the cell emf is:
C3 – Potentiometry 71

E(cell) = E*(F-) - (RT/F) ln (a (F-)) - E(SCE)

or, at 25oC:
E(cell) = k - 0.0257 ln (a(F-))
If a cell is constructed with a fluoride ISE combined with a SCE, calibration with
solutions of fluoride ions of known concentration enable the value of k to be
found. When the same cell is filled with an unknown solution, the fluoride ion
concentration may be determined. This is an example of direct potentiometry.
Other crystalline ion-selective electrodes use AgCl for Cl-, Ag2S for Ag+ and
S2-, Ag2S + CuS for Cu2+ and many more.
Liquid membrane electrodes are shown in Figure 3(a). Typically an
organophilic porous plastic membrane separates the internal reference from the
sample solution. Another reservoir contains a liquid ion-exchanger, which satu-
rates the pores in the membrane. The calcium electrode uses calcium dodecyl
phosphate [(RO)2 PO2]2 Ca where R is the dodecyl radical, dissolved in dioctyl
phenylphosphonate.
The electrode response is given by:
E(Ca2+) = k + (RT/2F) ln (a(Ca2+))

Ion selective ﬁeld effect transistors (ISFETs)

One disadvantage of the glass electrode for measuring pH is its fragility. A
modern development uses a ﬁeld effect transistor where the gate surface is
coated with a material, often aluminum oxide, whose surface is sensitive to pH.
As the potential at the surface changes with pH, the current induced through
the transistor varies. A temperature diode, incorporated in the electrode

Leads to meter
(a) (b)

Ag electrode

Reference
Liquid organic electrode
ion exchanger
Internal
solution

Aqueous solution Glass electrode

saturated with
O-ring
AgCl and MCl2
Gas-permeable
Porous plastic
membrane
membrane with
liquid ion exchanger

Fig. 3. (a) Liquid ion exchange membrane electrode for M2+ ions. (b) Gas sensing electrode
using glass ISE.
72 Section C – Analytical reactions in solution

simultaneously measures the temperature so that a rapid reading of pH,

compensated for temperature is obtained. The electrodes are robust and the
response time is a few seconds. An example is given in Topic H1.

Gas-sensing electrodes
If gases such as CO2 or NH3 are allowed to diffuse into the solution surrounding
a pH electrode, they will alter the pH. The construction is shown in Figure 3(b).
The pH electrode, often incorporating a reference electrode as well, is separated
from the sample solution by a microporous hydrophobic membrane, which
will allow gases but not water to diffuse through rapidly. For CO2 the overall
equilibrium occurs in 3 stages:
(i) carbon dioxide gas diffuses from the outer solution through the membrane
until inner and outer solutions are at the same concentration;
(ii) the solution of CO2 forms the acid H2CO3, which dissociates to form
hydrogen ions:
CO2 + 2H2O = H2CO3 + H2O = H3O+ + HCO3-
K = a(H3O+). a(HCO3-)/(p(CO2 (sample)))
(iii) if the internal solution within the membrane has a constant activity of
HCO3-, for example sodium hydrogen carbonate, then the pH may be
calculated:
pH = -log (H3O+) = log (K. p (CO2 (sample)))
so that by measuring the pH we can ﬁnd the concentration of CO2.
Similar arguments apply with an ammonia-sensing electrode.

Selectivity The ideal electrode should respond only to changes in concentration of a single
ion i. This is rarely true and the response of a real electrode system can be
described by the Nikolsky-Eisenmann equation:
Ei = Eni + S log[a(1) + SKpot1,2 (a(2))(z1/z2)]
where S is the slope of the emf -log(a) plot, which for Nernstian behavior
should be 0.0592/z(1) at 25∞C, a(1) is the activity of the ion selected of charge
z(1), a(2) is the activity of the interfering ion of charge z(2) and Kpot1,2 is the selec-
tivity coefﬁcient for ion 1 with respect to ion 2. The smaller the value of this
selectivity coefﬁcient, the smaller the interference by the second ion.

Direct By calibration, and the use of appropriate electrode response equations, it is

potentiometry possible to measure concentrations directly, as indicated in the example above for
ﬂuoride ion in tap water.
A calibration plot should indicate three things:
(i) whether the electrode responds to the correct ion;
(ii) whether that response is Nernstian;
(iii) what range of concentrations may be studied.
Figure 4 shows the calibration plot for a copper ion selective electrode, where a
total ionic strength adjustment buffer (TISAB), such as 1M NaNO3, has been
added to each solution so that the response is effectively at constant ionic
strength and constant activity coefﬁcient.
C3 – Potentiometry 73

150

100

Electrode potential (mV)

–50

–100
–7 –6 –5 –4 –3 –2 –1 0
log (a)

Fig. 4. Electrode potential response for a copper ISE. (a) Against activity (solid line); (b)
against concentration (dashed line).

This graph shows that over the concentration range 10-1 to 10-5 M the calibra-
tion is linear with a slope of 0.029 V/log (a(Cu2+)). Below about 10-6 M the line
curves, since the solubility of the crystal becomes signiﬁcant. Interferences from
Cu(I), Ag(I) and Hg(II) are troublesome.
If a calibration curve is constructed, direct measurement of solution concen-
trations within that range may be made.

Example
Calibration performed with a ISE selective to Mg2+ ions gave a value of S =
0.0296, and E = 0.411 V for a solution of a(Mg2+) = 1.77 ¥ 10-3 M.
What is the activity of Mg2+ ions when E = 0.439 V?
Substitution gives a(Mg2+) = 1.998 ¥ 10-4 M.

The method of standard additions (Topic B4) has the advantage that comparison
is made in the same matrix solution, so that interferences and other effects are
the same for the unknown and for the standard.
Potentiometric titrations are discussed in Topic C5.
Section C – Analytical reactions in solution

C4 pH AND ITS CONTROL

Key Notes
Deﬁnition of pH Since the concentrations of ions in aqueous solution vary over an
enormous range from greater than 1 molar down to 10-14 molar or less, it
is convenient to use a logarithmic function to describe them. For
hydrogen ions in aqueous solution, denoted by H3O+, often called the
hydronium ion, this is deﬁned as:
pH = -log(a(H3O+)) ⬇ -log(c( H3O+)/mol dm-3)

The pH scale Pure water contains equal amounts hydrogen ions and hydroxyl ions
OH-, at a concentration of 10-7 molar. Therefore, the pH of neutral water
is -log (10-7) = 7. Acids have pH values less than 7, down to less than 1,
while for alkalis, the pH is greater than 7, up to 13 or more.

Buffers For many analytical measurements it is necessary to control the pH so

that it varies as little as possible. Solutions that resist changes in pH are
known as buffer solutions. They usually consist of a mixture of a weak
acid and its salt for lower pH values, or a weak base and its salt for
higher pH values.

pH measurement pH may be estimated approximately using visual indicators, but more

accurate determination requires the use of pH meters. These must be
calibrated using standard buffers.

pH control This is most usually achieved with buffers. Enzyme reactions,

electrophoretic separations and also spectrometric and electrochemical
determinations may all require pH control.

The practical (or operational) deﬁnition of pH recognizes that it is deter-

mined using electrochemical cells having an electrode selective to hydrogen
ions. This has been discussed in Topics C2 and C3, but a typical cell is:
Reference electrode || Solution X | glass electrode.
This gives an output e.m.f , EX,
EX = E* + (RT/F) ln [a(H3O+)X]
The constant E* depends on the exact nature of the reference and glass elec-
trodes, and is best eliminated by calibration with a standard solution S which
has a pH that is accurately known.
ES = E* + (RT/F) ln [a(H3O+)S]
Subtracting these and converting the logarithms gives a practical deﬁnition of
pH:
pH(X) = pH(S) + (ES-EX)/(RT ln(10)/F)
Typical calibration buffers are discussed below.

The pH scale In all aqueous solutions, pH values may range between about 0 and 14 or more
as shown in Figure 1. Molar solutions of strong mineral acids, such as HCl,
HNO3 or H2SO4 have pH values less than 1. Weak acids, such as ethanoic or
citric acid in decimolar solution have a pH of around 3.
A useful standard is 0.05 M potassium hydrogen phthalate which, at 15∞C has
a pH of 4.00. Although pure water is neutral and has a pH of 7.00, freshly
distilled water rapidly absorbs carbon dioxide from the air to form a very dilute
solution of carbonic acid, and therefore has a pH of around 6.

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14

Very acid Acidic Neutral Alkaline Very alkaline

Fig. 1. The pH scale.

Another standard occasionally used is 0.05 M borax (sodium tetraborate,

Na2B4O7), which has a pH of 9.18 at 25∞C.
Dilute alkalis such as ammonia or calcium hydroxide (lime water) have pH
values near to 12, and for molar caustic alkalis, such as NaOH, the pH is over 13.

Buffers As many reactions depend greatly upon the concentration of hydrogen ions in
the solutions being used, it is important to control the pH. This is usually
achieved by using a solution which has a pH that is accurately known and that
resists any change in pH as solvent for the experiment. Such solutions are called
buffers.
The equilibria that govern the reactions of weak acids or bases in aqueous
solution will resist attempts to change them. This is known as Le Chatelier’s
principle. For example, the dissociation of ethanoic acid obeys the equation:
76 Section C – Analytical reactions in solution

CH3COOH [ CH3COO- + H3O+

and an equilibrium constant is written (in terms of concentrations)
Ka = c(Ac-) ¥ c(H3O+)/c(HAc) = 1.75 ¥ 10-5
using the abbreviation Ac for the CH3COO- group. Converting to logarithmic
form, and recalling that pK = -log(K):
pH = pKa + log [c(salt)/c(acid)]
This is the Henderson-Hasselbalch equation.
If we make a mixture containing both the free acid, HAc, and its salt sodium
ethanoate, NaAc, then the equilbrium and the concentrations of acid and salt
will determine the concentration of hydrogen ions and the pH.

Example 1
For a mixture of 50 cm3 of 0.1 M HAc with 40 cm3 of 0.1M NaAc, giving a total
volume of 90 cm3,
c(H3O+) = 1.75 ¥ 10-5 ¥ [(50 ¥ 0.1/90)/(40 ¥ 0.1/90)]
= 2.19 ¥ 10-5 M, so that
pH = 4.66
Addition of acid to this buffer shifts the above equilibrium to the left and most
of the added hydrogen ion combines with the anion. Adding 10 cm3 of 0.1 M
HCl lowers the pH only to about 4.45. If this amount of acid were added to
90 cm3 of water, the pH would be 2.0. Similarly, when alkali is added, the
hydroxyl ions react with the acid to produce more salt. 10 cm3 of 0.1 M NaOH
will raise the pH only to around 4.85. If this amount of alkali were added to
90 cm3 of water, the pH would rise to 12.

Weak bases and their salts behave in much the same way. For example,
ammonia and ammonium chloride:
NH3 + H2O [ NH4+ + OH-
Kb = c(OH-) ¥ c(NH4+)/c(NH3) = 1.75 ¥ 10-5
or, rewriting the Henderson-Hasselbalch equation:
pOH = pKb + log[c (salt)/c(base)]
or, since pH + pOH = 14.0
pH = 14.0 - pKb - log[c(salt)/c(base)]
For a mixture of equal amounts of 0.1 M ammonia and 0.1 M ammonium
chloride
pH = 14.0 - 4.75 = 9.25
A most useful range of buffers is obtained by using salts of a dibasic (or
tribasic) acid such as phosphoric acid, H3PO4 - for example, potassium
dihydrogen phosphate, KH2PO4, and disodium hydrogen phosphate, Na2HPO4.
The equilibrium involved here is:
H2PO4- [ H3O+ + HPO42-
For this equilibrium, the second dissociation constant of phosphoric acid, Ka2, is
close to 1 ¥ 10-7, or pKa2 = 7. Figure 2 shows the effect of adding acid or alkali on
C4 – pH and its control 77

the pH of a mixture containing 50 cm3 each of 0.1 M KH2PO4 and 0.1 M

Na2HPO4, which originally has a pH of 7.0. Adding 0.001 moles of acid to
100 cm3 of water would lower the pH to 2.
The more concentrated the buffer, the greater will be its buffer capacity. This
is the amount of acid (or alkali) that, when added to 1 liter of buffer, will change
its pH by 1 unit. In the above example, the buffer capacity is about 0.04 moles. If
we had used a more concentrated buffer, the capacity would be greater. Very
dilute buffers have little buffer capacity, and hence have limited use.
Table 1 gives a selection of buffers and standard solutions that are useful for
pH control. These solutions and others are often used to calibrate pH meters.

pH measurement Two important methods exist for pH measurement: visual, using indicators,
and potentiometric, by means of electrochemical cells.
Indicators for pH measurement are weak acids (or bases) where the color of
the acid form is different from that of the salt.

8
pH

4
–4 –3 –2 –4 0 1 2 3 4
Millimoles of acid (+) or alkali (–)

Fig. 2. The effect of adding acid (+) or alkali (-) to a phosphate buffer mixture, originally at
pH 7.

Table 1. Buffer solutions

Solution pH at 25∞C
0.05 M potassium hydrogen phthalate 4.008
0.1 M ethanoic acid, 0.1 M sodium ethanoate 4.640
0.025 M potassium dihydrogen phosphate, 6.865
0.025 M disodium hydrogen phosphate
0.01 M borax 9.180
0.1 M ammonia, 0.1 M ammonium chloride 9.250
0.025 M sodium bicarbonate, 10.012
0.025 M sodium carbonate
Saturated calcium hydroxide 12.454
78 Section C – Analytical reactions in solution

HIn [ H3O+ + In-

Colour 1 Colour 2
The dissociation constant of the indicator, KIn, is given by
KIn = c(H3O+) ¥ c(In-)/c(HIn)
For example, for methyl orange (Fig. 3):
Acid form, red [ Alkaline form, yellow + H+
As with buffers, the equilibrium and the concentration of hydrogen ions will
govern the ratio of color 2 to color 1.

–SO
3 NH N N(CH3)2+ = –SO
3 N N N(CH3)2 + H+

Red (pH<3) Yellow (pH>5)

Fig. 3. Structures of methyl orange at low and high pH (only one tautomeric form is shown).

Example 2
With the indicator bromocresol green, where KIn = 1.6 ¥ 10-5, or pKIn = 4.8 and
color 1 (acid) is yellow, while color 2 (salt) is blue, a solution of pH 4.0 will give:
log [c(In-)/c(HIn)] = log [c(color 2)/c(color 1)] = pH - pKIn = -0.8
Therefore, c(color 2)/c(color 1) = 0.16, which means about 14% blue, 86% yellow,
or visually a very yellowish green. When the pH is equal to the pKIn, there are
equal amounts of each form making a green color.
A wide range of indicators is available for titrations and other purposes and
these are discussed further in Topics C5 and C7.
This provides a useful and rapid method of estimating pH by eye; for
example, using litmus paper which is red below about pH 6 and blue above
pH 8. Both wide range and narrow range indicator papers are available to
enable a rapid estimation of pH. However, to determine the pH accurately using
indicators, careful spectrometric comparison would be needed and this is a
time-consuming method that is rarely used.

The pH meter uses a reference electrode and a glass electrode with a high-
resistance voltmeter and affords a rapid and accurate method of measuring pH
(Fig. 4). The calomel reference electrode is decribed fully in Topic C3.
The glass electrode is an example of a membrane ion-selective electrode and
is described in Topic C3. It responds to hydrogen ions:
E(glass) = E* + (RT/F) ln [a(H3O+)X]
Therefore the complete cell
Pt | Hg | Hg2Cl2 (s) | KCl (sat, aq) || Solution X | glass membrane | AgCl | Ag
has an e.m.f. at 25∞C equal to:
E (cell) = (E* - 0.241) + ( RT/F) ln [a(H3O+)X]
C4 – pH and its control 79

pH meter

Hg–Hg2Cl2 paste Ag–AgCl electrode

Saturated KCl Buffer solution

Porous ceramic plug Glass membrane

Unknown solution

Fig. 4. pH measurement system.

By using one of the standards described above, for example, 0.05 M potassium
hydrogen phthalate, which has a pH at 25oC of 4.008, we may eliminate E* and
measure the activity of hydrogen ions, and hence the pH, in the unknown
solution X.
pH(X) = pH(S) + (ES - EX)/(RT ln(10)/F)

pH control It is often necessary to control the pH of a solution, especially if hydrogen ions

are being generated or consumed, or if the nature of the species being analyzed
changes with the pH. A few examples will illustrate this problem.
(i) In recording the UV spectrum of a solution of a weak acid, such as a
phenol, the peak maxima occur at different wavelengths in an acid medium
compared with those in a basic medium. Comparisons may best be made
using a constant pH buffer.
(ii) In complexometric titrations (see Topic C7), such as the determination of
magnesium by EDTA, the complex is formed readily and completely at
high pH, so the titration is carried out using an ammonia-ammonium
chloride buffer to keep the solution at pH 10.
(iii) As noted in Topic C3, the ﬂuoride electrode detects free F- ions very well,
but OH- ions interfere, and H+ ions form undissociated H2F2. It is therefore
essential to make measurements in a buffer of about pH 5-6.
pH is most often controlled by performing the analysis in a suitable buffer
solution. Occasionally, where a reaction produces an acid (or alkali) the tech-
nique of pH-stat titration may be used. Here, the acid produced is detected, and
sufﬁcient alkali added to return the pH to the optimum value.
Section C – Analytical reactions in solution

C5 T ITRIMETRY I: ACID – BASE

TITRATIONS

Key Notes
Titrimetry Titrimetry is an analytical technique for the determination of the
stoichiometry of a reaction by the addition of controlled amounts of a
standard reagent.

Standard solutions Titrations usually involve the addition of controlled volumes of a

standard solution, whose concentration is known accurately, to a solution
of reactant of unknown concentration.

Equivalence points The theoretical amount of solution that must be added until the reaction
and end points is just complete is the equivalence point and the end point in a titration is
the point at which change is detected accurately. In an ideal case, these
points should be the same.

Indicators In order to detect the end point, a visual indicator may be added.
Instrumental methods may also be used.

Potentiometric The use of an electrochemical cell with an indicator electrode and a

titrations reference electrode to measure the concentration at each stage in the
titration facilitates the detection of the end point and the automation of
titrations.

Applications The major applications for acid-base reactions are the determination of
the concentrations or amounts of acids or bases.

Titrimetry In order to obtain accurate quantitative data for a reaction in solution, it is

necessary that the reaction be fast, complete and occur in ﬁxed, reproducible
amounts. The requirement for fast reaction is achieved readily when ionic
species are involved, although in some other cases, it is necessary to warm the
solutions or add a catalyst. The reaction will be complete provided the equilib-
rium constant is large (see Topic C1).
The technique of volumetric analysis is the simplest type of titrimetry, and
involves the addition of controlled volumes of a reagent solution, the titrant, to
a known volume of another solution, the titrand in a volumetric titration. This
procedure may be automated, and the changes detected instrumentally (Topics
C2, C3, C9 and C10). In some cases, excess of a reagent is added and the excess
measured by back titration.
The volumes and concentrations can be measured with high accuracy.
Calibration of volumetric glassware by discharge into weighed containers
C5 – Titrimetry I: acid–base titrations 81

allows the determination of volumes to 0.01 cm3. This corresponds to about

0.05% for a typical titration volume of 20 cm3.
Weighing solutes prior to dissolution gives comparable accuracy. Pre-deter-
mined volumes, for example 25 cm3 are measured with manual or mechanical
pipets, while cumulative volumes of titrant during addition are measured by a
manual or automated buret.
Any reaction used in titrimetry will cause the concentrations of the species in
solution to change. For acid-base reactions, the concentration of hydrogen ions,
and hence the pH, will alter and similar changes in the ionic concentrations
occur with other reactions.
Figure 1 shows the pH values in the titration of a strong base by a strong acid
as a function of the volume of acid added. pH is discussed in Topic C4. The
ﬁgure shows that there is only a small change in pH before about 24 cm3 where
there is still an excess of base, and also after 26 cm3 where there is an excess of
acid. However, in the region around 25 cm3, where the acid exactly neutralizes
the base, the change in pH is very large. This signiﬁes the end point of the
titration.
Provided that the end point coincides with the equivalence point or the stoi-
chiometry of the reaction studied, the amounts of titrant and titrand measured
should correspond to the actual amounts present.
For a general reaction:
aA + bB = cC + dD
a moles of A react with b moles of B to produce c moles of C and d moles of D.
A molar solution contains 1 mole of solute per 1000 cm3 (or 1 l) of solution.
Therefore, 1 cm3 of a 1 M solution contains 1 mmol of solute.

8
pH

0
0 10 20 30 40 50
Volume of acid/cm3

Fig. 1. Titration of 25 cm3 of 0.1 M NaOH with 0.1 M HCl (pH against volume added).
82 Section C – Analytical reactions in solution

Example 1
A 25.00 cm3 aliquot of a solution of a base of known concentration 0.1057 M is
titrated with an acid of unknown concentration. The reaction involved 1 mole of
base and 1 mole of acid. The end point was determined as 24.88 cm3 of acid
added. What is the concentration of the acid?
25.00 cm3 of the base solution contained 25.00 ¥ 0.1057 = 2.6425 mmol base.
From the known reaction, this should be equivalent to 2.6425 mmol of acid.
Since the volume of acid at the end point was 24.88 cm3, the concentration must
be (2.6425/24.88) = 0.1062 M.
In some cases, the end point detected does not correspond exactly with the
equivalence point. This may be due to problems with the reaction, or to the
small amount of reagent needed to react with additional materials (for example,
the added indicator) present in the titrand. In these cases, a blank titration must
be performed, or allowance made for the titration error.

Standard In order to achieve the highest accuracy, it is necessary to use well-established

solutions standard materials as reagents in the primary calibration or standardization of
the reacting solutions (see Topic A5).
The most important of these are called primary standards, and should be
easy to obtain, purify and dry, should be stable and not hygroscopic, but should
be readily soluble and react rapidly and stoichiometrically. They should ideally
have a high relative molecular mass to minimize weighing errors.
The above criteria mean that reagents such as sodium hydroxide, which is
hygroscopic and may react with carbon dioxide from the air, and potassium
permanganate, which slowly decomposes in air, are unsuitable as primary
standards.
Solutions used for quantitative analysis need to be checked and calibrated
frequently. For example, hydrochloric acid solutions should be checked against
sodium carbonate solution and sodium hydroxide against potassium hydrogen
phthalate.

Equivalence The end point of a titration is based upon experimental observation, whereas
points and end the equivalence point is the theoretical value dependent upon the reaction equa-
points tion. In an ideal case, these should be the same, but a check may be needed to
ensure that factors such as blank errors do not affect the results.
In any titration, the end point corresponds to rapid changes in the concentra-
tion of species. This may be detected in many ways. Instrumental methods are
discussed later, and visual indicators are discussed below.
For reactions such as a strong acid neutralizing a strong base (Fig. 1), the
change at the end point is large, and the rate of change with volume is very
great, as shown by the derivative plot in Figure 2(b). In other cases, such as the
reaction of weak acids with weak bases, the titration curve shows a much less
pronounced change, and the derivative plot may be needed to conﬁrm the end
point, as in Figure 2(b). The second derivative plot is also useful, but relies
greatly on very accurate measurements. Plots of the concentration of a species
(for example H+, or OH-), against the volume added, give straight lines inter-
secting at the end point.
If a mixture of acids, or a polybasic acid such as maleic acid, is titrated, then
two end points are obtained. Similarly, for mixtures of two titrands, two end
points will be detected, provided the species have sufﬁciently different equilib-
rium constants for the reaction (e.g. Ka1 Ka2).
C5 – Titrimetry I: acid–base titrations 83

(a) (b)
12 0.35

0.3
10

0.25
8
0.2

dpH/dV
pH

6
0.15

4
0.1

2 0.05

0 0
0 10 20 30 40 50 60 0 10 20 30 40 50 60
Volume of acid (cm3) Volume of acid (cm3)

Fig. 2. (a) Titration of 25.00 cm3 of 0.105 M ammonia with an ethanoic acid solution. (b) Derivative plot of d (pH)/dV
showing the end point as 26.80 cm3.

Indicators In general, indicators have two forms, which possess different colors. Indicators
for acid-base titrations are themselves weak acids or bases where the two forms
differ in color as shown in Figure 3 of Topic C4 for methyl orange.
HIn (color 1) = H+ + In- (color 2)
The choice of indicator depends upon the reaction to be studied. As noted in
Topic C2, the equilibrium constant of the indicator must match the pH range, or
electrode potential range of the species being titrated. Table 1 shows a selection
of indicators for acid-base reactions.
As a general rule it is noted that the color change takes place over the range:
pH = pKIn ± 1
Similarly for other reactions, the concentration at which the indicator changes
must match the concentrations at the end point.
For example, in the titration of the weak dibasic maleic acid with sodium
hydroxide, the ﬁrst end point, corresponding to sodium hydrogen maleate
occurs at pH = 3.5, while the second end point for disodium maleate is at
pH = 9. Since pKIn = 3.7 for methyl orange, this will change at the ﬁrst end point,
while phenolphthalein would change color around pH = 9.

Table 1. Typical visual indicators for acid-base titrimetry

Acid–base Low pH High pH pKIn
color color
Thymol blue Red Yellow 1.7
Methyl orange Red Yellow 3.7
Bromothymol blue Yellow Blue 7.0
Phenolphthalein Colorless Red 9.6
Alizarin yellow R Yellow Orange 11
84 Section C – Analytical reactions in solution

Potentiometric The use of electrodes, particularly the glass electrode for pH measurements and
titrations the wide range of other ion selective electrodes (ISE) described in Topic C3,
enables titrations to be studied throughout the addition of titrant, so that small
changes may be detected. It also allows automation of the titration.
To compare potentiometric titrations with those where the end point is
detected visually, it is useful to think of the ion selective electrode as an indi-
cator electrode responding to the ion to be detected, and to remember that it
must be combined with a reference electrode, the potential of which must not
be affected by the titration reactions.
In general, the titration will produce a graph such as Figure 2(a) where pH,
pIon or E(cell) is plotted against the volume of titrant added. The sharp change
at the end point is readily observed. Derivative plots such as Figure 2(b) are an
aid to ﬁnding the end point.
Three types of potentiometric titration may be recognized. Detailed discus-
sion of the reactions is given in Topics C6 and C7, but the principles will be
discussed here.
● R, or reagent sensed. A reagent is added, for example, some copper(II)
EDTA complex, which is sensed by a copper ISE. Addition of EDTA to a
solution containing Ca2+ (or Mg2+, Ni2+ etc.) ions gives an abrupt change at the
end point where the excess EDTA reduces the concentration of Cu2+ ions.
● S, or sample sensed. The ISE senses the ion in the solution being titrated. pH
titrations using a glass electrode belong to this type, as do titrations of
ﬂuoride with La3+.
● T, or titrant sensed. The titrant added to the sample is sensed by an ISE
responsive to that ion; for example a silver ion ISE for the titration of halide
ions by silver nitrate.
Potentiometric titrations for complex, precipitation and redox systems are
discussed in Topic C7.

Applications There are many applications for acid-base titrations, several of which are
routinely used analytical methods described in the appropriate topics.
● The determination of the concentration of acid in foods and pharmaceuticals.
● The measurement of acid number (or base number) during the course of a
reaction. For example, in the production of polyester resins by the reaction of
a glycol with maleic and phthalic acids, the total acid remaining is deter-
mined by titration of a weighed sample with potassium hydroxide using
phenolphthalein as indicator.
● The Kjeldahl method for nitrogen determination is a good example of a back
titration. The sample (for example, a food product) is oxidized by concen-
trated sulfuric acid to remove carbonaceous matter. Excess sodium
hydroxide solution is then added, and the ammonia released is carefully
distilled off into a known volume of standard acid, such as 0.1 M boric acid.
The excess acid is then titrated with standard alkali.
Automated titrations are important in producing rapid, reproducible results
in commercial and research laboratories. Samples may be prepared and loaded
using mechanical pipets or direct weighing and dissolution methods. Titrant is
added to the sample titrand solution using peristaltic pumps, or burets driven
by pressure or piston systems. The addition is very reproducible after accurate
calibration. The progress of the titration is most often followed by potentio-
metric measurements, as outlined above (see also Topic H2).
Section C – Analytical reactions in solution

C6 C OMPLEXATION , SOLUBILITY
AND REDOX EQUILIBRIA

Key Notes

Complexation A complex is formed by reactions between two or more species that are
capable of independent existence. Often this is between a metal ion, M
and a coordinating molecule L.

Solubility Ionic reactions producing a compound that is insoluble in the chosen

solvent used may be used for analysis.

Redox equilibria Where one species is reduced while the other is simultaneously oxidized,
the reactions are termed redox reactions and are useful in many
analytical methods.

Complexation The formation of stable compounds and complexes is important in analytical

chemistry, since many species may be formed in a real sample. The amounts
and nature of the species present are analyzed to study speciation. For example,
in a natural water sample, the metal ions may form complexes with water
molecules, carbonate species, plant acids or pollutants. Complexes may be used
for titrations, both directly and for masking unwanted reactions.
The formation of a complex compound between an acceptor species, most
usually a metal ion, M, and a coordinating species, or donor ligand, L, involves
the formation of coordinate bonds, for example, hexamino cobalt (III)
Co3+ + 6NH3 = Co (NH3)63+
The formation of such complexes involves interactions between the orbitals of
the central atom and suitable orbitals or lone pair electrons of the ligands. The
structure and stability of the complexes are discussed more fully in textbooks of
inorganic chemistry.
The formation constant for the equilibrium may be represented in two
ways. The stepwise formation constants, Kf, relates to each addition of a ligand
molecule:
M + L = ML Kf1 = c (ML)/(c (M) . c (L))
ML + L = ML2 Kf2 = c (ML2)/(c (ML) . c (L))
or, generally:
MLn-1 + L = MLn Kfn = c (MLn)/(c (MLn–1) . c (L))
The overall formation constant, b, relates to the formation of the entire complex
in one equation, so that for the complex with n ligands:
86 Section C – Analytical reactions in solution

M + nL = MLn
bn = c (MLn)/(c (M) . c (L) n)
The overall formation constant is the product of the stepwise formation
constants:
bn = Kf1 . Kf2. … Kfn or log bn = log Kf1 + log Kf2 +…+ log Kfn
When a ligand is used that can bond to several sites, it is referred to as a multi-
dentate ligand. One of the most important examples is ethylenediamine
tetracetic acid (EDTA):
(HOOCCH2)2-N-CH2-CH2-N- (CH2COOH)2
This tetrabasic acid, abbreviated to H4Y, has four acetate group and two nitro-
gens, which may complex to the central metal ion, as shown in Figure 1. It is
important to remember that for a satisfactory titration, the equilibrium constant
K of the reaction must be greater than 1000.
Since the concentrations of the various species containing Y (H4Y, H3Y-, H2Y2-
, HY3-, Y4-) will vary with the pH, a formation constant K MY may be written:
Mn+ + Y4- = MY(n-4)+
KMY = c(MY(n-4)+)/(c(Mn+).c(Y4-))
and c(Y4-) = cL ¥ a4
where a4 depends on the pH and the acid dissociation constants (K1, K2, K3, K4)
of EDTA, as shown in Table 1, and cL is the total concentration of all the ligand
species.

2–
O
C
OO H2CH2
C C
O
N
M CH2
O
N CH2
O C C
H2 O
CH2
C
O

Fig. 1. The structure of a metal-EDTA chelate showing its octahedral geometry.

Table 1. Values of a4 as a function of pH

pH a4 pH a4
-14
2 3.7 ¥ 10 8 5.4 ¥ 10-3
3 2.5 ¥ 10-11 9 5.2 ¥ 10-2
4 3.6 ¥ 10-9 10 3.5 ¥ 10-1
5 3.5 ¥ 10-7 11 8.5 ¥ 10-1
6 2.2 ¥ 10-5 12 9.8 ¥ 10-1
7 4.8 ¥ 10-4
C6 – Complexation, solubility and redox equilibria 87

To compare the formation constants at similar pH conditions the equilibrium

may be written:
K’MY = KMY a4 = c(MY(n-4)+)/(c(Mn+) . cL)

Example
The formation constant KMY for magnesium-EDTA is 5 ¥ 108. Should pH 5, or
pH 10 be used to titrate Mg?
At pH 5, for Mg:
K’MY = 5 ¥ 108 ¥ 3.5 ¥ 10-7 = 1.75 ¥ 102
At pH 10, for Mg:
K’MY = 5 ¥ 108 ¥ 3.5 ¥ 10-1 = 1.75 ¥ 108
Therefore, magnesium could be titrated only at pH 10, as the value of the equi-
librium constant is too small below this.
The fact that EDTA forms a number of ring systems adds considerably to the
stability of the complex. This is called the chelate effect.

Solubility The formation of insoluble compounds by reaction between two soluble species
is discussed in detail in Topic C8, where the amount of an insoluble product is
measured gravimetrically (Table 2).
The solubility equilibrium is described by the solubility product, Ksp:
Mn+(solvated) + An-(solvated) = MA(solid) + solvent

Ksp = a (Mn+) . a (An-)

(Note: This equilibrium constant is written in the inverse way to most others,
and thus a very small solubility product is desirable for complete reaction. The
activities of the pure solid MA and of the solvent are taken as 1.)
There are some general rules governing which cations will form precipitates
with which anions. Nitrates and perchlorates are generally soluble.
Various other reagents are also useful for gravimetric analysis, for example
dimethylglyoxime (a) for nickel and oxine (b) for aluminum and magnesium
(Fig. 2(a) and (b)).
If an excess of a precipitating ion is present, in order to keep the equilibrium
constant unchanged, the concentration of the other ion must decrease. This is
called the common ion effect. Occasionally, complex formation may occur, so
that the precipitate formed redissolves when excess reagent is added.
Since equilibrium constants are thermodynamic quantities, they should really
be written in terms of activities and the activity coefﬁcient, depending on the
ionic strength taken into account.

Table 2. Ions forming insoluble products (Ksp ~ < 10-5)

Anions Cations
- - -
Cl , Br , I Ag+, Pb2+, Cu+, Hg22+
CO32- Ca2+, Sr2+, Ba2+, Mg2+
SO42- Ca2+, Ba2+, Pb2+
OH- Al3+, Co3+, Cr3+, Fe3+, Mg2+, Mn,2+ Ni2+, Zn2+
S2- Ag+, Cd2+, Cu2+, Fe2+, Hg2+, Mn2+, Pb2+, Zn2+
88 Section C – Analytical reactions in solution

(a)
CH3C CCH3

NOH NOH

(b)

OH
Fig. 2. Reagents for the precipitation of metal ions. (a) Dimethylglyoxime. (b) Oxine.

Redox equilibria In order to establish a scale of oxidative power, it is necessary to have a stan-
dard, and since these reactions involve electrons, measurement of the reduction
electrode potential is a convenient way to do this. The details are given more
fully in Topic C3.
Some standard reduction electrode potentials, where the reagents are at unit
activity, at 25∞C are given in Table 3. These potentials allow the prediction of
which ions will oxidize other ions, under standard conditions, that is when the
concentrations are molar. A more poisitve electrode potential will oxidize a
more negative potential.
It was shown in Topic C2 that the electrochemical cell e.m.f. is related to the
free energy change, and hence to the equilibrium constant:
En = (RT/nF) ln K
Therefore, the larger the cell e.m.f, the larger the equilibrium constant, and the
more complete the reaction.

Example
For the reaction of cerium(IV) ions with iron(II) ions, what is the likely reaction,
and what is the equilibrium constant? Which reagent is the oxidizing agent, and
which the reducing agent?
Cell: Pt | Fe2+ , Fe3+ || Ce4+, Ce3+ | Pt
En (cell) = En (rhs) – En (lhs) = 1.44 - 0.77 = 0.67 V

Table 3. Standard reduction electrode potentials of some common redox systems at

25∞C
Reaction E n/V
H2O2 + 2H+ + 2e- = 2H2O 1.77
MnO4- + 8H+ + 5e- = Mn2+ + 5H2O 1.51
Ce4+ + e- = Ce3+ 1.44
Cr2O72-+14H+ + 6e- = 2Cr3+ + 7H2O 1.33
I2 + 2e- = 2I- 0.54
Fe3+ + e- = Fe2+ 0.77
S4O62- + 2e- = 2S2O32- 0.08
2CO2 + 2e- = C2O42- -0.49
C6 – Complexation, solubility and redox equilibria 89

ln K = 0.67/(8.314 ¥ 298/(1 ¥ 96485)) = 26.09

or K = 2.1 ¥ 1011
Cerium(IV) oxidizes iron(II) almost completely.
Examination of Table 3 shows that permanganate will oxidize iron(II) and
oxalate, iodine will oxidize thiosulfate to dithionite, but iodide will be oxidized
by iron(III) to iodine.
Section C – Analytical reactions in solution

C7 T ITRIMETRY II:
COMPLEXATION ,
PRECIPITATION AND REDOX
TITRATIONS

Key Notes

Complexation The techniques of titrimetry, using both visual and potentiometric end
titrations point detection, are used to measure species, particularly metal ions, in a
wide range of samples.

Precipitation Reactions producing an insoluble product are valuable analytical tools for
titrations the titrimetric determination of halide and other anions.

Redox titrations Oxidation and reduction titrations may be used to measure many species,
especially metals in high or low valency states, iodine and iodides, and
easily oxidized organic compounds.

Table 1. Metal-EDTA formation constants at 25∞C

Cation KMY log (KMY)
+ 7
Ag 2.0 ¥ 10 7.3
Mg2+ 4.9 ¥ 108 8.7
Ca2+ 5.0 ¥ 1010 10.7
Fe2+ 2.1 ¥ 1014 14.3
Fe3+ 1.0 ¥ 1025 25.1
Zn2+ 3.2 ¥ 1016 16.5
Cd2+ 2.9 ¥ 1016 16.5
V3+ 8.0 ¥ 1025 25.9
C7 – Titrimetry II: complexation, precipitation and redox titrations 91

It is possible to titrate two cationic species in a solution by performing the

titration at different pH values. However, if a solution of high pH must be used,
this might cause precipitation of metal hydroxides or other insoluble species. In
order to prevent this, secondary complexing agents can be added to retain the
metal ion in solution. Ammonium chloride and triethylamine are typical
reagents for this purpose.
Zinc ions, which might otherwise form insoluble Zn(OH)2 at pH of 10, may be
converted to soluble zinc amine complexes. These are less stable than the EDTA
complex and the zinc may then be reacted quantitatively.
Secondary complexing agents may also act as masking agents. Examples
of this are the use of cyanide ions to form stronger complexes with heavy
metal ions so that magnesium can be titrated, or masking Fe2+ and Mn2+ using
hydroxylamine in water hardness determinations.
Standard solutions of EDTA may be prepared from the dry disodium salt
(Na2H2Y, RMM 336 or the dihydrate, RMM 372), by dissolving a known amount
in water free of heavy metals. Alternatively, the solution may be standardized
by a standard magnesium salt solution.
In complex and precipitation titrations, as in others, the end point corre-
sponds to a rapid change in the concentration of species. This may be detected
by instrumental methods, particularly potentiometry (see Topic C3) and by
visual indicators discussed below. Using suitable indicators, or potentiometric
measurements, it is possible to detect two or more end points.
Complexometric indicators behave in a similar way to titrating complexing
agents such as EDTA. They generally change color with pH, but one species, for
example HIn2-, will react with excess metal ions Mn+:
HIn2- + Mn+ = MIn(n-3)+ + H+
blue red
Some selected indicators for complexometric titrations are given in Table 2.
In order to use potentiometric methods to study complexometric titrations,
an electrode speciﬁc to the metal ion may be used (see Topics C3 and C5), for
example, a copper ISE to follow the reaction of copper with EDTA.
Alternatively, a ‘J’-shaped electrode with a small mercury pool may be used
together with a small amount of added Hg-EDTA complex. This acts in a similar
way to the Class 2 electrodes, where the complexes determine the concentration
of ions in contact with the mercury pool:
Mn+ + HY3- = MY(n-4)+ + H+
Hg2+ + HY3- = HgY2- + H+
Hg2+ + 2e- = Hg

Table 2. Indicators for complexometric titrations

Indicator Free color Complex color* Metal ions
Eriochrome black T Blue Red Ba, Cd, Ca, Pb, Zn
Pyrocatechol violet Yellow Blue Al, Bi, Cd, Co, Cu
Fe, Mg, Mn, Ni, Zn
Xylenol orange Yellow Red Bio, Cd, Pb, Th, Zn
Calcon carboxylic acid Blue Red Ca, Cd, Mg, Mn, Zn
*varies with metal and pH
92 Section C – Analytical reactions in solution

The major application of complexation titrations is for the determination of

the concentrations or amounts of metallic elements in water, food and other
industrial samples.

Example
The calcium and magnesium ions in hard water may be determined. The solution
is adjusted to pH 12 with NaOH, when Mg(OH)2 is precipitated. The calcium is
then titrated with EDTA using calcon carboxylic acid as indicator. Both calcium
and magnesium are then determined in by titrating a sample with EDTA at pH 10
using eriochrome black T, and ﬁnding the magnesium by difference.

Precipitation For precipitation reactions, the change in the concentration of either ion
titrations forming the precipitate may be considered. Since the changes often involve
many orders of magnitude of concentration, it is again convenient to use the pX
notation. For example, for the reaction of silver ions with chloride to form an
insoluble silver chloride precipitate
Ag+ + Cl- = AgCl (s)
the concentration may be expressed as:
pAg = -log (a (Ag+)) ~ -log (c (Ag+))
Figure 1 shows the pAg values in the titration of sodium chloride by silver
nitrate as a function of the volume of silver nitrate added. This ﬁgure shows
that, before the end point pAg is very high (that is, the concentration of silver
ions is small) and changes little, because there is still an excess of chloride and
the silver is almost completely removed as precipitate. After the end point, there
is an excess of silver ions, the concentration increases and pAg decreases. In the
region around the end point, where the amounts are nearly equal, the change in
pAg with volume added is very large. If a mixture of iodide and chloride ions is
titrated (dashed line), the iodide, which is less soluble, precipitates ﬁrst and pAg
is even higher than for chloride. Then the chloride precipitates. Both end points
can be found.
As noted in Table 1 of Topic A5, silver nitrate, sodium chloride and potassium
chloride are primary standards for silver halide precipitation reactions. Other

8
pAg

4
(b)
(a)
2

0
0 5 10 15 20 25 30 35
Volume AgNO3 added
Fig. 1. Silver-halide titrations. (a) Chloride alone (solid line); (b) iodide plus chloride (dashed
line).
C7 – Titrimetry II: complexation, precipitation and redox titrations 93

precipitation titrations (e.g., barium with sulfate, zinc with ferrocyanide) are less
commonly performed.
Indicators for silver-halide precipitation titrations are of two types. The first
react specifically when an excess of titrant becomes present immediately after
the end point - for example, if a small amount of potassium chromate is added,
it will react with excess silver ions to produce deep red silver chromate in
neutral solutions (Mohr’s method). In acid solutions, the silver is titrated with
potassium thiocyanate (KCNS) solution (Volhard’s method). Iron (III) ammo-
nium sulfate solution is added and reacts with an excesss of thiocyanate to
produce a deep red iron thiocyanate species.
Adsorption indicators such as fluorescein adsorb onto the precipitate when
excess silver ions are present and the precipitate takes on a pinkish color.
As with other indicators, the change of color is detectable by eye over a range:
log (c(ion)) = ± 1
A selection of indicators for precipitation titrations is given in Table 3.

Table 3. Indicators for precipitation titrations

Precipitation Color 1 Color 2 Ions detected
CrO 4
2-
Yellow Deep red Ag+
Fe3+ Light brown Deep red CNS-
Fluorescein Green-yellow Pink Ag+

It is considerably easier to titrate mixed chloride and bromide in solution by

potentiometry. The use of a silver ion selective electrode, or even a silver wire,
together with a double junction reference electrode, since the chloride ions from
a calomel electrode would react, allows the determination of the silver ion
concentration. Other precipitation titrations may be followed using suitable ion
selective electrodes.
The major applications of precipitation titrations involve the determination
of halides with silver, or the reverse, or the determination of silver in acid
solutions with thiocyanate.

Example
An insecticide containing chlorine was digested in nitric acid to convert the
chlorine to soluble chloride. Silver nitrate was added in excess, and the excess
titrated with potassium thiocyanate by Volhard’s method. It is important to
know whether all the chlorine is converted to chloride.

Redox titrations Oxidation-reduction or redox titrations are used for determining metals with
two well-deﬁned oxidation states, and indirect methods for the determination of
organic compounds.
For redox reactions the concentrations both of the oxidized species, Ox, and
of the reduced species, Red, will change simultaneously. Considering a cell with
a redox electrode and a reference electrode:
SCE || a (Ox), a (Red) | Pt
the cell emf is given by:
E = En + (RT/F) ln (a (Ox)/a (Red)) - ESCE
94 Section C – Analytical reactions in solution

Therefore, as the reaction proceeds during titration, the ratio of the concentra-
tions will change and the emf will alter. The potentiometric titration curve will
resemble those described in Topic C5. A summary list of redox reagents is given
in Table 3 of Topic C6.
For standard solutions, sodium oxalate and iron(II) ammonium sulfate and
potassium iodate are suitable, but potassium permanganate and iodine solu-
tions decompose on standing and must be standardized before use.
The indicators for redox reactions are reagents whose oxidized and reduced
forms differ in color:
In(Ox) (color 3) + ne- = In(Red) (color 4)
An example of this is 1,10-phenanthroline iron (II)
[Fe (C12H8 N2)3]3+ + e- = [Fe (C12H8 N2)3]2+
oxidized form, pale blue = reduced form, deep red
In several cases, the indicator reaction additionally involves hydrogen ions,
so the change is pH dependent. Table 4 lists commonly used redox indicators.

Table 4. Indicators for redox titrations

Redox Oxidized Reduced EIn/V Solution
color color
1,10-phenanthroline iron(II) complex Pale blue Red 1.11 1 M H2SO4
Diphenylamine Violet Colorless 0.76 Dilute acid
Methylene blue Blue Colorless 0.53 1M acid
Phenosafranine Red Colorless 0.28 1M acid

For a redox indicator where one electron is involved, at 25∞C, the color
change takes place at electrode potentials in the range
E = EIn ± 0.059
One further useful indicator employed in redox titrations involving iodine is
starch, or more synthetic equivalent materials. The starch forms a blue-black
complex with iodine, which is rendered colorless when all the iodine has been
removed.
The applications of redox titrations include the determination of metals, with
two well-defined oxidation states, which are present in metallurgical samples
and ores. In order to dissolve the material, it may be necessary to use oxidative
conditions, for example, concentrated nitric acid. This will convert the majority
of the ions into their higher oxidation state, and in order to titrate them they
must first be reduced quantitatively. This may be done by passing the acidified
solution through a Jones reductor, which contains a zinc-mercury amalgam.
The effluent may then be titrated using a suitable oxidant. Some organic
compounds, such as phenols, may be determined by bromination with a
bromate/bromide mixture, followed by back titration of the excess using
thiosulfate.
Section C – Analytical reactions in solution

C8 G RAVIMETRY

Key notes

Gravimetry Gravimetry is the analytical technique of obtaining a stable solid

compound, of known stoichiometric composition so that the amount of
an analyte in the sample may be found by weighing.

Precipitation This involves treatment of an analytical sample, usually in solution, to

obtain a quantitative amount of an insoluble compound of known
composition.

Puriﬁcation The precipitate must be as pure as possible. Substances that are similar
and might precipitate under the same conditions must be removed, and
the analysis must be carried out so that no impurities are co-precipitated.

Drying and heating If precipitation is carried out from solution, the solid precipitate will have
solvent associated with it. This must be removed. Heating near the
boiling point of the solvent will do this, and further heating may be
needed to obtain a more stable compound whose formula is known.

Weighing The procedures of weighing the container initially and with the ﬁnal
sample are most important.

time consuming. Newer instrumental methods may determine several compo-

nents simultaneously, rapidly and are generally applicable down to trace levels.

Precipitation As noted in Topic C1, many elements will form compounds insoluble in water
or other solvents. Provided the compound is stable, or may be converted into a
stable compound easily, these insoluble precipitates may be used for analysis.
The technique for obtaining a precipitate may be summarized as follows:
(i) The sample should be dissolved as completely as possible in a suitable
solvent. Any residue that does not dissolve (for example, silica present in
the metal sample of the above example), may be filtered off at this stage.
(ii) Unless there are undesirable changes when the sample is heated, the solu-
tions should be warmed. This speeds up reactions and helps to form a more
granular precipitate.
(iii) The precipitating reagent must be chosen to give as insoluble a precipitate
as possible. Preferably, a reagent that will produce the largest mass of
precipitate should be used. For example, aluminum may be precipitated
and heated to give the oxide, Al2O3 when 10.0 mg of aluminum will
produce 18.9 mg precipitate. If ‘aluminon’ (8-hydroxyquinoline, C9H6ON) is
used, 170.0 mg of precipitate results.
(iv) The precipitating reagent should be added slowly, with stirring to the
warm solution. To check whether precipitation is complete, the precipitate
is allowed to settle, and more reagent added. If further precipitate does not
form, the reaction is complete. If the solution appears cloudy, it is possible
that a colloidal form of the solid is present. This may be coagulated by
further warming or adding more reagent.
(v) The reaction mixture is filtered. Various means may be used for this. The
simplest is a quantitative filter paper (ashless), which has been dried and
weighed previously. These may be dried, or burnt (‘ashed’).
Another useful filter is a sintered glass or porcelain crucible, dried and
weighed as before. Glass will withstand heating to about 300∞C and porcelain to
over 800∞C.

Puriﬁcation The precipitate should be washed to remove traces of solution. This may cause
difﬁculties, as the washing may redissolve the solid. Using a wash liquid with a
common ion reduces the solubility.
For example, if 0.18 g of a precipitate of lead sulfate is washed with 1 dm3 of
distilled water, it will dissolve 0.046 g of precipitate or 25%, as the solubility
product is Ksp = 2.3 ¥ 10-8 (mol dm-3)2. If the precipitate is washed with 1 dm3 of
0.01 M sulfuric acid, then the amount dissolved is much less, 0.7 mg or 0.4%.

Drying and Drying can be done in stages. To remove water, the ﬁltered solid in its container
heating is placed in a desiccator and left for a few hours. A vacuum desiccator is even
more efﬁcient for removing solvents at low temperature. Heating in ovens,
furnaces or directly with burners will raise the temperature to remove materials
or to decompose the precipitate to a more stable form. For example, ‘basic
aluminum succinate’ is a good precipitate for aluminum, but must be ignited to
constant weight at about 1200∞C to convert to aluminum oxide.
The stages involved in drying and decomposition can be studied using
thermogravimetry (see Topic G1).
C8 – Gravimetry 97

Weighing Modern balances can readily weigh samples directly, and masses from several
grams down to a few micrograms can be weighed accurately and quickly. It is
important that the conditions are the same for the initial weighing (crucible,
ﬁlter paper) as for the ﬁnal weighings. Temperature is especially important and
hot samples should never be placed directly onto a balance pan.
Section C – Analytical reactions in solution

C9 V OLTAMMETRY AND
AMPEROMETRY

Key Notes
Principles Voltammetry is the study of the variation of current with applied
potential in an electrolysis cell where the reactions are controlled by the
diffusion of the sample species. The current is proportional to the
concentration of the electroactive species and amperometric methods
involve current measurement.

Instrumentation The cell uses a working microelectrode, a reference electrode, and a

counter electrode, and a controlled voltage supply.

Applications These methods are used qualitatively to determine the nature of metal
and organic species and their reactions, and quantitatively to measure
trace levels of metals and organic compounds.

Principles Voltammetric techniques involve the electrolysis of the solution to be analyzed

using a controlled external power source and measuring the resultant current-
potential or current time curves to obtain information about the solution.
The species to be determined undergoes oxidation or reduction at a
working electrode. The voltage between the working electrode and an auxil-
lary or counter electrode is controlled by the external circuitry in order to
maintain a preselected potential difference at the working electrode, with
respect to the reference electrode, as a function of time. A typical voltam-
metric cell, shown in Figure 1, has a working electrode, reference electrode
and an auxillary electrode and contains the solution to be analyzed. Often the
solution is deaerated with nitrogen to prevent interference due to the reduc-
tion reactions of oxygen.
If there is no reaction at the working electrode, the potential changes greatly for
a very small increase in current. A mercury drop electrode, for example, has a
polarization range between +0.3 V and -2.7 V against the SCE and in the absence
of oxygen so that many reactions that occur in that range may be studied.
By controlling the potential of the working electrode, a particular reaction
may be selected. Suppose a cell has two inert, solid electrodes and a reference
electrode, which dip into an aqueous solution containing copper ions.
In order to cause any reaction, the applied potential must exceed the decom-
position potential. This may be calculated by considering the reactions at each
electrode, and adding the extra potentials or overpotentials due to polarization
effects at the electrodes and to the voltage needed to drive the current against
the resistance of the solution. In this example, in order to drive the cell reaction,
a voltage greater than about –2.54 V must be applied.
C9 – Voltammetry and amperometry 99

Potentiostat

V A

Nitrogen in Nitrogen out

Working
electrode Reference electrode

Auxillary electrode
Sample
solution Stirrer

Fig. 1. Basic voltammetric cell.

The reaction that occurs is:

Cu2+ + H2O = Cu(s) + 1⁄2 O2 + 2H+
The copper is then deposited on the cathode and oxygen is evolved at the
anode. This is the basis for electrogravimetry, where the copper is completely
deposited from solution and the increase in weight of the cathode determined.
The analysis may be conducted using either a controlled potential or a
controlled current.
Coulometric methods of analysis involve measuring the quantity of electricity
in coulombs needed to convert the analyte to a different oxidation state. If the
electrolysis occurs at 100% efﬁciency, Faraday’s laws may be applied and each
96485 C will bring about the reaction corresponding to 1 mole of electron transfer.
For example, using a silver anode, the passage of a current produces silver
ions, which react with any chloride in the solution. Bromine and acids may also
be generated coulometrically.
Polarographic methods employ a microelectrode, often a dropping mercury
electrode (DME), as the working electrode, plus a reference electrode (SCE) and a
mercury pool as auxillary electrode. The simplest potential-time regime, where
the potential increases regularly (linear potential sweep dc voltammetry) is
applied to the cell containing the analyte and a supporting electrolyte to carry
the majority of the current.
In these methods, the transport of ions to the electrodes depends on three
factors: diffusion, convection or stirring, and conduction. The effects of conduc-
tion of the ion that reacts at the electrode is minimized by using a concentration
of supporting electrolyte such as KCl about 50-times higher than that of the
analyte. Stirring and convection are minimized. The resulting polarographic
curve shows three regions.
(i) If a potential difference is applied across a cell and no reaction occurs, only
the residual current Ir will ﬂow.
(ii) If a reducible ion, say Cd2+, is present, it will migrate to the dropping
mercury cathode. If the applied potential exceeds its decomposition poten-
tial, ED, it will be reduced to the metal which dissolves in the mercury:
Hg
Cd2+ + 2e- = Cd(Hg)
100 Section C – Analytical reactions in solution

As the cadmium plates out, the layer around the electrode is depleted and
more cadmium ions must diffuse in from outside through the diffusion
layer of thickness d. This will cause a current, I, to ﬂow, which depends on
the concentration gradient between the bulk solution and the surface.
Eventually, the surface concentration becomes zero, and the limiting diffu-
sion current is reached:
Id = constant (c(bulk))/d = kS (c(bulk))
The constant, kS, depends on the number of electrons transferred, the diffu-
sion coefﬁcient of the ion in the solution, and the characteristics of the
cathode.
(iii) If the potential is increased further, the current does not increase unless
other reducible ions are present. These three regions are shown in Figure 2.
The potential difference, E, across the cell at any stage is:
E = ESCE - EDME or
n
E = ESCE - (E Cd + (RT/2F) ln [(a(Cd2+, surface)/a(Cd(Hg))]
From the equations above, the concentrations may be substituted by the
currents, since the concentration of reduced species in the mercury depends on
the current I and the diffusion constant in the amalgam, kA
I = kA(c(Cd(Hg))
E = ESCE - (EnCd + (RT/2F) ln (kA/kS)) + (RT/2F) ln [(Id - I)/I]
When I = 1⁄2 Id, that is at the half-wave position, the DME has the half-wave
potential, E ⁄ 1
2

EDME, ⁄ = ESCE - (E ⁄ + (RT/2F) ln [1⁄2 Id/1⁄2 Id] = E ⁄

1
2
1
2
1
2

The half-wave potential is usually quoted relative to the SCE, and, like the
standard electrode potential, is characteristic of the electrode reaction. Typical
values are shown in Table 1.

Limiting current
Current, I

Residual current
E1/2

–0.5 –0.7 –0.9 –1.1

E/(V)

Fig. 2. Current-voltage curve for 10-4 M cadmium sulfate solution.

C9 – Voltammetry and amperometry 101

Table 1. Value of the half-wave potential

Ion E ⁄ /V with respect to the SCE
1
2

in 0.1 M KCl in 1 M NH3, NH4Cl

Cd2+ -0.60 -0.81
Cu2+ +0.04 -0.24
-0.22 -0.51
Zn2+ -1.00 -1.35
K+ -2.14
C6H5NO2 -0.22
Note that copper(II) is reduced in two stages.

Instrumentation For most voltammetric and amperometric methods, the instrumentation

includes a working microelectrode, a reference electrode and an auxillary or
counter electrode, together with electronic equipment to control the voltage and
voltage sweep, plus a computer or recorder to collect data.
The earliest microelectrode used was the dropping mercury electrode
(DME), where pure mercury ﬂows through a ﬁne capillary, either due to gravi-
tational force, or by applied pressure. Drop times of a few seconds are usual.
This electrode has the advantages that:
● the surface area is small and is constantly refreshed so that products of
electrolysis do not accumulate;
● mercury has a high overpotential for hydrogen formation, which allows the
reduction of other species.
Other electrodes used are the static, or hanging mercury drop electrode,
where the drop is dislodged at a particular time and size, and solid micro-
electrodes, such as platinum and glassy carbon, which may be incorporated into
a rotating disc electrode.
Dissolved oxygen in the sample solution must be removed, since oxygen may
be reduced in two steps, giving waves that overlap with those of the sample.
O2 + 2H+ + 2e- = H2O2 E ⁄ = -0.05 V
1
2

+ -
H2O2 + 2H + 2e = H2O E ⁄ = -0.9 V
1
2

This is usually done by passing oxygen-free nitrogen through the sample

solution during the experiment.
Maxima on the waves are due to surface effects, and may be suppressed by
adding a small amount of surface-active agents, such as gelatin or Triton-X100.
Anodic stripping voltammetry is designed to measure trace amounts by
preconcentrating them onto a suitable electrode. The experiment has two stages:
(i) The sample is electrolyzed onto a hanging mercury drop, or a mercury film
deposited on a carbon electrode. By Faraday’s laws, (Topic C2), passing a
current of I amps for t seconds will produce a concentration cR in a mercury
film of thickness l, area A:
cR = It/nF lA
Because the current is limited by diffusion:
I = mnFD cB A
where m is a mass transfer coefficient, D the diffusion coefficient and cB the
bulk concentration.
102 Section C – Analytical reactions in solution

(ii) The reduced species (that is, the metal) is then oxidized out of the ﬁlm by
making the electrode increasingly anodic. A peak appears on the current-
potential plot, and the peak current can be shown to be:
Ip = k(cB n t)
where the constant k includes the diffusion and other constants, and n is the
rate of increase of the anodic potential. The peak potential at which an
active species is oxidized is characteristic of that species, and is close to its
half-wave potential.

Applications Polarographic techniques may be used in both qualitative and quantitative modes.
Since the half-wave potential is characteristic of the particular reaction that is
occurring at that potential, it is possible to identify the species involved. A
simple case is shown in Figure 3 where a mixture of metal ions was analyzed.
The two reduction waves for copper occur at –0.1 and –0.35 V, cadmium at
–0.69, nickel at -1.10 and zinc at –1.35 V. This illustrates an analysis that may
identify the species qualitatively and, by using a standard addition method, can
also determine the ions quantitatively.
Organic substances may be determined either in an aqueous or a nonaqueous
medium. For example, the concentration of nitrobenzene in commercial aniline
may be found by studying the reaction:
C6H5NO2 + 4H+ + 4e- = C6H5NHOH + H2O
The oxygen electrode is based on voltammetric principles and depends on
the diffusion and reduction of oxygen. It is also called the Clark sensor.
The cell has a lead anode and a silver cathode set close together in an alkaline
solution, often 1M KOH. At the anode, the reaction is
Pb(s) + 4OH- (aq) = PbO22- (aq) + 2H2O + 2e-
The silver cathode is inert, unless oxygen or another reducible species can
diffuse to it. A semipermeable membrane through which only oxygen can
diffuse surrounds the electrodes, and then the reduction reaction takes place.
O2(aq) + 2H2O + 4e- = 4OH-
Current, I

0 0.3 0.6 0.9 1.2 1.5

E (applied) (V)

Fig. 3. Polarogram of Cu2+, Cd2+, Ni2+ and Zn2+ ions at ~10-4 M.

C9 – Voltammetry and amperometry 103

Since the current depends on the diffusion of the oxygen to the electrode from
the external solution, and this diffusion is proportional to the concentration of
oxygen in the external solution, this electrode may be used to measure dissolved
oxygen.
Amperometric titrations are used to determine substances by measuring the
limiting diffusion current of a species as a function of the volume of a reagent
added to react with that species. Since Id is proportional to the concentration, it
will decrease as a species is used up, or increase as the excess of a species
becomes greater. For example, for the determination of Pb2+ with Cr2O72-:
2Pb2+ + Cr2O72- + H2O = 2PbCrO4 + 2H+
At an applied potential of 0.0 V, and at pH4, dichromate is reduced, but Pb2+ is
not, giving the graph shown in Figure 4.
Applications of anodic stripping voltammetry, are chieﬂy for the determina-
tion of trace amounts of amalgam-forming metals (Fig. 5), while cathodic strip-
ping voltammetry is used for determining species that form insoluble salts with
mercury. The preconcentration stage allows determination in the concentration
range 10-6 to10-8 M.

8
Current I (µA)

0
0 5 10 15 20 25 30 35 40
Volume dichromate added (ml)

Fig. 4. Amperometric titration of Pb(II) with dichromate at pH 4 and 0.0 V.

Current I (µA)

Pb
Cd

0 –0.2 –0.4 –0.6

E/V
Fig. 5. Anodic stripping voltammogram for 7.5 ppb Cu(II), 2.6 ppb Pb(II) and trace amounts
of Cd(II).
Section C – Analytical reactions in solution

C10 C ONDUCTIMETRY

Key Notes
Movement of ions Ions in solution or in molten ionic solids will move when an electric ﬁeld
is applied. The speed of movement will depend on the size and charge of
the ion.

Conductance In ionic solutions, the ions carry the current, but Ohm’s law still applies.
Increasing the number of ions increases the conductance.

Conductivity Since analysts are concerned with concentrations, it is preferable to

compare molar conductivities, which depend on the chacteristics of the
ions.

Conductimetric If the number or nature of ions present in a cell change, then the
titrations conductance will change. This is useful for many types of titration,
including some in nonaqueous media.

Table 1. Selected vales of mobility and ionic radius

Ion 108 ui/m2 s-1 V-1 ri /nm
+
Na 5.2 0.18
K+ 7.6 0.13
F- 5.5 0.17
Cl- 7.9 0.12
Br- 8.1 0.12
C10 – Conductimetry 105

Similar rules apply to anions. Also, potassium ions and chloride ions have very
similar mobilities. This is useful when ‘salt bridges’ are needed (see Topic C2).
If ions are present in other solvents, such as liquid ammonia, they will
conduct in a similar way, but their mobilities will be different.

Conductance Ohm’s law still applies to relate the current, I, to the applied voltage, E, and the
resistance, R, or to the conductance, G = 1/R.
I = E/R = EG
G is related to the concentration of ions and has units of ohms-1 or siemens (W-1
or S).
In order to measure the conductance of a solution, a conductance cell is used.
It has two electrodes, often platinum, coated with platinum black, separated by
a ﬁxed distance. Connecting these to a conductance meter, or more simply to a
Wheatstone bridge, the conductance may be measured directly. In order to
prevent electrolysis (see Topics C2, C9) taking place, which would change the
concentrations in the solution, the bridge uses alternating current. Plastic and
ﬂow-through conductance cells are also available
In any conductor, the conductance depends on the cell dimensions and on the
characteristics of the conducting medium. For a cell with electrodes of area A,
set a distance l apart:
G = k/(l/A)
where k is called the conductivity of the solution with units of W-1 m-1 or S m-1.
So that the results from different cells and solutions may be compared, the cell
constant, K = (l/A) is measured using a solution of known conductivity, often
0.1 M aqueous potassium chloride, which has a conductivity k =1.288 W-1 m-1
at 25∞C.

Conductivity Although it is often possible to work directly with conductance, it is useful to

know the conductivity and molar conductivity when results are to be compared.
From the equations above:
k = G(l/A)
Since the analyst is concerned with molar concentrations, and the conductivity
depends mostly on the amounts of ions present, the molar conductivity Lm is
deﬁned by:
Lm = k/c
where Lm has units of W m mol-1 provided that the concentration c is
-1 2

expressed in mol m-3 (= 1000 M).

Example
When immersed in a 0.1 M aqueous solution of KCl at 25∞C, a conductance cell
had G = 8.59 ¥ 10-3 W-1. For 0.1 M HCl, using the same cell, the conductance was
measured as 2.57 ¥ 10-2 W-1. Calculate the conductivity and molar conductivity
of the HCl.
For the KCl calibration:
k = G(l/A) = 8.59 ¥ 10-3 ¥ (l/A) = 1.288 W-1m-1
Thus (l/A) = 149.9 m-1 and for HCl, k = 3.85 W-1 m-1
106 Section C – Analytical reactions in solution

Lm = k/c = 3.85/100 = 0.0385 W-1 m2 mol-1

For strong electrolytes, it must be noted that, because of ionic interaction and
the effects of the ionic atmosphere, the molar conductivity decreases as the
concentration increases, according to the Onsager equation:
Lm = Lom - K c
where Lom is the molar conductivity at inﬁnite dilution, and K is a constant,
dependent on temperature and the solute.
For weak electrolytes, the ionic concentration depends on the dissociation
constant, and since the relationship of the dissociation constant, K to the degree
of ionization a is, for a weak acid, approximately:
K = ca2/(1 - a)
The molar conductivity of a weak electrolyte may be written:
Lm = aLom
These equations show that, while Lm for a strong electrolyte decreases almost
linearly with the (c)1/2, for weak electrolyes such as ethanoic acid, it decreases
approximately as (1/c)1/2. More details will be found in textbooks of physical
chemistry.
Direct measurement of concentration is possible. The presence of sodium
chloride, or other ionic species will increase conductance, and by knowing
the molar conductivity, the concentration may be found by measuring the
conductivity of the solution.
When very pure water is required for the preparation of pure organics, or the
manufacture of semiconductor components, conductance measurements are
most useful. Provided dissolved CO2 and other intrinsic factors are eliminated,
extremely pure water should have a conductivity of around 150 mS m-1 (or
1.5 mS cm-1) at 25∞C. In order to comply with regulations for some work, it
must be less than 200 mS m-1.

Conductimetric Whenever titrations of ionic solutions are carried out, the number and nature of
titrations the ions change throughout the entire titration. If a strong base, say NaOH, is
titrated with a strong acid, HCl, the reaction
Na+ + OH- + H+ + Cl- = Na+ + Cl- + H2O
will ﬁrst of all remove the OH- as nonconducting water, so the conductance will
decrease until the end point. Then excess HCl will increase the conductance.
This gives a ‘V’ shaped graph of conductance versus titer. If a mixture of acids is
used, they will be titrated in the order of their strength. Figure 1(a) shows the
conductimetric titration of a mixture of HCl, boric acid and ammonium chloride
using NaOH.
For precipitation reactions:
Ag+ + NO3- + Na+ + Cl- = AgCl(s) + Na+ + NO3-
The sodium and silver ions have similar conductivities, so adding NaCl solution
does not change the conductance much until the end point is reached. Then the
conductance rises with excess NaCl. This is shown in Figure 1(b). The titrant is
often about ten-times more concentrated than the sample solution, to avoid
dilution errors.
C10 – Conductimetry 107

(a) (b)
1.4 4

12 3.5
1000 x conductance

3
1

Conductance
2.5
0.8
2
0.6
1.5
0.4
1
0.2 HCl HB NH4Cl 0.5

0 0
0 0.5 1 1.5 2 0 1 2 3 4 5
Volume added (cm3) Volume added (cm3)

Fig. 1. (a) Conductimetric titration of mixed acids with 1.0 M NaOH. (b) Conductimetric titration of 0.1 M AgNO3 with
1.0 M KCl.

Conductimetric titrations may be carried out in nonaqueous media. For

example, phenols, which are very weak acids in water, have enhanced acidity
when liquid ammonia is used as the solvent, and may be titrated conductimetri-
cally with an ammoniacal solution of KOH.
Section D – Separation techniques

D1 S OLVENT AND SOLID - PHASE

EXTRACTION

Key Notes

Extraction techniques Solvent and solid-phase extraction are two techniques for separating
mixtures of substances, either by selective transfer between two
immiscible liquid phases or between a liquid and a solid phase.

Extraction efficiency Extraction efficiency is defined as the fraction or percentage of a

and selectivity substance that can be extracted in one or more steps. Selectivity is the
degree to which a substance can be separated from others in a mixture.

Solvent extraction Procedures are based on the extraction of nonpolar, uncharged species
from an aqueous solution into an immiscible organic solvent, or the
extraction of polar or ionized species into an aqueous solution from an
organic solvent.

Solid-phase sorbents Sorbents are particulate materials such as silica, chemically-modiﬁed

silica, alumina and organic resins that can interact with and retain
substances from solutions. Retained substances can be subsequently
released back into a suitable solvent under controlled conditions.

Solid-phase Sample solutions are passed through the sorbent under conditions where
extraction either the analyte(s) are retained and matrix components washed through
or the reverse. Retained analytes are removed with an alternative solvent
before completing the analysis.

Extraction The purpose of an extraction technique is physically to separate components of

techniques a mixture (solutes) by exploiting differences in their relative solubilities in two
immiscible liquids or between their afﬁnities for a solid sorbent. Substances
reach an equilibrium distribution through intimate contact between the two
phases, which are then physically separated to enable the species in either phase
to be recovered for completion of the analysis. An equilibrium distribution of
the solutes between the two phases is established by dissolving the sample in a
suitable solvent, then shaking the solution with a second, immiscible, solvent or
by passing it through a sorbent bed or disk. Where the equilibrium distributions
of two solutes differ, a separation is possible. The principal factors that deter-
mine how a solute will distribute between two phases are its polarity and the
polarities of each phase. Degree of ionization, hydrogen bonding and other
electrostatic interactions also play a part.
Most solvent-extraction procedures involve the extraction of solutes from an
aqueous solution into a nonpolar or slightly polar organic solvent, such as
hexane, methylbenzene or trichloromethane, although the reverse is also
110 Section D – Separation techniques

possible. Readily extractable solutes, therefore, include covalent neutral mole-

cules with few polar and no ionized substituents; polar, ionized or ionic species
will remain in the aqueous phase. A wide range of solid sorbents with different
polarities are used in solid phase extraction making this a very versatile tech-
nique, but the factors affecting the distribution of solutes are essentially the
same as in solvent extraction. Two examples illustrate these points.

Example 1
If an aqueous solution containing iodine and sodium chloride is shaken with a
hydrocarbon or chlorinated hydrocarbon solvent, the iodine, being a covalent
molecule, will be extracted largely into the organic phase, whilst the completely
ionized and hydrated sodium chloride will remain in the aqueous phase.

Example 2
If an aqueous solution containing a mixture of weakly polar vitamins or drugs
and highly polar sugars is passed through a hydrocarbon-modiﬁed silica
sorbent which has a nonpolar surface, the vitamins or drugs will be retained on
the surface of the sorbent whilst the sugars pass through.

Solvent extraction is governed by the Nernst Distribution or Partition Law.

This states that at constant temperature and pressure, a solute, S, will always be
distributed in the same proportions between two particular immiscible solvents.
The ratio of the equilibrium concentrations (strictly their activities (Topic C1)) in
the two phases deﬁnes a distribution or partition coefﬁcient, KD, given by the
expression
[S]org
KD =
[S]aq (1)

where [] denotes concentrations (activities) of the distributing solute species, S,

and the value of KD is independent of the total solute concentration. In practice,
the solute often exists in different chemical forms due to dissociation (ioniza-
tion), protonation, complexation or polymerization, so a more practically useful
expression which deﬁnes a distribution or partition ratio, D, is
(CS)org
D=
(CS)aq (2)

where (CS) represents the total concentration of all forms of the distributing
solute S in each phase. If no interactions involving S occur in either phase, then
D and KD would be identical. However, the value of D is determined by the
experimental conditions and it can be adjusted over a wide range to suit the
requirements of the analytical procedure.
A distribution ratio can also be deﬁned for solid phase extraction, i.e.
(CS)sorb
D(CS)liq (3)

where the numerator represents the solute concentration in the solid sorbent
and the denominator the solute concentration in the liquid phase.
Solutes with large values of D (e.g. 104 or more) will be essentially quantita-
tively extracted into the organic phase or retained by the sorbent, although the
nature of an equilibrium process means that 100% extraction or retention can
never be achieved.
D1 – Solvent and solid-phase extraction 111

Extraction The efﬁciency of an extraction depends on the value of the distribution ratio, D.
efﬁciency and For solvent extraction, it also depends on the relative volumes of the two liquid
selectivity phases and for solid-phase extraction on the surface area of the sorbent. With
solvent extraction, the percentage of a solute extracted, E, is given by the expres-
sion
100D
E (4)
[ (Vaq/Vorg)]
D +

where Vaq and Vorg are the volumes of the aqueous and organic phases, respec-
tively.
For solutes with small values of D, multiple extractions will improve the
overall efﬁciency, and an alternative expression enables this to calculated
n
(Caq)n = Caq[Vaq/(DVorg Vaq)] (5)

where Caq and (Caq)n are the amounts of solute in the aqueous phase initially and
remaining after n extractions, respectively.
The following example demonstrates the use of these formulae.

Example 1
A water sample contains 10 mg each of a halogenated pesticide and an ionic
herbicide which are to be separated by extraction of the pesticide into methyl-
benzene. Given that the pesticide distribution ratio, D, for methylbenzene/
water is 50, calculate the extraction efﬁciency for
(i) one extraction of 20 cm3 of water with 10 cm3 of methylbenzene
(ii) one extraction of 20 cm3 of water with 30 cm3 of methylbenzene
(iii) three extractions of the same 20 cm3 of water with 10 cm3 portions of
methylbenzene (30 cm3 in total)
(i) Substitution of the values for D, Vaq and Vorg in equation (4) gives
100 ¥ 50
E
[50 (20/10)] = 96.15%
(ii) Substitution of the values for D, Vaq and Vorg in equation (4) gives
100 ¥ 50
E
[50 (20/30)] = 98.68%
(iii) Substitution of the values for D, Vaq, Vorg, Caq and (Caq)n in equation (5) gives

(Caq)3 = 10 ¥ [20/((50 ¥ 10) + 20)]3

(Caq)3 = 10 ¥ [0.03846]3 = 5.6896 ¥ 10-4 mg
E = 99.99%
An extraction of 99.99%, as achieved in (iii), would be considered quantita-
tive, although the lower efficiencies obtained in (i) and (ii) might be acceptable
in the context of a defined analytical problem. It is clear that increasing the
volume of organic solvent, or extracting with the same volume divided into
several smaller portions, increases the overall efficiency of an extraction.
NB the ionic herbicide has a negligibly small distribution ratio, being very
polar and highly water soluble.
Selectivity in extraction procedures is the degree to which solutes in a
mixture can be separated by virtue of having different distribution ratios. For
112 Section D – Separation techniques

two solutes with distribution ratios D1 and D2, a separation or selectivity factor,
b, is defined as
b = D1/D2 where D1 >D2 (6)
4 5
Selectivity factors exceeding 10 or 10 (logb values exceeding four or five) are
necessary to achieve a quantitative separation of two solutes, as for most prac-
tical purposes, a separation would be considered complete if one solute could be
extracted with greater than 99% efficiency whilst extracting less than 1% of
another. The extraction of many solutes can be enhanced or supressed by
adjusting solution conditions, e.g. pH or the addition of complexing agents (vide
infra).

Solvent extraction Solvent extraction (SE) is used as a means of sample pre-treatment or clean-up
to separate analytes from matrix components that would interfere with their
detection or quantitation (Topic A4). It is also used to pre-concentrate analytes
present in samples at very low levels and which might otherwise be difficult or
impossible to detect or quantify. Most extractions are carried out batchwise in a
few minutes using separating funnels. However, the efficient extraction of
solutes with very small distribution ratios (<1) can be achieved only by continu-
ously exposing the sample solution to fresh solvent that is recycled by refluxing
in a specially designed apparatus.
Broad classes of organic compounds, such as acids and bases, can be sepa-
rated by pH control, and trace metal ions complexed with organic reagents can
be separated or concentrated prior to spectrometric analysis (Topic E10).

Extraction of organic acids and bases

Organic compounds with acidic or basic functionalities dissociate or protonate
in aqueous solutions according to the pH of the solution. Their extraction can,
therefore, be optimized by pH adjustments. The relation between pH and the
distribution ratio, D, of a weak acid can be derived in the following way:
A weak acid, HA, dissociates in water according to the equation
HA = H+ + A- (7)
The acid dissociation constant, Ka, is deﬁned as
[H]aq[A-]aq
Ka [HA]aq (8)

Only the undissociated form, [HA], can be extracted into a nonpolar or slightly
polar solvent such as diethyl ether, the distribution or partition coefﬁcient, KD,
being given by
[HA]ether
KD [HA]aq (9)

However, the distribution ratio takes account of both the dissociated and
undissociated forms of the acid in the aqueous phase and is given by
[HA]ether
D
([HA]aq [A-]aq) (10)

Re-arrangement of equation (8) and substitution for [A-]aq in equation (10) gives
D1 – Solvent and solid-phase extraction 113

[HA]ether
D
[HA]aq (1 Ka/[H]aq) (11)

and substituting KD for [HA]ether/[HA]aq from equation (9) gives

KD
D
(1 Ka[H+]) (12)

Equation (12) shows that at low pH, when the acid is undissociated, it is
extracted with the greatest efﬁciency as D @ KD, whereas as the pH is increased
the value of D decreases until at high pH the acid is completely dissociated into
the anion A-, and none will be extracted. This is shown graphically in Figure 1 as
pH versus E plots for two weak acids (curves 1 and 2) with different acid dis-
sociation constants (Ka or pKa values), curve 1 being for the stronger of the two
acids. For a weak organic base, such as an amine, protonation occurs at low pH
according to the equation
RNH2 + H+ = RNH3+ (13)
The relation between pH and the distribution ratio for a weak base (curve 3) is
therefore the opposite of a weak acid so that it is possible to separate acids from
bases in a mixture either by extracting the acids at low pH or the bases at high
pH. Separating mixtures of acids or mixtures of bases is possible only if their
dissociation or protonation constants differ by several pK units.

Extraction of metals
Metal ions in aqueous solutions are not themselves extractable into organic
solvents, but many can be complexed with a variety of organic reagents to form
extractable species. Some inorganic complex ions can be extracted as neutral
ionic aggregates by association with suitable ions of opposite charge (counter
ions). There are two principal metal extraction systems:
∑ Uncharged metal chelate complexes (ring structures that satisfy the
coordination requirements of the metal) with organic reagents
e.g. 8-hydroxyquinoline (oxine)
acetylacetone (AcAc)

100
1 2 3
80
Percent extracted (E)

0
0 1 2 3 4 5 6 7 8 9
pH

Fig. 1. Solvent extraction curves for two organic acids having different pKa values (curves 1
and 2) and a base (curve 3). Curve 1 represents the acid with the larger Ka (smaller pKa).
114 Section D – Separation techniques

1-(2-pyridylazo)-2-naphthol (PAN)
sodium diethyldithiocarbamate (NaDDTC)
(see also Table 1);
∑ Electrically neutral ion-association complexes
e.g. Cu(neocuproine)2+, NO3-
(C6H5CH2)3NH+, GaCl4-
[(C2H5)2O]3H+, FeCl4-
(see also Table 1).
The Nernst distribution law applies to metal complexes, but their distribution
ratios are determined by several interrelated equilibria. As in the case of organic
acids and bases, the efﬁciency of extraction of metal chelates is pH dependent,
and for some ion-association complexes, notably oxonium systems (hydrogen
ions solvated with ethers, esters or ketones), inorganic complex ions can be
extracted from concentrated solutions of mineral acids.
Reagents that form neutral metal chelate complexes (5- or 6-membered ring
structures) are weakly acidic and contain one or more additional co-ordinating
sites (O, N or S atoms). Protons are displaced according to the general equation

Mn+(H2O)x + nHR = MRn + nH+ + xH2O (14)

not extractable extractable
where HR is a weakly acidic and co-ordinating reagent (ligand), and the metal
ion Mn+ has a formal valency n. The removal of hydrogen ions is necessary to
drive the reaction to completion, and pH control, which is essential, is achieved
by buffering the aqueous solution. It is sometimes possible to improve the selec-
tivity of the procedure by adding an additional reagent, known as a masking
agent, that reacts preferentially with one of the metals to form a nonextractable
water-soluble complex. Typical masking agents include EDTA (ethylenedi-
aminetetraacetic acid), citrate, tartrate, ﬂuoride, cyanide and thiourea. Solvents
commonly used to extract metal chelate complexes include trichloromethane, 4-
methyl-pentan-2-one and methylbenzene.
Electrically neutral ion-association complexes consist of cationic (positively
charged) and anionic (negatively charged) species that form an overall neutral
aggregate extractable by an organic solvent. Either the cations or the anions

Table 1. A selection of reagents and extraction systems for the solvent extraction of
metals
Reagent Type of metal complex
8-Hydroxyquinoline (oxine) Neutral metal chelate complexes,
Di-alkyldithiocarbamates extractable into organic solvents. Intense
e.g. sodium diethyldithiocarbamate color of many facilitates colorimetric
(NaDDTC) determinations.
1,10-Phenanthroline (o-phen) Ion-association complexes. Metals as
2,9-Dimethyl-1,10-phenanthroline cationic or anionic chelated complexes
(neocuproine) extracted with suitable counter ion.
Ethylenediaminetetraacetic acid (EDTA)
Oxonium systems: Ion-pairs with anionic metal halide or
i.e. protons solvated with alkyl ethers, thiocyanate complexes. Chloride complexes
ketones, esters or alcohols extractable from strong HCl solutions.
D1 – Solvent and solid-phase extraction 115

should be bulky organic hydrophobic groups to provide high solubility in

nonpolar solvents. The metal ion can be a cationic or an anionic complex, and
can be an inorganic species such as FeCl4-, MnO4- or a chelated organic complex
such as Fe(1,10-phenanthroline)32+ or UO2(oxine)3-. Suitable counter-ions of
opposite charge to the metal-complex ions include (C4H9)4N+, ClO4-
(C6H5CH2)3NH+, [(C4H9O)3P=O]H+ and an oxonium ion such as [(C2H5)2O]3H+.
A list of metal chelate complexing agents and ion-association complexes is
given in Table 1. Most complexing agents react with a large number of metals
(up to 50 or more), but pH control, the use of masking agents and a variety of
ion-association systems can enable the selective extraction and separation of just
one or two metals to be accomplished.

Solid-phase These are generally either silica or chemically-modified silica similar to the
sorbents bonded phases used in high-performance liquid chromatography (Topics D6
and D7) but of larger particle size, typically 40–60 mm diameter. Solutes interact
with the surface of the sorbent through van der Waals forces, dipolar interac-
tions, H-bonding, ion-exchange and exclusion. The four chromatographic sorp-
tion mechanisms described in Topic D2 can be exploited depending on the
sorbent selected and the nature of the sample. Sorbents can be classified
according to the polarity of the surface. Hydrocarbon-modified silicas are
nonpolar, and therefore hydrophobic, but are capable of extracting a very wide
range of organic compounds from aqueous solutions. However, they do not
extract very polar compounds well, if at all, and these are best extracted by
unmodified silica, alumina or Florisil, all of which have a polar surface. Ionic
and ionizable solutes are readily retained by an ion-exchange mechanism using
cationic or anionic sorbents. Weak acids can be extracted from aqueous solu-
tions of high pH when they are ionized, and weak bases from aqueous solutions
of low pH when they are protonated. It should be noted that this is the opposite
way around compared to solvent extraction into non-polar solvents. However,
by suppressing ionization through pH control, extraction by hydrocarbon-
modified silica sorbents is possible. Sorbents of intermediate polarity, such as
cyanopropyl and aminopropyl modified silicas may have different selectivities
to nonpolar and polar sorbents. Some SPE sorbents are listed in Table 2 along
with the predominant interaction mechanism for each one.

Solid-phase Compared to solvent extraction, solid-phase extraction (SPE), is a relatively

extraction new technique, but it has rapidly become established as the prime means of
sample pre-treatment or the clean-up of dirty samples, i.e. those containing high
levels of matrix components such as salts, proteins, polymers, resins, tars etc. In
addition to being potential sources of interference with the detection and quan-
titation of analytes, their presence can be detrimental to the stability and perfor-
mance of columns and detectors when a chromatographic analysis is required.
The removal of interfering matrix components in general and the pre-concentra-
tion of trace and ultra-trace level analytes are other important uses of SPE which
is versatile, rapid and, unlike solvent extraction, requires only small volumes of
solvents, or none at all in the case of solid phase microextraction (vide infra).
Furthermore, SPE sorbents are cheap enough to be discarded after use thus
obviating the need for regeneration. The analysis of environmental, clinical,
biological and pharmaceutical samples have all beneﬁted from the rapid growth
in the use of SPE where it has largely replaced solvent extraction. Speciﬁc exam-
ples include the determination of pesticides and herbicides in polluted surface
116 Section D – Separation techniques

Table 2. Typical SPE sorbents and interaction mechanisms

Sorbent Polarity Interaction mechanisms
Silica SiO2 Polar Adsorption;
H-bonding
Florisil, MgSiO3 Polar H-bonding
alumina Al2O3 Polar H-bonding
Bonded phases (modiﬁed silica)
-C18H37 (C18 or ODS) Nonpolar Van der Waals interactions
-C8H17 (C8 or octyl) Nonpolar Van der Waals interactions
-C6H5 (phenyl) Nonpolar Van der Waals interactions
and p–p
interactions
-(CH2)3CN (cyanopropyl) Polar Polar interactions;
H-bonding
-(CH2)3NH2 (aminopropyl) Polar H-bonding
-(CH2)3C6H4SO3H Ionic Cation exchange
-(CH2)3N(CH3)3Cl Ionic Anion exchange
Chiral (cyclodextrin) Polar Adsorption;
H-bonding
dipolar interactions
steric effects
Styrene/divinyl benzene co-polymer Nonpolar Size exclusion

waters and soils, polycyclic aromatic hydrocarbons (PAHs) in drinking water,

polluted industrial and urban atmospheres, and drugs in biological fluids.
Sorbents are either packed into disposable cartridges the size of a syringe
barrel, fabricated into disks or incorporated into plastic pipette tips or well
plates. Most SPE is carried out using a small packed bed of sorbent (25–500 mg)
contained in a cartridge made from a polypropylene syringe barrel, the sorbent
being retained in position by polyethylene fritted disks. The sorbent generally
occupies only the lower half of the cartridge, leaving space above to accommo-
date several millilitres of the sample solution or washing and eluting solvents. A
typical cartridge procedure is illustrated in Figure 2 and consists of four distinct
steps:
● Sorbent conditioning. The cartridge is flushed through with the sample
solvent to wet the surface of the sorbent and to create the same pH and
solvent composition as those of the sample, thus avoiding undesirable
chemical changes when the sample is applied.
● Sample loading or retention. The sample solution is passed through the
cartridge with the object of either retaining the analytes of interest whilst the
matrix components pass through or retaining the matrix components whilst
the analytes pass through. In some procedures, the analyte(s) and one or
more of the matrix components are retained whilst the remainder of the
matrix components pass through.
● Rinsing. This is necessary to remove all those components not retained by
the sorbent during the retention step and which may remain trapped in the
interstitial solvent.
● Elution. This final step is to recover retained analytes, otherwise the matrix-
free solution and rinsings from the second and third steps are combined for
quantitative recovery of the analytes before completion of the analysis.
D1 – Solvent and solid-phase extraction 117

Conditioning Retention
Conditioning the sorbent prior to sample Adsorbed isolate
application ensures reproducible retention Undesired matrix constituents
of the compound of interest (the isolate) Other undesired matrix components

Rinse Elution
Rinse the columns to remove undesired Undesired components remain
matrix components Purified and concentrated isolate ready
for analysis
Fig. 2. Diagrammatic representation of a cartridge-based solid phase extraction procedure.

SPE can be semi- or fully-automated to increase sample throughput and to

improve both precision and accuracy. The degree of automation ranges from the
parallel off-line processing of batches of up to about 10 samples using a vacuum
manifold to provide suction, to on-line autosamplers, xyz liquid handlers and
robotic workstations.
Several alternative formats for SPE are available. These include
● Disks, which have relatively large cross-sectional areas compared to packed-
bed cartridges and thin sorbent layers (0.5–1 mm thick) containing about
15 mg of material. This reduced bed-mass results in low void volumes (as
little as 15 ml) thus minimizing solvent consumption in the rinsing and
elution steps, improving selectivity and facilitating high solvent ﬂow rates
when large volumes of sample are to be processed.
● Plastic pipette-tips incorporating small sorbent beds designed for processing
118 Section D – Separation techniques

very small volumes of sample and solvents rapidly, and having the advan-
tage of allowing flow in both directions if required.
● Well plates containing upwards of 96 individual miniature sample-
containers in a rectangular array and fitted with miniature SPE packed beds
or disks. Well plates are used in xyz liquid handlers for processing large
numbers of samples prior to the transfer of aliquots to analytical instruments,
particularly gas and liquid chromatographs and mass spectrometers.
● Solid-phase microextraction (SPME) is an important variation of SPE that
allows trace and ultra-trace levels of analytes in liquid or gaseous samples to
be concentrated. The sorbent is a thin layer of a polymeric substance such as
polydimethylsiloxane (PDMS) coated onto a fused-silica optical fibre about
1 cm long and attached to a modified microsyringe (Fig. 3). The fibre is
exposed to the sample and then inserted directly into the injection port of a
gas or liquid chromatograph to complete the analysis. An advantage of
SPME over SPE is the avoidance of solvents, but good precision for quantita-
tive determinations is more difficult to achieve, and automated systems are
only just being developed. SPME is finding particular use in water analysis,
the analysis of fragrances and volatiles in foodstuffs by headspace sampling
(Topic D5), the detection of drugs and their metabolites in urine, blood and
breath samples, and the monitoring of air quality in working environments.

Polymer coated Septum piercing

fused silica fiber needle Barrel Plunger

Stainless steel Viewing Z-slot

tube Adjustable needle window
guide/depth gauge

Fig. 3. Diagram of a solid phase microextraction device.

Section D – Separation techniques

D2 P RINCIPLES OF
CHROMATOGRAPHY

Key Notes

Chromatographic Chromatography is the process of separating the components of mixtures

separations (solutes) that are distributed between a stationary phase and a ﬂowing
mobile phase according to the rate at which they are transported through
the stationary phase.

Solute migration Solutes migrate through a stationary phase at rates determined by their
and retention relative afﬁnities for each phase, and are characterized by deﬁned
retention parameters.

Sorption processes During a chromatographic separation, solute species are continually

transferred back and forth between the mobile and stationary phases by
the process of sorption followed by desorption. Several mechanisms by
which this occurs give rise to different modes of chromatography.

Peak proﬁles and Individual solutes migrating through a stationary phase develop an
peak broadening approximately symmetrical band or peak proﬁle which becomes broader
as a function of time and distance travelled.

Peak asymmetry Peak proﬁles are fundamentally symmetrical but can become
asymmetrical, or skewed, as solutes migrate, due to changes in sorption
behavior.

Efﬁciency and The quality of a chromatographic separation is measured in terms of the

resolution degree of band broadening, or efﬁciency, measured for individual peaks,
and the degree of separation, or resolution, of adjacent peaks.

Qualitative and Chromatographic analysis can provide qualitative information in terms

quantitative analysis of characteristic retention parameters and quantitative information in
terms of peak areas or peak heights.

Chromatographic Chromatography was originally developed by the Russian botanist Michael

separations Tswett in 1903 for the separation of colored plant pigments by percolating a
petroleum ether extract through a glass column packed with powdered calcium
carbonate. It is now, in general, the most widely used separation technique in
analytical chemistry having developed into a number of related but quite
different forms that enable the components of complex mixtures of organic or
inorganic components to be separated and quantiﬁed. A chromatographic
120 Section D – Separation techniques

separation involves the placing of a sample onto a liquid or solid stationary

phase and passing a liquid or gaseous mobile phase through or over it, a
process known as elution. Sample components, or solutes, whose distribution
ratios (vide infra) between the two phases differ will migrate (be eluted) at
different rates, and this differential rate of migration will lead to their separa-
tion over a period of time and distance.
Chromatographic techniques can be classiﬁed according to whether the sepa-
ration takes place on a planar surface or in a column. They can be further sub-
divided into gas and liquid chromatography, and by the physical form, solid or
liquid, of the stationary phase and the nature of the interactions of solutes with
it, known as sorption mechanisms (vide infra). Table 1 lists the most important
forms of chromatography, each based on different combinations of stationary
and mobile phases and instrumental or other requirements.
Paper chromatography (PC) is simple and cheap but lacks the separating
power and versatility of thin-layer chromatography (TLC) which has largely
replaced it. Both require only inexpensive equipment and reagents, and, unlike
the various forms of column chromatography, comparisons can be made
between a number of samples and standards chromatographed simultaneously.
Gas (GC) and high performance liquid chromatography (HPLC) are comple-
mentary techniques best suited to the separation of volatile and nonvolatile
mixtures, respectively. Both these techniques are instrumentally-based and
computer-controlled, with sophisticated software packages and the ability to
separate very complex mixtures of up to 100 or more components. HPLC is
particularly versatile, having several alternative modes suited to different types
of solute. For example, ion-exchange (IEC) and ion chromatography (IC) are

Table 1. A classiﬁcation of the principal chromatographic techniques

Technique Stationary Mobile Format Principal
phase phase sorption
mechanism
Paper chromatography (PC) Paper (cellulose) Liquid Planar Partition (adsorption,
ion-exchange,
exclusion)
Thin-layer chromatography (TLC) Silica, cellulose, Liquid Planar Adsorption (partition,
ion-exchange resin, ion-exchange,
controlled porosity exclusion)
solid
Gas chromatography (GC)
Gas-liquid chromatography (GLC) Liquid Gas Column Partition
Gas-solid chromatography (GSC) Solid Gas Column Adsorption
Liquid chromatography (LC)
High-performance liquid Solid or bonded-phase Liquid Column Modiﬁed partition
chromatography (HPLC) (adsorption)
Size-exclusion chromatography (SEC) Controlled porosity Liquid Column Exclusion
solid
Ion-exchange chromatography (IEC) Ion-exchange resin or Liquid Column Ion-exchange
Ion chromatography (IC) bonded-phase
Chiral chromatography (CC) Solid chiral selector Liquid Column Selective adsorption
D2 – Principles of chromatography 121

modes that enable mixtures of either anionic or cationic solutes to be separated.

Size-exclusion (SEC) and chiral chromatography (CC) are two additional
modes used for separating mixtures of high relative molecular mass solutes and
enantiomers, respectively.
Earlier forms of liquid chromatography, used for relatively large scale separa-
tions and known as classical column chromatography, are based on large glass
columns through which the mobile phase ﬂows by gravity compared to the
pressurized systems used in HPLC.

Solute migration The rate of migration of a solute through a stationary phase is determined by its
and retention distribution ratio, D, which in turn is determined by its relative afﬁnity for the
two phases. In the context of chromatography, D is deﬁned as the ratio of the
total solute concentration, CS, in the stationary phase to that in the mobile phase,
CM, i.e.
CS
D
CM (1)

Thus, large values of D lead to slow solute migration, and small values of D
lead to rapid solute migration. Solutes are eluted in order of increasing distribu-
tion ratio. The larger the differences between the distribution ratios of the
solutes in a mixture, the more easily and quickly they can be separated. Because
the interaction of solutes with the stationary phase slows down their rate of
migration relative to the velocity of the mobile phase, the process is described as
retardation or retention.
● Column separations (GC and LC). For the separation of mixtures on
columns, solutes are characterized by the length of time they take to pass
through, i.e. their retention time, tR, or by a retention factor, k, that is directly
proportional to D. The retention time and the retention factor are related by
the expression
tR = tM(1 + k) (2)
where tM (sometimes written as t0 and known as the dead time) is the time
taken by a nonretained solute to pass through the column. A nonretained
solute migrates at the same velocity as the mobile phase, having a distribu-
tion ratio and retention factor of zero; hence tR = tM. Solutes whose D and k
values are greater than zero are proportionately retarded, having retention
times longer than tM, e.g.
if k = 1, tR = 2 tM
if k = 2, tR = 3 tM etc.
Chromatographic conditions are generally adjusted so that k values are
less than about 20, otherwise retention times become unacceptably long.
Practical values of k are easily calculated using a re-arranged equation (2)
(tR - tM)
k= tM (3)

For size-exclusion chromatography (Topic D7), the retention volume, VR,

is used to characterize solutes. This is the volume of mobile phase required to
elute the solute from the column. Retention times are directly proportional to
retention volumes at a constant ﬂow rate, so equation (2) can be re-written as
122 Section D – Separation techniques

冢冣
VS
VR VM 1 D (4)
VM
or VR VM DVs (5)
where D has been substituted for k using the relation

冢冣
VS
kD (6)
VM
and the volumes of the stationary and mobile phases in the column are VS
and VM, respectively.
Equation (5) is regarded as a fundamental equation of column chromato-
graphy as it relates the retention volume of a solute to its distribution ratio.
● Planar separations (PC and TLC). Separations are normally halted before the
mobile phase has travelled completely across the surface, and solutes are char-
acterized by the distance they have migrated relative to the leading edge of the
mobile phase (solvent front). A solute retardation factor, Rf, is deﬁned as
1
Rf
1k (7)

The maximum value of Rf is 1, which is observed for a solute having a distri-

bution ratio and retention factor of zero, and therefore migrating at the same
velocity as the mobile phase. Solutes whose D and k values are greater than
zero are proportionately retarded, the minimum value for Rf being zero,
which is observed when the solute spends all of the time in the stationary
phase and remains in its original position on the surface. Practical values of
Rf are evaluated from the ratio
Rf = (distance moved by solute) / (distance moved by the solvent front)

Sorption Sorption is the process whereby solute species are transferred from the mobile
processes to the stationary phase, desorption being the reverse process. These processes
occur continually throughout a chromatographic separation, and the system is
therefore described as being in a state of dynamic equilibrium. A solute is
repeatedly re-distributed between the phases as the mobile phase advances, in
an attempt to maintain an equilibrium corresponding to its distribution ratio, D.
There are four basic sorption mechanisms, and it is common for two or more
to be involved simultaneously in a particular mode of chromatography, viz:
adsorption; partition; ion-exchange; exclusion.
● Adsorption is a surface effect, not to be confused with absorption, which is
a bulk effect. Surface adsorption involves electrostatic interactions such as
hydrogen-bonding, dipole–dipole and dipole-induced dipole attractions.
Solute species compete with the mobile phase for a limited number of
polar sites on the surface of the adsorbent of which silica gel is the most
widely used. Its surface comprises Si-O-Si and Si-OH (silanol) groups, the
latter being slightly acidic as well as being polar, which readily form
hydrogen bonds with slightly-polar to very-polar solutes. Water in the
atmosphere can de-activate an adsorbent surface by itself being adsorbed,
thereby blocking adsorption sites. This can be overcome by drying the
adsorbent if a more active material is required, although reproducibility
may be difﬁcult to achieve unless ambient temperature and humidity are
carefully controlled. Some common adsorbents are listed in Table 2. Suitable
D2 – Principles of chromatography 123

Table 2. Adsorbents for chromatographic separations

(listed in order of decreasing polarity)
Alumina (most polar)
Charcoal
Silica gel
Molecular sieve
Magnesium silicate
Cellulose
Polymeric resins (least polar)
(styrene/divinyl benzene)

mobile phases for TLC and HPLC are to be found listed in Table 1, Topic
D6.
The more polar the solute, the more tenaciously it will be adsorbed onto
the surface of an adsorbent. Nonpolar solutes (e.g. saturated hydrocarbons)
have little or no affinity for polar adsorbents, whilst polarizable solutes (e.g.
unsaturated hydrocarbons) have weak affinities arising from dipole/induced
dipole interactions. Polar solutes, especially those capable of hydrogen-
bonding, are adsorbed strongly and require fairly polar mobile phases to
elute them. An approximate order of increasing polarity and therefore order
of elution (increasing distribution ratio) among classes of organic solutes is
alkanes<alkenes<aromatics<ethers<esters<ketones and
aldehydes<thiols<amines and amides<alcohols<phenols<acids
Adsorption-based chromatography is particularly useful for separating
mixtures of positional isomers such as 1,2-, 1,3- and 1,4-di-substituted
aromatic compounds with polar substituents, whereas members of a homolo-
gous series have similar polarities and cannot generally be separated at all.
Chiral chromatography, a mode of HPLC, depends on differences in the
adsorptive interactions of two or more enantiomers of a compound with a
chiral stationary phase. Gas solid chromatography (GSC) is an adsorption-
based mode of GC.
● Partition is a sorption process analogous to solvent extraction (Topic D1),
the liquid stationary phase being thinly coated or chemically bonded onto
an inert solid. Where the liquid is bonded to the supporting solid, it is
debateable as to whether it behaves as a liquid and whether the sorption
process should be described as modified partition, because adsorption may
also be involved. In true partition, solutes are distributed according to their
relative solubilities in the mobile and stationary phases, but the exact
mechanism for bonded phases is not clear. The use of bonded phases has
become widespread in all forms of chromatography, and a pure partition
mechanism probably occurs only in gas liquid chromatography (GLC)
where the stationary phase is not chemically bonded to the column wall
(Topic D4).
Bonded phases are described in more detail in Topics D4 and D6.
● Ion-exchange is a process whereby solute ions in the mobile phase can
exchange with counter-ions carrying the same charge and associated with
oppositely charged groups chemically bound to the stationary phase. The
stationary phase is a permeable polymeric solid, such as an insoluble organic
resin or a chemically modified silica, containing fixed charge groups and
124 Section D – Separation techniques

mobile counter-ions. Both cationic and anionic ion-exchangers are available,

the exchange processes being represented by the following equations
cation-exchange: nR-H+ + Xn+ = (R-)nXn+ + nH+
anion-exchange: nR+Cl- + Yn- = (R+)nYn- + nCl-
where R represents the polymeric resin or silica, and Xn+ and Yn- are solute
cations and anions respectively of valency n.
The factors affecting ion-exchange equilibria and selectivity are described
in Topic D7.
● Exclusion differs from the other sorption mechanisms in that no specific inter-
actions between solute species and the stationary phase are necessary or desir-
able. The stationary phase is a controlled-porosity silica or polymer gel with a
range of pore sizes, and solutes remain in the mobile phase throughout the
separation, merely diffusing through the porous structure to different extents
depending on their size and shape. Solutes whose size exceeds the diameter of
the largest pores are entirely excluded from the structure and migrate at the
same rate as the mobile phase. Solutes smaller than the diameter of the smallest
pores can diffuse throughout the structure and have the slowest rate of migra-
tion. Solutes of an intermediate size can diffuse through some pores but not
others, migrating at rates between those of the largest and smallest species.
Size-exclusion chromatography (SEC) is a mode of HPLC (Topic D7) and
is also a classical technique employing large columns of silica or polymeric
gel particles and gravity flow of the mobile phase. It is sometimes described
as gel permeation or gel filtration chromatography.

Peak profiles and During a chromatographic separation, individual solutes develop a symmetrical
band broadening or Gaussian concentration profile (Topic B2) in the direction of flow of the
mobile phase. The profiles, known as bands or peaks, gradually broaden and
often become asymmetrical as the solutes continue to migrate through the
stationary phase. The principal underlying reasons accounting for the peak
shapes and the observed broadening can be summarized as follows:
● continual sorption and desorption of a solute between a mobile and a
stationary phase inherently produces a Gaussian concentration profile which
broadens as the solute migrates further. (This can be demonstrated by a math-
ematical treatment of a solvent extraction procedure to separate mixtures,
developed in 1952, and known as Craig Countercurrent Distribution);
● solute species travel slightly different total distances through a particulate
stationary phase, causing concentration profiles to broaden symmetrically,
this being known as the multiple-path effect;
● solute species spread by diffusion in all directions when they are in the
mobile phase. Diffusion in both the direction of flow of the mobile phase and
directly counter to it (longitudinal or axial diffusion) contributes to the
symmetrical broadening of the peak profile;
● sorption and desorption, or mass transfer, between the stationary and
mobile phases, are not instantaneous processes, and are sometimes kineti-
cally slow. Because the mobile phase moves continuously, a true equilibrium
distribution of a solute is never established, and the concentration profile in
the stationary phase lags slightly behind that in the mobile phase causing
further peak broadening. Slow desorption can also result in the peak
becoming asymmetrical or skewed (vide infra);
D2 – Principles of chromatography 125

● variations in the distribution ratio of a solute with its total concentration also
leads to asymmetrical or skewed peaks (vide infra).
Figure 1 illustrates the symmetrical nature of a chromatographic peak and
symmetrical broadening. Figure 2 illustrates the mutiple-path, longitudinal
diffusion and mass transfer effects.

Flow

Mobile phase

Interface

Stationary phase

Equilibrium concentration
Actual concentration
Fig. 1. The symmetrical nature and broadening of a chromatographic peak.

(a) Stationary phase particles

Mobile phase

(b) Concentration
profile of band Forward and backward
diffusion in mobile phase
as band moves along

Mobile phase

Analyte band

(c) Mobile phase

Movement off SP Movement onto SP

Stationary Analyte attracted onto SP

phase (SP)

Fig. 2. Illustration of the three principal causes of band broadening: (a) multiple-path effect; (b) longitudinal diffusion
effect; (c) mass-transfer (non-equilibrium) effect. Reproduced from A. Braithwaite & F.J. Smith, Chromatographic
Methods, 5th edn, 1996, ﬁrst published by Blackie Academic & Professional.
126 Section D – Separation techniques

Peak asymmetry The concentration proﬁle of a migrating solute is fundamentally symmetrical

(Gaussian), only if the solute distribution ratio, D (deﬁned by eqn. (1)), remains
constant over the concentration range of the entire peak, as shown by a linear
sorption isotherm, which is a plot of the solute concentration in the stationary
phase, CS, against that in the mobile phase, CM (Fig. 3(a)). However, curved
isotherms, resulting from changes in the solute distribution ratio at higher
concentration levels, lead to two types of peak asymmetry, or skew, described
as tailing and fronting.
Both tailing and fronting are undesirable, as closely eluting peaks will be less
well separated and retention data less reproducible. Where either occur,
reducing the amount of solute chromatographed will generally improve the
peak shape, but slow de-sorption may still cause some tailing.

(a) Linear (b) Convex (c) Concave

CS CS CS

CM CM CM

Gaussian Tailing Fronting

t t t

Fig. 3. Sorption isotherms and the resulting peak proﬁles. (a) Linear isotherm, symmetrical
peak. (b) Curvature towards the CM axis, tailing peak. (c) Curvature towards the CS axis,
fronting peak.

Efficiency and Two means of assessing the quality of a chromatographic separation are to
resolution measure the extent of band broadening of individual solute peaks (efficiency)
and the degree of separation of adjacent peaks (resolution).
For column chromatography, a plate number, N (based on the theoretical
plate concept of distillation columns), is used as a measure of efficiency.
Assuming a Gaussian peak profile, N is defined in terms of the solute retention
time, tR, and the peak width as given by the standard deviation, st (Fig. 4), i.e.

冢冣
tR 2
N (8)
st
In practice, it is much easier to measure either the baseline width of a peak,
Wb, or the width at half height, Wh/2, and two alternative expressions derived
from equation (8) are

冢冣
tR 2
N 16 (9)
Wb
and
D2 – Principles of chromatography 127

Concentration axis
Peak height
Injection

Wh/2

One-half
peak height

Wb
(4s)
Time axis
Fig. 4. Measurement of chromatographic efﬁciency from a Gaussian peak.

冢冣
tR 2
N 5.54 (10)
Wh/2
Some laboratories favor the use of eqn. (9), but some favour eqn. (10) on the
grounds that peak width at half height can be measured with greater accuracy
than the base width. To make valid comparisons, the same formula should
always be used.
An alternative measure of efﬁciency, which is independent of the length of a
chromatographic column, is the plate height, H (or Height Equivalent of a
Theoretical Plate, HETP), and given by
L
H
N (11)

where L is the column length, usually expressed in millimetres or centimetres.

NB N is a dimensionless number and to ensure correct computation, the
units of tR and Wb or Wh/2 must be the same. Most chromatography data systems
have software to perform the calculations. Plate numbers for solutes separated
by GC and HPLC are of the order of several thousands to several hundreds of
thousands.
Columns giving very high plate numbers are capable of separating multicom-
ponent mixtures, but it is their resolving power, as measured by the resolution,
RS, that is of prime importance. This is deﬁned as the difference between the
retention times of two adjacent solute peaks, DtR, divided by their average base-
widths, (W1 + W2)/2 (Fig. 5)
2DtR
RS
(W1 W2) (12)

As in the case of the efficiency formulae, Gaussian peaks are assumed and all
measurements must be in the same units as RS is a dimensionless number.
Values of RS approaching or exceeding 1.5, which is defined as baseline resolu-
tion, are deemed satisfactory for most purposes.
The following is an example of the calculation of efficiency and resolution
using the above formulae:
128 Section D – Separation techniques

tR,1 tR, 2
DtR

Concentration axis

W1 W2

Time axis

Fig. 5. Measurement of the resolution of two adjacent Gaussian peaks.

Example. A gas chromatographic method for the separation of a mixture of

cyclohexane, t-butanol and benzene on a 10 m capillary column gave the follow-
ing data:

Parameter Cyclohexane t-Butanol Benzene

tR 3 m 20 s 3 m 30 s 3 m 45 s
Wb 8s 9s 11 s
Wh/2 4.6 s 5.1 s 6.2 s

Calculate (a) the plate numbers, using both plate number formulae, for each
solute, (b) the plate heights, and (c) the resolution between adjacent pairs of
solutes.

冢冣冢冣
tR 2 tR 2
(a) plate numbers N 16 N 5.54
Wb Wh/2

N 16冢冣 10 000 N 5.54冢冣 10 473

200 2 200 2
cyclohexane
8 4.6

冢 2109 冣 N 5.54冢冣 9393

210
2 2
t-butanol N 16 8711
5.1

冢 225
11 冣
N 5.54冢冣 7296
225
2 2
benzene N 16 6694
6.2

10 000
(b) plate heights H N
cyclohexane H = 1.0 mm H = 0.95 mm
t-butanol H = 1.15 mm H = 1.06 mm
benzene H = 1.49 mm H = 1.37 mm

2DtR
(c) resolution RS
(W1 W2)
2DtR
cyclohexane/t-butanol RS
(W1 W2) = 20/17 = 1.2
D2 – Principles of chromatography 129

2DtR
t-butanol/benzene RS
(W1 W2) = 30/20 = 1.5

Note that (i) The plate numbers are slightly higher when half-height peak
widths are used to calculate them due to peak tailing increasing
Wb, hence comparisons of efﬁciencies are valid only if the same
formula is used throughout.
(ii) The cyclohexane and t-butanol peaks are not fully resolved
(RS = 1.2), but the t-butanol and benzene peaks have baseline
resolution (RS = 1.5).

Qualitative and There are three approaches to qualitative chromatographic analysis, viz
quantitative
● Comparison of retention data for unknown solutes with corresponding data
analysis
for standards (known substances) obtained under identical conditions.
For planar chromatography (PC and TLC), retardation factors, Rf values,
for standards and unknowns are compared by chromatographing them
simultaneously so as to eliminate variations in laboratory materials and
conditions. For column separations, retention times, tR, or volumes, VR, are
compared by chromatographing standards and unknowns sequentially
under stable conditions with as little time between runs as possible.
● Spiking samples with known solutes.
For column separations where samples are known to contain certain
solutes, a comparison is made between two or more chromatograms run
under identical conditions. The first is of the original sample, and subsequent
ones are obtained after adding a spike of one of the known solutes. Any peak
in the original chromatogram that is of increased size in a subsequent one
can then be identified as the corresponding spiked solute. However,
unambiguous identifications may not be possible.
● Interfacing the chromatograph with a spectrometer (Section F).
For column separations, this provides spectral information for each sepa-
rated solute in addition to retention data. Spectra of unknown solutes can be
compared with those in computerized library databases or interpreted
manually, even when pure standards are not available.
Comparisons of retention data alone are not always reliable as many
substances can have identical chromatographic behavior. Two or more
comparisons made under different chromatographic conditions, e.g. different
stationary or mobile phases, reduce the chances of making an incorrect identifi-
cation.
For quantitative chromatographic analysis, in addition to ensuring stable
and reproducible conditions for sample preparation and chromatography,
further specific requirements must also be met viz:
● the analyte (solute) must be identified and completely separated from other
components in the chromatogram;
● standards of known purity must be available;
● a recognized calibration procedure must be used.
For planar chromatography, solute spot areas or densities can be measured
in situ, or the solute spots can be removed, dissolved and measurements made
by another analytical technique such as UV spectrometry (Topic E9).
For column separations, quantitation can be by peak area, peak height or
130 Section D – Separation techniques

peak height ¥ retention time. Integrated peak areas are directly proportional to
the amount of analyte chromatographed when working within the linear range
of the detector, and are the most reliable. They can be measured by computing-
integrators, by triangulation (1/2-base ¥ height of the triangle that approximates
to the Gaussian peak proﬁle) or by cutting out and weighing peaks drawn by a
chart recorder. Detector responses for an analyte are established by the prepara-
tion of a calibration graph using chromatographed standards, with or without
the addition of an internal standard, by standard addition or by internal
normalization. These calibration procedures are described in Topics A5 and B4.
Section D – Separation techniques

D3 T HIN - LAYER
CHROMATOGRAPHY

Key Notes

Principles and Thin-layer chromatography is a technique where the components of

procedures mixtures separate by differential migration through a planar bed of a
stationary phase, the mobile phase ﬂowing by virtue of capillary forces.
The solutes are detected in situ on the surface of the thin-layer plate by
visualizing reagents after the chromatography has been completed.

Stationary phase A variety of ﬁnely-divided particulate sorbents are used as thin-layer

stationary phases. These include silica-gel, cellulose powder, ion-
exchange resins, restricted pore-size materials, and chiral selectors.

Mobile phase Single solvents or blends of two or more solvents having the appropriate
overall polarity necessary to achieve the required separation are used as
mobile phases. They range from nonpolar hydrocarbons to polar
alcohols, water, and acidic or basic solvents.

Solute detection Methods of visualizing solutes include spraying the surface of the thin-
layer plate with a chromogenic reagent, or viewing it under a UV lamp if
the sorbent has been treated with a ﬂuorescent indicator.

Alternative TLC Alternative development procedures aimed at improving

procedures chromatographic performance have been introduced, and new stationary
phases are becoming available.

Applications of TLC Thin-layer chromatography is used primarily as a qualitative analytical

technique for the identiﬁcation of organic and inorganic solutes by
comparisons of samples with standards chromatographed
simultaneously. Quantitative analysis is possible but precision is
relatively poor.

Principles and Thin-layer chromatography is a form of planar chromatography similar to

procedures paper chromatography, but the stationary phase is a finely-divided sorbent
spread as a thin layer on a supporting flat plastic, aluminum or glass plate.
Solutes migrate through the stationary phase at rates determined by their distri-
bution ratios (Topic D2), those with the largest values moving the least, if at all,
whilst those with the smallest values moving with the advancing mobile phase,
or solvent front. A typical TLC procedure consists of the following steps:
● sufficient mobile phase to provide about a 0.5 cm depth of liquid is poured
into a development tank, or chamber, which is then covered and allowed to
132 Section D – Separation techniques

stand for several minutes to allow the atmosphere in the tank to become
saturated with the solvent vapor;
● small volumes of liquid samples and standards, or solutions, are spotted
onto the sorbent surface of a TLC plate along a line close to and parallel with
one edge (the origin). The plate is then positioned in the tank with this edge
in contact with the mobile phase and the cover replaced (Fig. 1(a));
● the mobile phase is drawn through the bed of sorbent from the edge of the
plate, principally by capillary action, and this development process is halted
shortly before the solvent front reaches the opposite side of the plate. Sample
components and standards migrate in parallel paths in the direction of flow
of the mobile phase, separating into discrete zones or spots;
● the plate is removed from the development tank, dried in a current of warm
air, and solute spots located by appropriate methods (vide infra);
● each solute is characterized by the distance migrated relative to the solvent
front, i.e. its Rf value, which will lie between 0 and 1 (Topic D2), and
unknowns are identified by comparisons with standards run simultaneously.
Figure 1(b) illustrates a developed and visualized TLC plate with Rf values
shown alongside. Note that the shapes of some spots have become slightly
elongated in the direction of flow of the mobile phase, which is an example of
tailing (Topic D2). This is caused by slow desorption as the solute migrates, or
saturation of adsorption sites by high concentrations of the solute leading to a
convex sorption isotherm, and is most likely to occur where adsorption is the
principal chromatographic sorption mechanism (Topic D2).

Stationary phase Stationary phases used in TLC are microparticulate sorbents with particle
diameters of between 10 and 30 mm. The smaller the mean particle size and the
narrower the size range, the better the chromatographic performance in terms of
band spreading (efﬁciency) and resolution (Topic D2). Thin-layer chromatog-
raphy plates are prepared by coating sorbents onto rectangular plastic,
aluminum or glass sheets in adherent uniform layers approximately 250 mm

(b)

Rf
1.00
(a)
Lid

TLC plate
Developing 0.59
coated with
absorbent tank

Origin line Sample spots 0.27

Mobile phase

Fig. 1. TLC plates (a) during and (b) after development and visualization; Rf values are 0
shown alongside. a, Reproduced from R.J. Hamilton & S. Hamilton, Thin-Layer
Chromatography, 1987. b, Reproduced from R.M. Smith, Gas and Liquid Chromatography
in Analytical Chemistry, 1988. © John Wiley & Sons Ltd. Reproduced with permission.
D3 – Thin-layer chromatography 133

thick. Commercially produced plates are available in several sizes between 5 cm

and 20 cm square and may incorporate an insoluble ﬂuorescent reagent to facili-
tate the detection of solute spots (vide infra). The most commonly used sorbents
are silica and powdered cellulose, and the corresponding sorption mechanisms
are adsorption and partition, respectively. Thin layers can also be made of
chemically-modiﬁed silicas, ion-exchange resins, exclusion gels and cyclo-
dextrins that display chiral selectivity. Some of these sorbents are similar to the
bonded phases used in HPLC which are discussed in Topics D6 and D7. Some
TLC sorbents are listed in Table 1.

Table 1. Stationary phases (sorbents) for thin-layer chromatography

Sorbents Chromatographic mechanism Typical applications
Silica gels Adsorption Amino acids, hydrocarbons, alkaloids,
vitamins
Hydrocarbon modiﬁed silicas Modiﬁed partition Nonpolar compounds
Cellulose powder Partition Amino acids, nucleotides, carbohydrates
Alumina Adsorption Hydrocarbons, alkaloids, food dyes,
lipids, metal ions
Kieselguhrs (diatomaceous earths) Partition Sugars, fatty acids
Ion-exchange celluloses Ion-exchange Nucleic acids, nucleotides, halide and
metal ions
Sephadex gels Exclusion Polymers, proteins, metal complexes
b-Cyclodextrins Stereo-adsorptive interactions Mixtures of enantiomers

Mobile phase The range of mobile phases used in TLC is extremely wide and they are often
selected empirically. Blends of two solvents are common because the solvent
strength, or eluting power, can be easily adjusted to optimize a separation by
altering solute distribution ratios. Some general guidelines in selecting and
optimizing the composition of a mobile phase are
● Solvents should be of the highest purity as TLC is a very sensitive analytical
technique;
● Mobile phase eluting power should be adjusted so that solute Rf values fall
between 0.2 and 0.8 so as to maximize resolution;
● For separations on silica gel and other polar adsorbents, the overall polarity
of the mobile phase determines solute migration rates and hence their Rf
values; small additions of a slightly polar solvent, such as diethyl ether, to a
nonpolar solvent, such as methylbenzene, will signiﬁcantly increase Rf
values;
● Polar and ionic solutes are best separated using a blend of a polar organic
solvent, such as n-butanol, with water; the addition of small amounts of
ethanoic acid or ammonia to the water increases the solubilities of basic and
acidic solutes, respectively.
A helpful guide to solvent strength in adsorption and partition-based separa-
tions is an eluotropic series, an example of which is given for HPLC in Table 1
of Topic D6. The solvents are listed in order of increasing solvent strength for
adsorption-based separations. The order for partition-based separations is
broadly similar.
134 Section D – Separation techniques

Solute detection Solutes separated by TLC remain on the surface of the plate after development.
As the majority of solutes are colorless, their spots must be located using a
chemical or physical means of visualization. Chemical methods include:
● Spraying the plate with a locating, or chromogenic reagent that reacts chemi-
cally with all solutes or those containing specific functional groups or struc-
tural features to give colored spots. The plates sometimes need to be warmed
to accelerate the chromogenic reaction and intensify the spots. Examples are
given in Table 2;
● Viewing the plate under a UV lamp set at an emission wavelength of 254 nm
or 370 nm to reveal the solutes as dark spots or bright fluorescent spots on a
uniform background fluorescence. Commercial plates can be purchased with
an insoluble fluorescent substance incorporated into the stationary phase to
provide the background fluorescence, or the plate can be sprayed with a
fluorescent reagent after development;
● Spraying the plate with concentrated sulphuric or nitric acid and heating to
oxidize and char organic solutes which are revealed as brown to black spots;
● Exposing the plate to iodine vapor in a sealed chamber when many organic
solutes develop dark brown colorations;
● Scanning across the surface of the plate with a densitometer, an instrument
that measures the intensity of radiation reflected from the surface when irra-
diated with a UV or visible lamp. Solutes that absorb radiation are recorded
as peaks on a recorder trace or VDU screen;
● Radiolabelled solutes can be detected by autoradiography (blackening of a
photographic film sensitive to X-rays), liquid scintillation counting on
scraped-off areas of the stationary phase, or monitoring the surface of the
plate with a Geiger-Müller tube.

Table 2. Some examples of TLC locating reagents

Method of detection Color of solute spots Application
General reagents
Phosphomolybdic acid + heat Dark blue Many organics
Conc. sulphuric acid + heat Charred brown-black All organics
Iodine vapor Brown Many organics
Selective reagents
Ninhydrin Pink to purple Amino acids and amines
2,4-Dinitrophenylhydrazone Orange/red Carbonyl compounds
Bromocresol green/ blue Yellow Organic acids
2,7-Fluorescein Yellow-green Most organics
Vanillin/ sulphuric acid Blue, green, pink Alcohols, ketones
Rhodamine-B Red ﬂuorescence Lipids
Anisaldehyde/antimony trichloride Various Steroids
Diphenylamine/zinc Various Pesticides

Alternative TLC Variations of the basic development procedure and new stationary phases
procedures aimed at improving resolution, sensitivity, speed, reproducibility and selectivity
have been developed from time to time. Two of the more signiﬁcant and useful
ones are:
● Two-dimensional TLC. This is a means of improving resolution in samples
where the component solutes have similar chemical characteristics and hence
D3 – Thin-layer chromatography 135

Rf values, e.g. amino acids. A single sample is spotted close to one corner of
the plate and, after development, there is a partial separation of the solutes
along one edge. The plate is dried, turned through 90∞ and developed a
second time, but using a different mobile phase composition for which the
solutes have different distribution ratios compared to those for the ﬁrst
mobile phase. This results in the partially-separated solutes separating
further because of changes to their migration rates. Visualization gives a 2-D
map or ﬁngerprint of the sample components. The procedure for the separa-
tion of 14 amino acids is shown diagrammatically in Figure 2.
● High-performance plates (HPTLC). These are coated with a thinner layer
(100 mm thick) of a 5 mm particle diameter stationary phase with a very
narrow range of sizes. Sensitivity and resolution are improved because solute
spots are compact, development is much faster, partly because smaller plates
can be used, and solvent consumption is reduced. HPTLC plates with silica
and bonded-phase silicas are commercially available.

Applications of TLC is applicable to a very wide range of mainly organic solutes. It is used
TLC primarily in the biochemical, pharmaceutical, clinical and forensic areas for
qualitative analysis by the comparison of Rf values of sample solutes with those
of standards run on the same plate. It is especially useful for checks on purity, to
monitor the course of reactions and production processes, and to characterize
complex materials such as natural products. The screening of samples for drugs

1st a b a b Solvent front b 2nd b Solvent front

solvent solvent

Sample Origin Turn 90 a Origin a

After 1st development After 2nd development

LEU

VAL Ph-ALA
a-ABA Solvent 1 n-butanol/acetic
GLU acid/water (12:3:5)
GLY THR ALA PRO-OH
PRO

SER LYS ARG

TAU
Sample
application

Solvent 2 Phenol/H2O

Fig. 2. Two-dimensional TLC of a mixture of 14 amino acids. Top panel, reproduced from R.J. Hamilton & S. Hamilton,
Thin-Layer Chromatography, 1987. © John Wiley & Sons Ltd. Reproduced with permission. Bottom panel, reproduced from
A. Braithwaite & F.J. Smith, Chromatographic Methods, 5th edn, 1996, ﬁrst published by Blackie Academic & Professional.
136 Section D – Separation techniques

in clinical and forensic studies, and testing for the presence or absence of
specific substances (limit tests) are additional applications. Tables 1 and 2 include
further examples.
Quantitative TLC by the measurement of spot areas or by computerized
scanning reflectance densitometry (Topic E8) is possible, but, unless HPTLC
plates are used, the relative precision attainable is generally only 5–10%. The
principal source of error is in applying sample spots to the plate, although auto-
mated systems can reduce this.
TLC has a number of advantages over GC and HPLC:
● the ability to run 10–20 or more samples simultaneously for immediate and
direct comparison with standards, which represents a considerable saving in
time;
● the basic technique is very cheap, versatile and quick;
● all solutes, including those that do not migrate from the origin, are detectable.
Disadvantages are the limited reproducibility of Rf values due primarly to
changes in the composition of the mobile phase during development of the
plate, and increasing spot diffusion (band broadening) because the flow rate of
the mobile phase slows as it travels across the plate.
Section D – Separation techniques

D4 G AS CHROMATOGRAPHY:
PRINCIPLES AND
INSTRUMENTATION

Key Notes
Principles Gas chromatography is a technique for the separation of volatile
components of mixtures by differential migration through a column
containing a liquid or solid stationary phase. Solutes are transported
through the column by a gaseous mobile phase and are detected as they
are eluted.

Mobile phase The mobile phase is an inert gas, generally nitrogen or helium, supplied
from a cylinder via pressure and ﬂow controls, and passing through
puriﬁcation cartridges before entering the column.

Sample injection Gaseous, liquid and solid samples are introduced into the ﬂowing mobile
phase at the top of the column through an injection port using a
microsyringe, valve or other device.

Column and Columns are either long, narrow, capillary tubes with the stationary
stationary phase phase coated onto the inside wall, or shorter, larger diameter tubes
packed with a particulate stationary phase. Stationary phases are high-
boiling liquids, waxes or solid sorbents.

Temperature control The column is enclosed in a thermostatically-controlled oven that is

maintained at a steady temperature or programmed to increase
progressively during a separation.

Solute detection Solutes are detected in the mobile phase as they are eluted from the end
of the column. The detector generates an electrical signal that can be
ampliﬁed and presented in the form of a chromatogram of solute
concentration as a function of time.

Instrument control A dedicated microcomputer is an integral part of a modern gas

and data processing chromatograph. Software packages facilitate the control and monitoring
of instrumental parameters, and the display and processing of data.

Principles Gas chromatography (GC) is a separation technique where volatile, thermally

stable solutes migrate through a column containing a stationary phase at rates
dependent on their distribution ratios (Topic D2). These are inversely propor-
tional to their volatilities, which in turn are determined by their partial vapor
138 Section D – Separation techniques

pressures and hence their boiling points. Solutes are therefore generally eluted
in order of increasing boiling point, except where there are specific interactions
with the stationary phase. The gaseous mobile phase elutes the solutes from the
end of the column where they pass through a detector that responds to each
one. An elevated temperature, usually in the range 50–350∞C, is normally
employed to ensure that the solutes have adequate volatility and are therefore
eluted reasonably quickly.
There are two modes of gas chromatography:
∑ Gas-liquid chromatography (GLC), which employs a liquid stationary phase
in which solutes can dissolve, the sorption process being partition. Specific
interactions of solutes with the stationary phase may alter the order of
elution from that of increasing boiling points. GLC is by far and away the
more widely used mode of GC, the large number of alternative stationary
phases enabling many types of sample to be analyzed.
∑ Gas-solid chromatography (GSC) employs a solid, sometimes polymeric,
sorbent as the stationary phase, the sorption process being surface adsorp-
tion. GSC has limited specialist applications, being used mainly for
analyzing mixtures of gases or solvents with relatively low relative molecular
masses.
A schematic diagram of a gas chromatograph is shown in Figure 1. It consists of
five major components:

Microliter syringe
(or autosampler,
sample valve)
Sample
Gas flow control system

Carrier gas Detector

Septum Amplifier

Injector
Makeup flow

Capillary column

Split flow

Column oven

Chromatogram
Fig. 1. Schematic diagram of a capillary column gas chromatograph. Reproduced from D.W. Grant, Capillary Gas
Chromatography, 1996. © John Wiley & Sons Ltd. Reproduced with permission.
D4 – Gas chromatography: principles and instrumentation 139

● gas supply and controls;

● sample injection port;
● column housed in a thermostatically-controlled oven;
● detection and recording system;
● microcomputer with control and data processing software.
These are described in the following sections.

Mobile phase The mobile phase is known as the carrier-gas because its sole purpose is to
transport solutes through the column, thus not contributing to chromatographic
selectivity. It should be inert, non toxic, non flammable and preferably cheap.
Helium and nitrogen are invariably used routinely, the former with capillary
(open tubular) columns and the latter with packed columns (vide infra). Helium
gives better chromatographic efficiency (reduced band broadening) due to faster
mass transfer (Topic D2). The carrier gas must be purified by passing it through
suitable adsorbents so as to avoid undesirable chemical changes to sample
components and stationary phases, or adverse effects on detector performance.
The most common contaminants and the means of removing them are:
● Air or oxygen at levels above about 10 ppm, which can oxidize sample
components and liquid stationary phases, especially at high column tempera-
tures. These can be removed by a cartridge containing molecular sieve.
● Hydrocarbons, which affect detector performance by contamination or
producing a large and constant background signal. These can be removed by
a cartridge containing activated carbon.
● Water vapor, which can affect some solid and bonded-liquid stationary
phases, and the performance of some detectors. This can also be removed
with molecular sieve.
The carrier gas is supplied from a cylinder via a pressure-reducing valve at
10–45 psi (0.7–3 bar) which provides flow rates of between 1 and 50 cm3 min-1
depending on the type of column in use. A mass-flow controller ensures
constant flow rates regardless of back-pressure and temperature (the viscosity of
gases increases with temperature).

Sample injection Samples should preferably be injected into the flowing mobile phase rapidly so
as to occupy the smallest possible volume when vaporized. This ensures a
narrow initial sample band that maximizes column efficiency and resolution.
There are a number of methods of injection and designs of injection port
available, and the choice is determined by the type of column in use and the
nature of the sample. Small volumes of liquids or solutions (0.1–10 ml) are gener-
ally injected into a heated injection port, through which the carrier gas continu-
ously flows, from a calibrated microsyringe used to pierce a self-sealing
silicone-rubber septum. Gases are introduced via a gas sampling valve or gas
syringe, and solids with volatile components as solutions. Septa must be
replaced regularly to avoid leakage and can be a source of contamination by
previously injected samples or the bleeding of plasticizers into the gas stream,
especially when operating at very high temperatures. A separate septum-purge
gas stream vented to the atmosphere, or septumless valves can overcome the
latter problem. Capillary (open tubular) columns require specially designed
injection ports to prevent overloading them with samples, which can severely
impair efficiency and resolution.

Daniel C. Harris: Quantitative Chemical Analysis, 9th Ed.: Books and Software in Review
0% (2)
Daniel C. Harris: Quantitative Chemical Analysis, 9th Ed.: Books and Software in Review
3 pages
Asia Pacific Business Schools
100% (1)
Asia Pacific Business Schools
11 pages
Reviewing Data Empower
0% (1)
Reviewing Data Empower
48 pages
(De Gruyter Textbook) Vic Soffiantini - Analytical Chemistry - Principles and Practice-De Gruyter (2021)
100% (1)
(De Gruyter Textbook) Vic Soffiantini - Analytical Chemistry - Principles and Practice-De Gruyter (2021)
314 pages
Assignement 1 Question Pool
No ratings yet
Assignement 1 Question Pool
6 pages
Test of Hypotheses: Hypothesis
100% (2)
Test of Hypotheses: Hypothesis
31 pages
Analytical Chemistry Book
No ratings yet
Analytical Chemistry Book
353 pages
Analytical Chemistry
92% (24)
Analytical Chemistry
353 pages
Analytical Chemistry by D. Kealey and P. J. Haines
No ratings yet
Analytical Chemistry by D. Kealey and P. J. Haines
353 pages
Instant Notes in Analytical Chemistry
100% (1)
Instant Notes in Analytical Chemistry
353 pages
Analytical Chemistry
No ratings yet
Analytical Chemistry
295 pages
Analytical Chemistry by D Kealey and P J Haines
No ratings yet
Analytical Chemistry by D Kealey and P J Haines
353 pages
Download full (Ebook) BIOS Instant Notes in Analytical Chemistry by David Kealey, P J Haines ISBN 9780203645444, 9780203693292, 9781859961896, 0203645448, 1859961894, 0203693299 ebook all chapters
No ratings yet
Download full (Ebook) BIOS Instant Notes in Analytical Chemistry by David Kealey, P J Haines ISBN 9780203645444, 9780203693292, 9781859961896, 0203645448, 1859961894, 0203693299 ebook all chapters
81 pages
Introduction to pharmaceutical analytical chemistry Second Edition. Edition Bente Gammelgaard download
100% (1)
Introduction to pharmaceutical analytical chemistry Second Edition. Edition Bente Gammelgaard download
67 pages
Analytical Chemistry Principles and Practice (Sec 1)
No ratings yet
Analytical Chemistry Principles and Practice (Sec 1)
150 pages
Bioanalytical Chemistry 2nd Revised Edition Andreas Manz - Read the ebook online or download it to own the full content
No ratings yet
Bioanalytical Chemistry 2nd Revised Edition Andreas Manz - Read the ebook online or download it to own the full content
89 pages
Analytical Chemistry Lab Manual 2021
No ratings yet
Analytical Chemistry Lab Manual 2021
86 pages
Introduction to pharmaceutical analytical chemistry Second Edition. Edition Bente Gammelgaard pdf download
100% (2)
Introduction to pharmaceutical analytical chemistry Second Edition. Edition Bente Gammelgaard pdf download
61 pages
4.119 MSC Chem Analytical Sem IIIand IV
No ratings yet
4.119 MSC Chem Analytical Sem IIIand IV
25 pages
4.119 MSC Chem Analytical Sem IIIand IV
No ratings yet
4.119 MSC Chem Analytical Sem IIIand IV
25 pages
Analytical Chemistry Handbook
100% (3)
Analytical Chemistry Handbook
300 pages
Download Complete (Ebook) Bioanalytical Chemistry by Andreas Manz, Petra S Dittrich, Nicole Pamme, Dimitri Iossifidis ISBN 9781783266715, 1783266716 PDF for All Chapters
No ratings yet
Download Complete (Ebook) Bioanalytical Chemistry by Andreas Manz, Petra S Dittrich, Nicole Pamme, Dimitri Iossifidis ISBN 9781783266715, 1783266716 PDF for All Chapters
81 pages
Libros Departamento
No ratings yet
Libros Departamento
21 pages
Complete Download Basic Concepts of Analytical Chemistry 3rd Edition S.M. Khopkar PDF All Chapters
No ratings yet
Complete Download Basic Concepts of Analytical Chemistry 3rd Edition S.M. Khopkar PDF All Chapters
41 pages
The HPLC Solvent Guide 2nd Edition - Paul C. Sadek
No ratings yet
The HPLC Solvent Guide 2nd Edition - Paul C. Sadek
334 pages
The HPLC Solvent Guide 2nd Edition (Paul C. Sadek)
100% (1)
The HPLC Solvent Guide 2nd Edition (Paul C. Sadek)
334 pages
Download ebooks file Introduction to pharmaceutical analytical chemistry Second Edition. Edition Bente Gammelgaard all chapters
100% (5)
Download ebooks file Introduction to pharmaceutical analytical chemistry Second Edition. Edition Bente Gammelgaard all chapters
47 pages
Method Project
No ratings yet
Method Project
97 pages
Handbook of Spectroscopy
From Everand
Handbook of Spectroscopy
G¿nter Gauglitz
No ratings yet
Methods in Physical Chemistry
From Everand
Methods in Physical Chemistry
Rolf Schäfer
5/5 (1)
Comprehensive Chiroptical Spectroscopy: Instrumentation, Methodologies, and Theoretical Simulations
From Everand
Comprehensive Chiroptical Spectroscopy: Instrumentation, Methodologies, and Theoretical Simulations
Nina Berova
No ratings yet
Practical High-Performance Liquid Chromatography
From Everand
Practical High-Performance Liquid Chromatography
Veronika R. Meyer
No ratings yet
Handbook of Nanoscopy
From Everand
Handbook of Nanoscopy
Gustaaf van Tendeloo
No ratings yet
Handbook of Magnetic Resonance Spectroscopy In Vivo: MRS Theory, Practice and Applications
From Everand
Handbook of Magnetic Resonance Spectroscopy In Vivo: MRS Theory, Practice and Applications
Paul A. Bottomley
No ratings yet
Voltammetry for Sensing Applications
From Everand
Voltammetry for Sensing Applications
J.G. Manjunatha
No ratings yet
Mercury Cadmium Telluride: Growth, Properties and Applications
From Everand
Mercury Cadmium Telluride: Growth, Properties and Applications
Peter Capper
No ratings yet
Transverse Disciplines in Metrology: Proceedings of the 13th International Metrology Congress, 2007 - Lille, France
From Everand
Transverse Disciplines in Metrology: Proceedings of the 13th International Metrology Congress, 2007 - Lille, France
French College of Metrology
No ratings yet
Handbook of Measurement in Science and Engineering
From Everand
Handbook of Measurement in Science and Engineering
Myer Kutz
No ratings yet
Capillary Electrophoresis in Food Analysis
From Everand
Capillary Electrophoresis in Food Analysis
María Castro-Puyana
No ratings yet
Iridium(III) in Optoelectronic and Photonics Applications
From Everand
Iridium(III) in Optoelectronic and Photonics Applications
Eli Zysman-Colman
No ratings yet
Intelligent Technologies for Research and Engineering
From Everand
Intelligent Technologies for Research and Engineering
S. Kannadhasan
No ratings yet
Fiber Optics Installer (FOI) Certification Exam Guide
From Everand
Fiber Optics Installer (FOI) Certification Exam Guide
Bill Woodward
No ratings yet
Nanotechnology: A Quick Guide to Materials and Technologies
From Everand
Nanotechnology: A Quick Guide to Materials and Technologies
Divya Bajpai Tripathy
No ratings yet
Analytical Methods in Supramolecular Chemistry
From Everand
Analytical Methods in Supramolecular Chemistry
Christoph A. Schalley
No ratings yet
Introduction to Carbon Nanomaterials
From Everand
Introduction to Carbon Nanomaterials
PublishDrive
No ratings yet
Absorbed Dose Determination in External Beam Radiotherapy: An International Code of Practice for Dosimetry Based on Standards of Absorbed Dose To Water
From Everand
Absorbed Dose Determination in External Beam Radiotherapy: An International Code of Practice for Dosimetry Based on Standards of Absorbed Dose To Water
IAEA
No ratings yet
Slow Contact Opening Circuit Breakers
From Everand
Slow Contact Opening Circuit Breakers
Kesorn Pechrach
No ratings yet
Magnetic Resonance Imaging: Physical Principles and Sequence Design
From Everand
Magnetic Resonance Imaging: Physical Principles and Sequence Design
Robert W Brown
3/5 (1)
Economic Control of Quality of Manufactured Product
From Everand
Economic Control of Quality of Manufactured Product
Walter A. Shewhart
No ratings yet
Computational Inorganic and Bioinorganic Chemistry
From Everand
Computational Inorganic and Bioinorganic Chemistry
Edward I. Solomon
5/5 (1)
Characterization of Solid Materials and Heterogeneous Catalysts: From Structure to Surface Reactivity
From Everand
Characterization of Solid Materials and Heterogeneous Catalysts: From Structure to Surface Reactivity
Michel Che
No ratings yet
Two-dimensional X-ray Diffraction
From Everand
Two-dimensional X-ray Diffraction
Bob B. He
No ratings yet
Electrochromic Materials and Devices
From Everand
Electrochromic Materials and Devices
Roger J. Mortimer
No ratings yet
Handbook of Multiphase Polymer Systems
From Everand
Handbook of Multiphase Polymer Systems
Abderrahim Boudenne
No ratings yet
Nuclear and Radiochemistry: Fundamentals and Applications
From Everand
Nuclear and Radiochemistry: Fundamentals and Applications
Jens-Volker Kratz
3/5 (1)
Characterization of Polymer Blends: Miscibility, Morphology and Interfaces
From Everand
Characterization of Polymer Blends: Miscibility, Morphology and Interfaces
Sabu Thomas
No ratings yet
qnmr: the handbook
From Everand
qnmr: the handbook
Michael Bernstein
No ratings yet
Nanomaterials: Evolution and Advancement towards Therapeutic Drug Delivery (Part I)
From Everand
Nanomaterials: Evolution and Advancement towards Therapeutic Drug Delivery (Part I)
PublishDrive
No ratings yet
Mechanical Characterization of Materials and Wave Dispersion: Instrumentation and Experiment Interpretation
From Everand
Mechanical Characterization of Materials and Wave Dispersion: Instrumentation and Experiment Interpretation
Yvon Chevalier
No ratings yet
Advanced Carbon Materials and Technology
From Everand
Advanced Carbon Materials and Technology
Ashutosh Tiwari
No ratings yet
Quantum Information: From Foundations to Quantum Technology Applications
From Everand
Quantum Information: From Foundations to Quantum Technology Applications
Dagmar Bruss
No ratings yet
Chemical Engineering in the Pharmaceutical Industry, Active Pharmaceutical Ingredients
From Everand
Chemical Engineering in the Pharmaceutical Industry, Active Pharmaceutical Ingredients
David J. am Ende
No ratings yet
Graphene-based Carbocatalysis: Synthesis, Properties and Applications: Volume 1
From Everand
Graphene-based Carbocatalysis: Synthesis, Properties and Applications: Volume 1
Pinki Bala Punjabi
No ratings yet
Functional Supramolecular Architectures: For Organic Electronics and Nanotechnology
From Everand
Functional Supramolecular Architectures: For Organic Electronics and Nanotechnology
Paolo Samorì
No ratings yet
Mata30 Final 2012w
No ratings yet
Mata30 Final 2012w
3 pages
Susan Rose, Nigel Spinks, Ana Isabel Canhoto - Management Research - Applying The Principles of Business Research Methods-Routledge - Taylor & Francis Group (2024)
100% (1)
Susan Rose, Nigel Spinks, Ana Isabel Canhoto - Management Research - Applying The Principles of Business Research Methods-Routledge - Taylor & Francis Group (2024)
482 pages
Chapter 3 Analysis and Adjustment of Observations
100% (1)
Chapter 3 Analysis and Adjustment of Observations
67 pages
Kami Export - Manuel Cervantes - Alg 2 LT1a Day 1 - Algebra 1 Review Properties of Linear Functions
No ratings yet
Kami Export - Manuel Cervantes - Alg 2 LT1a Day 1 - Algebra 1 Review Properties of Linear Functions
5 pages
Stress and Stiffness Matrix
No ratings yet
Stress and Stiffness Matrix
6 pages
02b Theory of Equations
No ratings yet
02b Theory of Equations
4 pages
Mth603 Finalterm Subjective: Question No: 50 (Marks: 5)
No ratings yet
Mth603 Finalterm Subjective: Question No: 50 (Marks: 5)
43 pages
Unit II
No ratings yet
Unit II
37 pages
Instant ebooks textbook Artificial Intelligence in Value Creation Andrzej Wodecki download all chapters
100% (3)
Instant ebooks textbook Artificial Intelligence in Value Creation Andrzej Wodecki download all chapters
51 pages
CHAPTER 5 Exploratory Research Design: Qualitative Research
No ratings yet
CHAPTER 5 Exploratory Research Design: Qualitative Research
3 pages
Table: Joint Coordinates Joint Coordsys Coordtype Xorr Y Z Specialjt Globalx Globaly
No ratings yet
Table: Joint Coordinates Joint Coordsys Coordtype Xorr Y Z Specialjt Globalx Globaly
6 pages
Chomba Bupe - Quora
No ratings yet
Chomba Bupe - Quora
4 pages
E1251 Aug 3:0 Linear and Nonlinear Optimization: Instructor
No ratings yet
E1251 Aug 3:0 Linear and Nonlinear Optimization: Instructor
2 pages
Unit-5 Index Tracking Complete PDF
No ratings yet
Unit-5 Index Tracking Complete PDF
11 pages
COM3706 Jadie-Lee Moodly Exam Portfolio 03
No ratings yet
COM3706 Jadie-Lee Moodly Exam Portfolio 03
24 pages
Activity Network Diagram
No ratings yet
Activity Network Diagram
21 pages
Continuity and Differentiability MCQ JEE
No ratings yet
Continuity and Differentiability MCQ JEE
5 pages
Unit - II Image Transforms
No ratings yet
Unit - II Image Transforms
60 pages
Dr. Rabi N. Subudhi Professor, School of Management KIIT University Bhubaneswar
No ratings yet
Dr. Rabi N. Subudhi Professor, School of Management KIIT University Bhubaneswar
20 pages
Translation of Meteorological Metaphors On The Mat
No ratings yet
Translation of Meteorological Metaphors On The Mat
8 pages
Journal List
No ratings yet
Journal List
146 pages
Numerical Modeling in Acoustics
No ratings yet
Numerical Modeling in Acoustics
9 pages
The Graphical Interpretation of The Function Properties: Increasing, Decreasing, and Constant Functions
No ratings yet
The Graphical Interpretation of The Function Properties: Increasing, Decreasing, and Constant Functions
3 pages
20 PPM of Methyl Red Solution, Activated Carbon and Distilled Water
No ratings yet
20 PPM of Methyl Red Solution, Activated Carbon and Distilled Water
4 pages
In A Chi-Square Test With A P Value of 0.05 and A Degrees of Freedom Value of 2, You Should REJECT Your Null Hypothesis If
No ratings yet
In A Chi-Square Test With A P Value of 0.05 and A Degrees of Freedom Value of 2, You Should REJECT Your Null Hypothesis If
6 pages
3.1 Global-Descent-Based Error Backpropagation: W W Given by
No ratings yet
3.1 Global-Descent-Based Error Backpropagation: W W Given by
28 pages

Instant Notes in Analytical Chemistry(sec 1)

Uploaded by

Instant Notes in Analytical Chemistry(sec 1)

Uploaded by

11

The INSTANT NOTES series

The INSTANT NOTES Chemistry series

First published 2002 (ISBN 1 85996 189 4)

This edition published in the Taylor & Francis e-Library, 2005.

ISBN 0-203-64544-8 Master e-book ISBN

ISBN 0-203-68109-6 (Adobe eReader Format)

BIOS Scientiﬁc Publishers Ltd

Section A – The nature and scope of analytical chemistry 1

Section B − Assessment of data 21

Section C − Analytical reactions in solution 55

Section D − Separation techniques 109

Section E − Spectrometric techniques 189

Section F − Combined techniques 283

Section G − Thermal methods 305

Section H – Sensors, automation and computing 323

Further reading 337

AAS atomic absorption spectrometry ICP-MS ICP-mass spectrometry

A1 A NALYTICAL CHEMISTRY, ITS

Purpose The purpose of chemical analysis is to gather and interpret chemical

Scope and Quality control in manufacturing industries, the monitoring of clinical

Related topics Analytical problems and Computer control and data

Deﬁnition Analytical chemistry involves the application of a range of techniques and

ﬁnished products. Progress and research in many areas is dependent on estab-

Analytical problems Selecting or developing and validating appropriate methods of analysis

Related topics Analytical chemistry, its functions Automated procedures (H2)

Analytical The stages or steps in an overall analytical procedure can be summarized as

● Sample pre-treatment or conditioning. Conversion of the sample into a form

Analytical Chemical or physico-chemical processes that provide the basis for

Method validation A process whereby an analytical method is checked for reliability in

Related topic Quality in analytical laboratories (A6)

Analytical An analytical method consists of a detailed, stepwise list of instructions to be

Table 1. Analytical techniques and principal applications

Table 2. Spectrometric techniques and principal applications

Table 3. Separation techniques and principal applications

lists of chemicals and reagents to be used, laboratory apparatus and glassware,

● the degree of selectivity available − methods may be selective for a small

Step 1 Performance characteristics of detector

Step 2 Process repeated for mixed analyte

Step 3 Process repeated for analyte calibration

Step 4 Process repeated for analyte calibration

Step 5 Analysis of 'spiked' simulated sample

Step 6 Field trials in routine laboratory with

Fig. 1. Flow chart for method validation.

A4 S AMPLING AND SAMPLE

Sample Preliminary treatment of a sample is sometimes necessary before it is in a

Related topic Analytical problems and procedures (A2)

Representative The importance of obtaining a representative sample for analysis cannot be

measured, they may need to be physically separated before laboratory samples

The distribution of toxic heavy metals or organic compounds in a land rede-

● increases in temperature leading to the loss of volatile analytes, thermal or

In addition, containers may leak or allow contaminants to enter.

Conversely, sample solutions containing organic solvents and other organic

● drying at 100°C to 120°C to eliminate the effect of a variable moisture content;

Sample clean-up in relation to matrix interference and to protect special-

Sample A laboratory sample generally needs to be prepared for analytical measurement

Table 1. Some methods for sample decomposition and dissolution

Calibration Calibration or standardization is the process of establishing the response

Chemical standard A chemical standard is a material or substance of very high purity

Reference material A reference material is a material or substance, one or more properties of

Related topic Calibration and linear regression (B4)

Table 1. Some primary standards used in titrimetric analysis

Bureau of Analytical Standards. CRMs or SRMs are produced in various forms

Related topics Analytical techniques and Quality control and chemometrics

● provision of full documentation and traceability of results to recognized

Table 1. Accreditation organizations and their quality standards

Accumulated errors Where several different measurements are combined to compute an

Related topic Assessment of accuracy and precision (B2)

Determinate errors can be proportional to the size of sample taken for

Constant determinate errors are independent of sample size, and therefore

Relative error (%)

Fig. 3. Effect of a constant error on titers of differing magnitudes.

physical or physico-chemical measurement, but no one speciﬁc source can be

Fig. 4. The normal error or Gaussian distribution curve.

Relative standard Relative standard deviation (coefﬁcient of variation) is the standard

Overall precision An estimate of the overall precision of an analytical procedure can be

冦冤冥冤冥冧 = {(9 ¥ 0.3 + 9 ¥ 0.23 )/18}