Determination of Metals in Natural Waters, Sediments, and Soils

299 Pages·2015·2.495 MB·English

by  T. R. Crompton

Determination of Metals
in Natural Waters,
Sediments, and Soils
T. R. Crompton
Chemical Analysis Department
Water Authority, Leeds
Yorkshire, UK
AMSTERDAM • BOSTON • HEIDELBERG • LONDON • NEW YORK • OXFORD
PARIS • SAN DIEGO • SAN FRANCISCO • SINGAPORE • SYDNEY • TOKYO
Elsevier
Radarweg 29, PO Box 211, 1000 AE Amsterdam, Netherlands
The Boulevard, Langford Lane, Kidlington, Oxford OX5 1GB, UK
225 Wyman Street, Waltham, MA 02451, USA
Copyright © 2015 Elsevier Inc. All rights reserved.
No part of this publication may be reproduced or transmitted in any form or by
any means, electronic or mechanical, including photocopying, recording, or any
information storage and retrieval system, without permission in writing from the
publisher. Details on how to seek permission, further information about the Publisher’s
permissions policies and our arrangements with organizations such as the Copyright
Clearance Center and the Copyright Licensing Agency, can be found at our website:
www.elsevier.com/permissions.
This book and the individual contributions contained in it are protected under copyright
by the Publisher (other than as may be noted herein).
Notices
Knowledge and best practice in this field are constantly changing. As new research
and experience broaden our understanding, changes in research methods, professional
practices, or medical treatment may become necessary.
Practitioners and researchers must always rely on their own experience and knowledge
in evaluating and using any information, methods, compounds, or experiments
described herein. In using such information or methods they should be mindful of
their own safety and the safety of others, including parties for whom they have a
professional responsibility.
To the fullest extent of the law, neither the Publisher nor the authors, contributors, or
editors, assume any liability for any injury and/or damage to persons or property as a
matter of products liability, negligence or otherwise, or from any use or operation of
any methods, products, instructions, or ideas contained in the material herein.
ISBN: 978-0-12-802654-0
British Library Cataloguing in Publication Data
A catalogue record for this book is available from the British Library
Library of Congress Cataloging-in-Publication Data
A catalog record for this book is available from the Library of Congress
For Information on all Elsevier publications
visit our website at http://store.elsevier.com/
Preface
This book is concerned with a discussion of methods currently available in the
world literature up to early 2014 for the determination of metals, and m etalloids
in natural waters, sediments, and soils. The occurrence of metals, many of
which are toxic, can have profound effects on the ecosystem.
In the case of soils, the presence of deliberately added or adventitious
metallic compounds can cause contamination of the tissues of crops grown on
the land or animals feeding on the land and, consequently, can cause adverse
toxic effects on man, animals, birds, and insects. Also drainage of these
substances from the soil can cause pollution of adjacent streams rivers and
eventually the oceans. Some of the substances included in this category are
fertilizers, crop sprays, sheep dips, etc. In the case of river sediments, a major
input of metals is a consequence of industrial activity. Many industries dis-
charge metal-containing e ffluents to rivers or directly to sea or discharge of
such wastes by ship dumping.
Surface water drainage is another input. The presence of metallic com-
pounds in river sediments is due, in part, to man-made pollution and monitoring
the levels of these substances in the sediment and sediment cores provides an
indication of the time dependence of their concentration over large time spans.
Contamination of sediments is found not only in rivers but also in estuarine
and oceanic sediments and thus sediment analysis provides a means of tracking
metallic compounds from their source through the ecosystem.
Sediments have the property of absorbing such contaminants from waters
within their bulk (accumulation) and, indeed, it has been shown that the concen-
tration, for example, for some types of metallic contaminants in river sediments
can be up to one million times greater than it occurs in the surrounding water.
Most of the metallic elements in the periodic table have been found to
be present in natural waters, and soil sediments and sludges, some naturally
occurring and others as a result of industrial activity of one kind or another. To
date, insufficient attention has been given to the analysis of sediments and one
of the objects of this book is to draw attention of analysts and others concerned
to the methods available and their sensitivity and limitations.
The use of correct sampling and sample preservation procedures is manda-
tory in the analysis of solid materials and this is discussed fully in C hapters 1
and 2 which discuss aspects of sampling including sample homogeneity,
comminution and grinding of samples, sample digestion procedures. The appli-
cation of nondestructive methods of analysis to solid samples is also discussed.
xxi
xxii Preface
Chapters 3 to 5, respectively, deal with the determination of trace metal
concentrations in river waters, groundwaters, and aqueous precipitation.
Chapter 6 deals with the sampling of sediments and Chapters 7 and 8, respec-
tively, deal with the determination of “traces of metals in sediments and soils.”
Examination of solid samples for metals combines all the exciting features
of analytical chemistry. First, the analysis must be successful and in many cases
must be completed quickly. Often the nature of the substances to be analyzed
for is unknown, might occur at exceeding low concentrations and might, indeed,
be a complex mixture. To be successful in such an area requires analytical skills
of a high order and the availability of sophisticated instrumentation.
The work has been written with the interests of the following groups of
people in mind: management and scientists in all aspects of the water industry,
river management, fishery industries, sewage effluent treatment and disposal,
land drainage, and water supply; also management and scientists in all branches
of industry. It will also be of interest to agricultural chemists, agriculturalists
concerned with the ways in which metals used in crop or soil treatment perme-
ate through the ecosystem, the biologists and scientists involved in fish, plant,
insect, and plant life, and also to the medical profession, toxicologists and
public health workers and public analysts. Other groups or workers to whom
the work will be of interest include oceanographers, environmentalists and,
not least, members of the public who are concerned with the protection of our
environment.
Finally, it is hoped that the work will act as a spur to students of all subjects
mentioned and assist them in the challenge that awaits them in ensuring that
the pollution of the environment is controlled so as to ensure that into the new
millennium we are left with a worthwhile environment to protect.
Chapter 1
Metals in Natural Water Samples
Sampling Techniques
Chapter Outline
1.1 Introduction 1 1.3 F iltration of Water Samples for
1.2 Sampling Devices 2 Trace Metal Determination 5
References 7
1.1 INTRODUCTION
Heavy metals are among the most toxic and persistent pollutants in freshwa-
ter systems. Many research and monitoring efforts have been conducted to
determine sources, transport, and fate of these metals in the aquatic environ-
ment. However, studies have shown that contamination artifacts have seriously
compromised the reliability of many past and current analyses,1 and in some
cases, metals have been measured at 100 times true concentration.2 These
induced errors are of great concern, since artifact-free data are necessary to
detect trends and to identify factors that control the transport and fate of toxic
heavy metals. In addition, without accurate and reliable data, it is impossible
to accurately monitor the effect of costly regulations aimed at reducing metal
emissions. To avoid these problems, and to enhance the quality of trace metal
data, laboratories are putting substantial effort into improving protocols for
sample collection, handling, and analysis.3 This greater level of effort devoted
to clean methods is costly in both money and time.
The problems caused by contamination when measuring trace metals were
first brought to the attention of the scientific community by Patterson in his
investigation of stable isotopes in the 1960 and 1970s.4–6 Largely through his
influence, clean methods became part of the standard operating procedures used
by chemical oceanographers starting in the mid-1970s.5,6 Freshwater chemists
were slow to adopt these same techniques, with the notable exception of the
long series of investigations on lead cycling by Patterson and co-workers.7–11
With few exceptions,12–15 limnologist have begun to use clean techniques only
in the last 10 years. This was spurred, in part, by oceanographers who began to
study freshwater systems such as the Mississippi River,16–18 Great Lakes,19 and
Amazon River.20 The result of this activity has been to cast serious doubt on earlier
routine measurements. Thus, Flegal and Coale21 have questioned surveys of lead
Determination of Metals in Natural Waters, Sediments, and Soils.
http://dx.doi.org/10.1016/B978-0-12-802654-0.00001-5 1
Copyright © 2015 Elsevier Inc. All rights reserved.
2 Determination of Metals in Natural Waters, Sediments, and Soils
in surface waters,22,33 and Windom et al.2 disputed the reliability of the United
States Geological Survey (USGS) National Stream Quality Accounting Network.
Ahlers et al.4 and Benoit24 implied that most previously reported results may be
in error because of failure to follow appropriate clean protocols. A parallel can
be drawn with chemical oceanography, where virtually all uncensored (i.e., other
than nondetects) trace metal data from before about 1975 are considered invalid.
Common to all analytical methods is the need for correct sampling. It is still
the most critical stage with respect to risks to accuracy in aquatic trace metal
chemistry, owing to the potential introduction of contamination. Systematic
errors introduced here will make the whole analysis unreliable.
Surface-water samples are usually collected manually in pre-cleaned poly-
ethylene bottles (from a rubber or plastic boat) from the sea, lakes, and rivers.
Sample collection is performed in front of the bow of the boat, against the wind.
In the sea or in larger inland lakes, a sufficient distance (about 500 m) in an
appropriate wind direction has to be kept between the boat and the research
vessel to avoid contamination. The collection of surface water samples from the
vessel itself is impossible, considering the heavy metal contamination plume
surrounding each ship. Surface-water samples are usually taken at 0.3 to 1 m
depth, in order to be representative and to avoid interference by the air–water
interfacial layer in which organics and consequently bound heavy metals accu-
mulate. Usually, sample volumes between 0.5 and 2 L are collected. Substan-
tially larger volumes could not be handled in a sufficiently contamination-free
manner in the subsequent sample pre-treatment steps.
Reliable deep-water sampling is a special and demanding art. It usually has
to be done from the research vessel. Special devices and techniques have been
developed to provide reliable samples.
Samples for mercury analysis should preferably be taken in pre-cleaned glass
flasks. If, as required for the other ecotoxic heavy metals, polyethylene flasks are
commonly used for sampling, then an aliquot of the collected water sample for the
mercury determination has to be transferred as soon as possible into a glass bottle,
because mercury losses with time are to be expected in polyethylene bottles.
Luettich et al.25 have described an instrument system for remote measurement
of physical and chemical parameters in shallow water.
Martin et al.26 compared surface grab and cross-sectionally integrated
sampling and found that very similar concentrations of dissolved metals were
obtained by these procedures.
Cox and McLeod27 have discussed difficulties associated with preserving
the integrity of chromium in water samples.
1.2 SAMPLING DEVICES
The job of the analyst begins with taking the sample. The choice of sampling
gear can often determine the validity of the sample taken; if contamination is
introduced in the sampling process itself, no amount of care in the analysis can
save the results.
Metals in Natural Water Samples Chapter | 1 3
Sampling the subsurface waters present some not completely obvious
problems. For example, the material from which the sampler is constructed
must not add any metals or organic matter to the sample. To be completely
safe, then, the sampler should be constructed either of glass or of metal. All-
glass samplers have been used successfully at shallow depths; these samplers
are generally not commercially available.28,29 To avoid contamination from
material in the surface film, these samplers are often designed to be closed
while they are lowered through the surface, and then opened at the depth of
sampling. The pressure differential limits the depth of sampling to the upper
100 m; below this depth, implosion of the sampler becomes a problem.
Implosion at greater depths can be prevented either by strengthening the
container or by supplying pressure compensation. The first solution has been
applied in the Blumer sample.30 The glass container is actually a liner inside an
aluminum pressure housing; the evacuated sampler is lowered to the required
depth, where a rupture disc brakes, allowing the sampler to fill. Even with the
aluminum pressure casing, however, the sampler cannot be used below a few
1000 m without damage to the glass liner.
Another approach to the construction of glass sampling containers involves
equalization of pressure during the lowering of the sampler. Such a sampler has
been described by Bertoni and Melchiorri-Santolini.31 Gas pressure is supplied
by a standard diver’s gas cylinder, through an automatic delivery valve of the
type used by scuba divers. When the sampler is opened to the water, the pressur-
izing gas is allowed to flow out as the water flows in. The sampler in its original
form was designed for used in Lake Maggioŕe, where the maximum depth is
about 200 m; but in principle it can be built to operate at any depth.
Stainless steel samplers have been devised, largely to prevent organic
contamination. Some have been produced commercially. The Bodega-Bodman
sampler and the stainless steel Niskin bottle, formerly manufactured by G eneral
Oceanics, Inc, are examples. These bottles are both heavy and expensive.
The Bodega-Bodman bottle, designed to take very large samples, can only be
attached to the bottom of the sampling wire; therefore, the number of samples
taken at a single station is limited by the wire time available, and depth profiles
require a great deal of station time.
The limitations of the glass and stainless steel samplers have led many
workers to use the more readily available plastic samplers, sometimes with a
full knowledge of the risks and sometimes with the pious hope that the effects
resulting from the choice of sampler will be small compared with the amounts
of organic matter present.
Smith32 has described a device for sampling immediately above the sedi-
ment water interface of the ocean. The device consists of a nozzle supported
by a benthic sled, a hose and a centrifugal deck pump, and is operated from
a floating platform. Water immediately above the sediment surface is drawn
through the nozzle and pumped through the hose to the floating platform, where
samples are taken. The benthic sled is manipulated by means of a hand winch
and a hydrowire.
4 Determination of Metals in Natural Waters, Sediments, and Soils
Basu33 has described a device for collecting water samples at depth com-
prising a plastic cylinder (60 × 60 cm) attached to a light-alloy bracket and
shaft. The lower end of the cylinder terminates in an inverted cone and nozzle
(0.5 cm diameter). The upper end, fitted with a sealing ring, is closed by a
plastic ball (6.3 cm diameter) attached to the inside wall of the cylinder by an
elastic cable. The ball is held in the open position by one arm of a pivoted lever;
a wire attached to the bracket passes through a hole in the other arm of the lever.
Weights at the lower end of the cylinder keep it steady at the sampling depth.
The apparatus is lowered into position with both ends of the cylinder open; a
lead weight then slides down the wire, strikes the lever arm, and releases the
ball, which closes the top of the cylinder; and the sampler is carefully with-
drawn. A small orifice, closed by a thumb-screw, controls the discharge of
water from the sampler.
Bhagat et al.34 Evans and Edgar,35 and Noth36 reviewed sampling in the
aquatic environment, automated sampling, and sampling of frozen water,
respectively.
Gibs et al.37 used a multiport sampler with seven screened intervals to study
vertical variations in water chemistry.
Powell and Puls38 studied differences in ground water chemistry between the
casing and screened interval values of four wells. Tracer experiments were used
to study the differences in natural flushing between the casing and screened
interval volumes.
Benoliel39 has reviewed the storage and preservation of natural water
samples.
Salbu and Oughton40 have reviewed strategies for the sampling, fraction-
ation, and analysis of natural waters.
Droppo and Jaskot41 have studied the effect of river transport characteristics
on contaminant sampling.
Hall et al.42 have studied the effects of four different filter membranes on
concentrations of 28 dissolved elements in five different natural water matrices.
Benoit et al.43 have demonstrated that clean techniques are necessary for
reliable measurement of trace metals in freshwaters at ambient, though not
necessarily regulatory, concentrations. These workers compared conventional
sample-handling methods to clean techniques for 35 individual steps used in pro-
tocols for analysis of filtrate and filter-retained forms of silver, cadmium, copper,
and lead. Approximately two-thirds of all steps contributed statistically signifi-
cant amounts of contamination in the measurement of dissolved and particulate
cadmium, copper, and lead. Average contamination for a single contributing step
was 300%, 141%, and 200% for the three metals, respectively (where 100%
represents no added contamination). Relative copper contamination tended to
be lower, partly because real levels in water are higher for this metal. Contami-
nation generally was not a problem for silver when it was present in water at
higher than background levels. With that exception, it does not seem possible to
develop clean technique protocols, even when measuring trace metals in polluted
Metals in Natural Water Samples Chapter | 1 5
freshwaters where levels are moderately high. The expectation of Benoit et al.43
is that most other metals (e.g., zinc, chromium or nickel) will have contamina-
tion behavior that is similar to cadmium, copper, and lead, rather than the rare
metal silver.
Brite et al.23 have described a downhole groundwater sampler to reduce
bias and error due to sample handling and exposure while introducing minimal
distribution to natural flow conditions in the formation and well. This in situ
sealed (ISS) or “snap” sampling device includes removable/flab-ready sample
bottles, a sampler device to hold double end-opening sample bottles in an open
position, and a line for lowering the sampler system and triggering closure of
the bottles downhole.
Before deploying, each bottle is set open at both ends to allow flow-through
during installation and equilibration downhole. Bottles are triggered to close
downhole without well purging; the method is therefore passive or nonpurge.
The sample is retrieved in a sealed condition and remains unexposed until
analysis. Data from six field studies comparing ISS sampling with traditional
methods indicate ISS samples typically yield higher volatile organic compound
(VCO) concentrations; in one case, significant chemical-specific differentials
between sampling methods were discernible. For arsenic, filtered and unfiltered
purge results were negatively and positively biased, respectively, compared
to ISS results. Inorganic constituents showed parity with traditional methods.
Overall, the ISS is versatile, avoids low VOC recovery bias, and enhances
reproducibility while avoiding sampling complexity and purge water.
1.3 FILTRATION OF WATER SAMPLES FOR TRACE METAL
DETERMINATION
Filtration and centrifugation are the principal methods available for the
separation of dissolved and undissolved trace metal fractions in waters; in situ
dialysis has also been proposed as a means of effecting such a separation,44 but
interpretation of the results is not straightforward and the technique requires
further evaluation.
Since it is essential that the separation should be carried out immediately
after sample collection (to avoid trace metal redistribution on storage), centrifu-
gation is generally inconvenient and filtration is usually the only practicable
procedure.
Cheeseman and Wilson45 have presented a general discussion of filtration
and filter media in relation to trace metals. The use of membrane filters with a
pore size of about 0.5 μm is generally considered to give a separation of practical
utility, but glass fiber filters and membrane filters of other pore sizes have also
been used for trace-metal studies.
The U.S. Environmental Protection Agency has chosen 0.45-μm membrane
filters as the basis of its standard separation technique,46 and similar filters have
been recommended in a number of authoritative texts.47–49
6 Determination of Metals in Natural Waters, Sediments, and Soils
Paper filters are not recommended for trace metal studies,45 although their
use for determination of dissolved iron appears to be endorsed by the American
Society for Testing and Materials.50 However, the latter publication r ecommends
the use of 0.45-μm membranes for the determination of dissolved copper and
manganese.
Hunt51 have discussed the effect of filtration of water samples prior to
trace metal determination. From this work it is evident that separation of the
dissolved and undissolved fractions of trace metals in water samples is not
simple. Filtration, usually the only practicable procedure, may be associated
with problems of contamination and adsorption, in addition to being subject
to the basic difficulty of incomplete separation.
The adsorption of trace metals during sample filtration has received very
little study. It does not appear to be mentioned as a possible cause of error in
a number of analytical manuals. The limited information available suggests,
however, that it can cause serious difficulties.
Thus, it is not possible at present to recommend a filtration procedure known
to be suitable for all metals. This position is clearly unsatisfactory.
Hunt51 gives the following guidelines for conducting and reporting tests of
filtration systems for trace metal analysis.
1. Tests should be designed to assess both contamination and adsorption. For
this reason, tests involving only filtration of blanks to assess contamination
are inadequate.
2. As a minimum, a blank and a solution containing the maximum dissolved
determinand concentration expected in routine samples should be used in
the tests. It is desirable that a low-concentration solution should also be
used, since adsorption may be more important at lower concentrations.
3. Ideally, actual samples should be used for the tests, in which case prior
filtration is necessary to ensure that the determinand is present only in a
filterable form. This filtration should be carried out using filters of smaller
effective pore-size than those to be tested.
However, if wide variations in a sample matrix are expected, preliminary
studies using solutions of defined composition and appropriate determinand
concentrations may be more useful, particularly for screening a number of filters
and filter holders. Such solutions should not contain matrix components likely
to produce wholly atypical dissolved metal specification patterns, because trace
metal adsorption may be influenced by the adsorbate speciation, and should be
undersaturated with respect to all solid phases (to avoid retention by the filters
of determinand contained in, or adsorbed on, precipitated material). A test with
actual samples should, however, precede routine application of the procedure.
4. It is essential, in view of the likely effects of pH on trace metal a dsorption,
that test samples or solutions should cover the anticipated pH range of
samples. Tests at the extremes of the natural water pH range are inadequate,
since adsorption may be greatest at intermediate pH values.