Tải bản đầy đủ

(Advances in agronomy 104) donald l sparks (eds ) advances in agronomy academic press (2009)








Advisory Board



University of Kentucky

University of Minnesota



University of California,

Texas A&M University

Emeritus Advisory Board Members



University of Delaware

Iowa State University



North Carolina State

Cornell University

Prepared in cooperation with the
American Society of Agronomy, Crop Science Society of America, and Soil
Science Society of America Book and Multimedia Publishing Committee














Department of Plant and Soil Sciences
University of Delaware
Newark, Delaware, USA

Academic Press is an imprint of Elsevier

Academic Press is an imprint of Elsevier
525 B Street, Suite 1900, San Diego, CA 92101-4495, USA
30 Corporate Drive, Suite 400, Burlington, MA 01803, USA
32 Jamestown Road, London, NW1 7BY, UK
Radarweg 29, PO Box 211, 1000 AE Amsterdam, The Netherlands
First edition 2009
Copyright # 2009 Elsevier Inc. All rights reserved.
No part of this publication may be reproduced, stored in a retrieval system or transmitted
in any form or by any means electronic, mechanical, photocopying, recording or
otherwise without the prior written permission of the publisher
Permissions may be sought directly from Elsevier’s Science & Technology Rights
Department in Oxford, UK: phone (+44) (0) 1865 843830; fax (+44) (0) 1865 853333;
email: permissions@elsevier.com. Alternatively you can submit your request online by
visiting the Elsevier web site at http://elsevier.com/locate/permissions, and selecting
Obtaining permission to use Elsevier material
No responsibility is assumed by the publisher 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. Because of rapid advances in the medical sciences, in particular, independent
verification of diagnoses and drug dosages should be made
ISBN: 978-0-12-374820-1
ISSN: 0065-2113 (series)
For information on all Academic Press publications
visit our website at elsevierdirect.com
Printed and bound in USA
09 10 11 12
10 9 8 7 6 5 4 3 2 1



1. Advances in Assessing Bioavailability of Metal(Loid)s
in Contaminated Soils



Kirk G. Scheckel, Rufus L. Chaney, Nicholas T. Basta, and James A. Ryan
1. Introduction
2. Metal Risks in Soil
3. Biological Metal Uptake
4. Metal Extractability to Predict Availability
5. Metal Chemistry
6. Understanding Metal Bioavailability, Bioaccessibility, and Speciation
7. Conclusions

2. Nitrogen in Rainfed and Irrigated Cropping Systems in the
Mediterranean Region



John Ryan, Hayriye Ibrikci, Rolf Sommer, and Ann McNeill
1. Introduction
2. Mediterranean Agroecosystems
3. Perspective on Nitrogen in Agriculture
4. Fertilizer use Trends in the Mediterranean Region
5. Response of Rainfed Crops to Nitrogen Fertilizer
6. Assessing Soil Nitrogen Status for Crop Yields
7. Nitrogen Fixation Under Mediterranean Dryland Conditions
8. Potential Losses of Nitrogen in Dryland Cropping
9. Integrated Cropping Systems: Implications for Nitrogen
10. Nitrogen in Supplemental Irrigation Systems
11. Nitrogen Tracer use in Rainfed Cropping Systems
12. Modeling of Nitrogen in Rainfed Cropping Systems
13. Future Perspective




3. Biogeochemical Processes Controlling the Fate and Transport
of Arsenic: Implications for South and Southeast Asia


Scott Fendorf and Benjamin D. Kocar
1. Introduction
2. Arsenic Aqueous Chemistry
3. Arsenic Surface and Solid Phases
4. Desorption of Arsenic in Soils and Sediments
5. Biogeochemical Processes
6. Processes Controlling Arsenic Concentrations in South(east) Asia
7. Summary and Conclusions

4. Inorganic and Organic Constituents and Contaminants
of Biosolids: Implications for Land Application



R. J. Haynes, G. Murtaza, and R. Naidu
1. Introduction
2. Sewage Treatment Processes
3. Composition of Biosolids
4. Nutrient Content and Release
5. Heavy Metal Contaminants
6. Organic Contaminants
7. Synthesis and Conclusions



Numbers in Parentheses indicate the pages on which the authors’ contributions begin.

Nicholas T. Basta (1)
School of Environment and Natural Resources, The Ohio State University,
Columbus, Ohio, USA
Rufus L. Chaney (1)
USDA-ARS, Environmental Management and Byproduct Utilization Laboratory,
Beltsville, Maryland, USA
Scott Fendorf (137)
Stanford University, Stanford, California, USA
R. J. Haynes (165)
School of Land, Crop and Food Sciences/CRC CARE, The University of
Queensland, St Lucia, Australia
Hayriye Ibrikci (53)
Soil Science Department, Faculty of Agriculture, C
¸ ukurova University, Balcali,
Adana, Turkey
Benjamin D. Kocar (137)
Stanford University, Stanford, California, USA
Ann McNeill (53)
Adelaide University, Roseworthy Campus, Adelaide, South Australia, Australia
G. Murtaza (165)
Centre for Environmental Risk Assessment and Remediation, Division of Information Technology, Engineering and the Environment, University of South Australia, Mawson Lakes Campus, South Australia, Australia and Institute of Soil and
Environmental Sciences, University of Agriculture, Faisalabad, Pakistan
R. Naidu (165)
CRC CARE, Salisbury, South Australia, Australia
James A. Ryan (1)
USEPA, Cincinnati, Ohio, USA
John Ryan (53)
International Center for Agricultural Research in the Dry Areas (ICARDA),
Aleppo, Syria



Kirk G. Scheckel (1)
USEPA, National Risk Management Research Laboratory, Cincinnati, Ohio, USA
Rolf Sommer (53)
International Center for Agricultural Research in the Dry Areas (ICARDA),
Aleppo, Syria


Volume 104 contains four outstanding reviews on timely topics that will
be of interest to plant, soil, and environmental scientists. Chapter 1 is a
comprehensive review on frontiers in assessing the bioavailability of metal
(loids) in contaminated soils. Topics that are covered include metal risks in
soils, biological metal uptake, metal chemistry, metal extractability and
prediction of availability, and advances in understanding metal bioavailability,
bioaccessibility, and speciation. Chapter 2 discusses nitrogen in rainfed and
irrigated cropping systems in the region including Mediterranean agroecosystems, fertilizer use trends, response of rainfed crops to nitrogen
fertilizer, nitrogen fixation under Mediterranean dryland conditions, and
modeling of nitrogen in rainfed cropping systems. Chapter 3 covers the
biogeochemical processes that impact the fate and transport of arsenic with
specific emphasis on South and Southeast Asia. Processes that are critical
include ion displacement, desorption, reduction of arsenate to arsenite, and
reductive dissolution of Fe- and Mn-(hydr)oxides. Chapter 4 provides a
thorough treatment on inorganic and organic contaminants in biosolids and
impacts on application to land. Discussions on sewage sludge treatment
processes, nutrient content and release, heavy metal contaminants, and
organic contaminants are provided.
The authors are congratulated on their first-rate reviews.
Newark, Delaware, USA




Advances in Assessing Bioavailability
of Metal(Loid)s in Contaminated Soils
Kirk G. Scheckel,* Rufus L. Chaney,† Nicholas T. Basta,‡
and James A. Ryan§,1

1. Introduction
2. Metal Risks in Soil
2.1. Bioavailability and soil element risks
2.2. Phytotoxicity risks from soil elements
2.3. Risks to soil organisms
3. Biological Metal Uptake
3.1. Risks through soil ingestion
3.2. How much soil do children ingest?
3.3. Food-chain transfer and risks
4. Metal Extractability to Predict Availability
4.1. In vitro bioaccessibility
4.2. Common soil extractions to predict risk of phytotoxicity
or food-chain risk
5. Metal Chemistry
5.1. Metal equilibrium in soils
5.2. Metal speciation in soils
6. Understanding Metal Bioavailability, Bioaccessibility,
and Speciation
6.1. Lead
6.2. Arsenic
7. Conclusions




USEPA, National Risk Management Research Laboratory, Cincinnati, Ohio, USA
USDA-ARS, Environmental Management and Byproduct Utilization Laboratory, Beltsville,
Maryland, USA
School of Environment and Natural Resources, The Ohio State University, Columbus, Ohio, USA
USEPA, Cincinnati, Ohio, USA

Advances in Agronomy, Volume 104
ISSN 0065-2113, DOI: 10.1016/S0065-2113(09)04001-2


2009 Elsevier Inc.
All rights reserved.



Kirk G. Scheckel et al.

The term bioavailability has many different meanings across various disciplines
of toxicology and pharmacology. Often bioavailability is concerned with human
health aspects such in the case of lead (Pb) ingestion by children. However,
some of the most contaminated sites are found in nonpublic access facilities
(Department of Defense or Energy) or in remote regions as a result of mining or
industrial practices in which ecoreceptors such as plants, animals, and soil
organisms are the primary concerns as well as the potential for food-chain
transfer. In all cases, the endpoint requires movement of the element across a
biological barrier. The still utilized approach to base risk assessment on total
metal content in soils is an outdated endeavor and has never been proved to be
scientifically sound. Yet to reverse this trend, much work is required to establish
baseline bioavailability measurements and to develop complementary methods
that are capable of predicting bioavailability across a whole range of impacted
media in a cost-efficient manner. Thus, regulators have recognized site-specific
human health risk assessments play a key role in decision-making processes at
contaminated sites.
Bioavailability issues surrounding metal-contaminated soils and media have
been an area of intense research. For obvious ethical reasons, we cannot solicit
humans, in particular the sensitive population of children, from the general
population for experimental purposes to examine the long-term harmful effect
of metals in soils. However, some adult human feeding studies have been
accomplished under tight medical supervision and with very small doses. One
option to understand and relate bioavailability in humans is to employ animal
surrogates; however, the physiology of most animals is different than that of
humans but good correlations have been achieved despite the dose–response
paradigm not being identical. The biggest drawback of in vivo studies to examine
metal bioavailability to an appropriate ecoreceptor, be it human, plant, or soil
organism, is the tremendous cost and time involved relative to chemical and
physical surrogates. Chemical surrogate methods generally only require knowledge of the total metal content so that a percent bioaccessible number can be
generated from in vitro extractions that simulate digestive systems or mimic
responses to sensitive ecoreceptors. However, there is not a consensus as to
which of the many in vitro methods is the best analogy to an ecoreceptor uptake
and the same can be said for in vivo animal models to mimic human response as
well. Further, there is yet to be a single in vitro method that can account for more
than a few elements for a specific exposure pathway (e.g., Pb and/or arsenic (As)
for human health). These in vitro tests require honest and accurate validation
against in vivo bioavailability measurements, but most of all would benefit from
metal speciation methods to identify the forms of metals allowing their release.
Adaptation of spectroscopic speciation techniques to identify metal(loid) phases
is extremely beneficial in bioavailability research to understand the variability of
biologically available metal uptake, to manipulate the ecosystem to reduce
bioavailability via in situ amendments, to monitor the long-term stability of
elements to ensure bioavailability indicators do not change over time, and to
develop comprehensive predictive models based on speciation.

Advances in Assessing Bioavailability of Metal(Loid)s in Contaminated Soils


1. Introduction
In most cases, the toxicity of contaminants depends on how much of it
is absorbed into the body or taken up by plants. For soil contaminants where
human exposure is by ingestion of soil or plants and organisms produced on
the soil, toxicity depends on absorption into the gastrointestinal (GI) system.
Information on how well a contaminant is absorbed into the GI system is
important to determining how much of a contaminant humans can be
exposed to before health effects occur. Because typical health effect dose–
response assessments (and resulting oral reference doses (RfDs) and cancer
slope factors (CSFs)) are generally expressed in terms of ingested dose (rather
than absorbed dose into the organism), accounting for potential differences
in absorption between different exposure media can be important to site risk
assessment (USEPA, 1989). Thus, if the oral RfD for a particular metal is
based on bioavailability studies in water, risks from ingestion of the metal in
soil or plant produced on the soil might be (likely is) overestimated. Minor
adjustments in oral bioavailability based on nonrelevant exposure pathways
can have significant impacts on estimated risks and cleanup goals for
hazardous waste sites (USEPA, 1989).
It is increasingly recognized that the response of an at-risk population is
not controlled by the total metal concentration, but instead is controlled by
only the biologically available portion, which is dependent on the route of
exposure, the pharmacokinetics of the organism, and the speciation of the
In spite of the earlier understanding, the complexity of metal-contaminated sites has and continues to be simplified to a measure of the total metal
content. Regulations on the fate and effects of metals in the environment
based solely on total concentrations are no longer (perhaps never has been)
valid, state-of-the-art, or scientifically sound. A vast amount of knowledge
clearly illustrates the decisive role of metal speciation when metal bioavailability and phytoavailability in the environment have to be assessed (McNear
et al., 2007; Ryan et al., 2004). While total metal content is a critical
regulatory measure in assessing risk of a contaminated site, total metal
content alone does not provide predictive insights on the bioavailability,
mobility, and fate of the metal contaminants. Thus, a better understanding of
the nature of the chemical and physical interactions of contaminants with soil
constituents can increase the scientific understanding and lead to regulatory
and public confidence in the use of bioavailability adjustments. Predictions
of long-term stability rely on a mechanistic understanding of how
contaminants are stored or sequestered within the soil.
Bioavailability processes are defined as the individual physical, chemical,
and biological interactions that determine the transfer of chemicals
associated with soils to plants and animals. Bioavailability processes are


Kirk G. Scheckel et al.

embedded within existing human health and ecological risk frameworks to
reduce uncertainty in exposure estimates and improve risk assessment
(USEPA, 2007b). In both ecological and human health risk assessment,
bioavailability is usually reflected in default values or site-specific data that
are inserted into exposure equations. Although a multitude of processes can
affect bioavailability, a typical risk assessment generates one value that is used
to adjust the applied dose. For this reason, many bioavailability processes are
hidden within risk assessment, and assumptions made about these processes
are sometimes not clear. Although long employed in toxicology and agricultural sciences, the concept of bioavailability has recently sparked the
interest of the hazardous waste industry as an important consideration in
deciding how much waste to clean up. This interest stems from observations
that some contaminants in soils appear to be less available to cause harm to
humans and ecological receptors than is suggested by their total concentration, such that cleanup levels expressed as total concentrations poorly
correlate with actual risk. Correct characterization of bioavailability in
contaminated soils and sediments may indicate that greater levels
of contamination can be left untouched without increased risks, thus,
reducing cleanup costs and reducing volumes of contaminated media
requiring intrusive remedial options (USEPA, 2007c). However, in order
to pursue this concept in risk assessment critical knowledge of bioavailability
processes and spectroscopic speciation techniques are required to develop a
mechanistic understanding of the bioavailability processes to improve the
science of risk assessment to develop predictive models derived from sound
research. Further, chemical, environmental, and regulatory factors must
align in considering bioavailability processes that influence risk-based
decision-making (NRC, 2003).
Because the fraction of a soil element which can actually be absorbed by
an organism to cause harm depends on the chemical forms present and
physical/chemical properties of the soil, in both risk assessment and remediation evaluation, the fraction of a soil element which can actually cause
harm must be identified. This fraction is ultimately defined as the bioavailable
fraction, and because measurement of the bioavailable fraction is timeconsuming and expensive via in vivo animal feeding studies, in vitro chemical
methods are being developed to estimate the bioavailable fraction. In the
case of ingestion of soil, the in vitro or chemical estimation method has been
labeled ‘‘bioaccessible’’ (to avoid confusion with ‘‘bioavailable’’) and is a
measure of the amount of metal that can be liberated from the soil matrix,
thus not a measure of the amount of metal that moves across the
GI epithelium to harm internal target tissues and organs. Extensive progress
has been made in development of soil Pb and As bioavailability testing in
conjunction with bioaccessibility methods (Drexler and Brattin, 2007;
Rodriguez et al., 2003; Ruby et al., 1993, 1996). Additionally, great effort
has been wasted in planning inconsequential research efforts to develop

Advances in Assessing Bioavailability of Metal(Loid)s in Contaminated Soils


bioaccessibility methods, which try to match all digestion processes without
a valid bioavailability endpoint as a comparison. In the end, an in vitro
bioaccessibility method only needs to be well correlated with an acceptable
in vivo bioavailability model. Actually, the simpler and less expensive the
bioaccessibility method can be made, the better, as long as the correlation
with bioavailability is high. Further, it is necessary for the tests to be
reproducible in laboratories across the globe, which has not been the case
for many of the bioaccessibility methods available today. Further, for such
methods to be relevant to testing of remediation methods, changes in
bioavailability due to field treatments should be reflected in the bioaccessibility test results. In the case of soil Pb, in situ remediation using phosphate
and other treatments have been proved to reduce bioavailability to pigs, rats,
and humans, but the bioaccessibility test conducted at pH 1.5 does not
measure this 69% reduction in bioavailability to human adults while testing
at pH 2.2 or 2.5 does reflect the effectiveness of the soil treatment
(Ryan et al., 2004). Other simple chemical tests have been shown to suffer
significant flaws in that the extraction causes changes in chemical speciation
during the test, and have not been shown to correlate with bioavailability
changes due to soil treatments. Further, it is necessary to have a valid
measure of why the bioavailability or bioaccessibility of samples are different
and whether the changes are persistent; thus, the need for metal speciation.
For sensitive ecoreceptors (plants, animals, and soil organisms), where
testing with the organism to be protected is more readily conducted,
chemical methods have been developed which integrate potential toxicity
across soil properties including pH which often strongly affects bioavailability.
Mild neutral salt extractions (similar to the first extraction step of a
sequential extraction procedure) are often found to be effective methods.
However, assessment of potential toxicity by adding metal salts to uncontaminated soils substantially fails to mimic field contaminated soils because
elements react with soils, and metal salt additions alter soil pH and do not
account for the aging effect of metals in soils. Traditional toxicology
approaches of adding element salts and immediately measuring toxicity are
clearly inappropriate, and can cause serious artifacts due to pH change
resulting from the metal salt addition, or formation of soluble metal
complexes which temporarily increase or decrease element bioavailability.
Thus, testing of potential toxicity has as many problems as testing of bioaccessibility. It seems clear that by taking present knowledge into account,
effective toxicity testing, bioaccessibility evaluation, and risk assessment can
provide massive savings to the public in dealing with contaminated soils.
The extent to which metals are bioavailable has significant implications
on human and ecological health following exposure and on potential remediation of contaminated sites. Characterization via speciation of insoluble
metal phases in contaminated soils and sediments may indicate that greater
levels of contamination can be left untouched without increased risks, thus,


Kirk G. Scheckel et al.

driving reduced cleanup costs and limited volumes of contaminated media
through less intrusive remedial options. A mechanistic understanding of the
bioavailability process in relation to metal speciation will allow development
of predictive models and improvement of risk assessment. Further, chemical,
environmental, and regulatory factors must align in considering bioavailability
processes that influence risk-based decision-making (NRC, 2003).
In both ecological and human health risk assessment, bioavailability is
usually reflected in default values or site-specific data that are inserted into
exposure equations. Although a multitude of processes can affect bioavailability,
a typical bioavailability assessment generates one value that is used to adjust the
applied dose. The Risk Assessment Guidance for Superfund, Volume I: Human
Health Evaluation (Part A) (RAGS) (USEPA, 1989) supports the consideration
of bioavailability in the determination of site-specific human health and environmental risks. This guidance has been used to support bioavailability adjustments across different routes of exposure at contaminated sites. However, the
use of bioavailability information in site-specific risk assessment has not been
widespread (due to limited data, uncertain methodologies, and lack of method
validation). The primary impediment to the broad use of bioavailability data in
risk assessment and decision-making is the absence of rapid and inexpensive
tools that can generate reliable relative bioavailability (RBA) estimates in the
receptors of concern. It is in this context that coupling in vivo bioavailability,
in vitro bioaccessibility, and speciation research can fill many data gaps to aid in
understanding and predicting bioavailability.
The speciation, or chemical form, of metals governs their fate, toxicity,
mobility, and bioavailability in contaminated soils, sediments, and water.
Different chemical forms of metals, for example, can differ greatly in the
amounts taken up by organisms. The varying bioavailability values of different
metal species is a large reason for the wide range of bioaccessibility values
measured using standardized in vitro analyses of different soils. Other interactions between metals and soil components also govern speciation and affect
bioavailability. The influence of the soil matrix on metal(loid) availability is in
constant dynamic equilibria with multiple independent variables such as solid
mineral phases, exchangeable ions and surface adsorption, nutrient uptake by
plants, soil air, organic matter, and microorganisms, and water flux.
However, determining speciation is not a trivial task, particularly at low
concentrations in a complex matrix such as soil. To assess these chemical
properties and to accurately gauge their impact on human health and the
environment we need to characterize metals at the atomic level with
spectroscopic techniques. This research must move beyond operationally
defined sequential extraction methods and utilize analytical instruments that
are capable of identifying metal species (D’Amore et al., 2005) Researchers
have used advanced synchrotron radiation methods to elucidate the true,
in situ speciation of metal contaminants. Synchrotron techniques include
X-ray absorption near-edge spectroscopy (XANES), which identifies the

Advances in Assessing Bioavailability of Metal(Loid)s in Contaminated Soils


oxidation state and first coordination shell and X-ray absorption fine structure (XAFS) spectroscopy provides information on coordination environment of a selected element as well as interatomic bond distances and identity
of nearest-neighboring atoms to determine speciation. These methods can
also be used in conjunction with statistical methods (principal component
analysis and linear combination fitting) to determine chemical phases via a
finger printing process with a library of known reference standards.
Although most soil criteria and regulations for metals are still based on the
total concentration of the metal in question, it is becoming more and more
evident that spectroscopic speciation is vital for regulatory risk assessment of
environmentally relevant metals in conjunction with in vivo animal data and
validated in vitro extractions for human health effects and plant uptake/foodchain transfer for sensitive ecoreceptors. These innovative research tools are
expanding our ability to directly identify the role of metal speciation on
many dynamic processes that influence bioavailability and risk.
The application of synchrotron techniques for the speciation of metals to
assess bioavailability seems logical to this chapter, but to the common
regulator a simpler approach has been to pick a fractional number relative
to the total concentration of a metal in order to establish a cleanup standard.
Fortunately from a human health perspective, the common regulator
approach is significantly conservative almost to a point that hinders common
sense for site remediation. A good example of this is arsenic which is
regulated assuming 100% bioavailability, yet several studies have demonstrated that absolute bioavailability of arsenic at most sites can be as low as
20% through matrix effects or natural attenuation processes. If a lower
bioavailability value can be utilized at a site, then the effective cleanup
standard is raised resulting in significant savings in remedial clean up costs
without harm to human health or the natural environment. However, few
speciation studies have truly taken on the task of addressing bioavailability
from start to finish—meaning many synchrotron-based studies will broadly
state that their results support an understanding or prediction of bioavailability but provide no real data on bioavailability to support the claim.
There is much speciation research needed to complement in vivo and
in vitro research on metal bioavailability that can lead to effective predictive
models on the long-term fate of contaminants.

2. Metal Risks in Soil
2.1. Bioavailability and soil element risks
The focus of this chapter is on the potential for adverse effects of soil
elements to organisms; specifically soil organisms, plants, livestock, wildlife,
and humans which ingest soils and crops grown on soil. The most common


Kirk G. Scheckel et al.

understanding of bioavailability of a soil element is the fraction of total soil
element which can be absorbed into an organism and cause an adverse or
beneficial effect in the exposed organism. In its concern with direct
ingestion of soil, the USEPA has defined bioavailability as the fraction of
an ingested dose that crosses the GI epithelium and becomes available for
distribution to internal target tissues and organs (USEPA, 2007b). From this
definition, bioavailability can be divided into two kinetic steps: (1) dissolution and liberation of the metal in GI fluids and (2) absorption of the metal
across the GI epithelium into the blood stream. Either of which can be ratelimiting to element bioavailability. Combining the variability of geochemical
forms of elements in contaminated soils with dissolution chemistry and
biological absorption processes in the GI tract is a complex endeavor but
should be a call to arms for the many researchers pursuing this effort. The
scientific and regulatory communities must push further convoluting of this
complexity by recognizing that each element has its own specific environmental toxicology; meaning the organism to which a specific element can
cause an adverse effect at the lowest environmental exposure and the
interaction of other factors with that element such as Ca with Pb, Zn
with Cd, Fe with As, and Cu with Mo. In some cases, the key interaction
which affects element risk is related to dissolution from ingested soil, while
in other cases, interaction during intestinal absorption is the key process
which controls risk from an element. This understanding must come from
assessment of the specific pathway from soil to organism for each element
which can harm a sensitive exposed organism. Often children are the most
exposed and sensitive organisms with respect to contaminated soils in urban
areas, but for remote contaminated areas, it is wildlife, plants, or soil
organisms that are likely to be the most exposed and sensitive organisms.
But each element has its specific chemistry in soils, potential for uptake by
plants or soil organisms, and potential to affect consumers of plants or soils.

2.2. Phytotoxicity risks from soil elements
The most sensitive adverse effect of some elements in soils is phytotoxicity.
It seems clear that the first limiting effect of Zn, Ni, Cu, Mn, Al, and
possibly some other elements are phytotoxicity to sensitive plants. Of
course, plant species vary in tolerance of soil elements. And soil properties
can strongly affect phytotoxicity. For cations, acidic soil pH strongly
promotes element toxicity, and the elements react over time increasingly
strongly to lower phytoavailable forms. In the case of Ni, it was shown
by Singh and Jeng (1993) that Ni was about 10-fold less accumulated by
perennial ryegrass over a 3-year test period using experimental methods
which are highly defensible. Initially, such results were explained in terms of
adsorption and diffusion into micropores of the sesquioxides (Bruemmer
et al., 1988). Since then, research has shown that new mineral phases may

Advances in Assessing Bioavailability of Metal(Loid)s in Contaminated Soils


form in Ni enriched soils, both Ni–Al layered double hydroxides (LDH)
and Ni-silicates (Scheckel and Sparks, 2001). And although Zn can also
form such LDH species, the Zn forms are weaker than the Ni forms
(Roberts et al., 2003). Cu and Cd apparently do not form the LDH species
in soils; however, Co may form them (Scheinost and Sparks, 2000).
Figure 1 summarizes known soil reactions of Ni in relation to plant
uptake and Ni phytotoxicity. Some industrial compounds can land on soils
and persist for long periods. For example, NiO dissolves very slowly, with a
half-life of 20.4 years at pH 7.25, and slower with larger particle size
(Ludwig and Casey, 1996). A study of Ni species in a smelter-contaminated
soil at Port Colborne, Ontario found particles of NiO remaining in the soil
more than 30 years after smelting ceased (McNear et al., 2007). They also
found that Ni-LDH had formed in these soils over time, confirming the
practical significance of Ni-LDH formation in contaminated soils.
Ni-sulfides deposited on soils can be oxidized by microbes. Other Ni phases
enter into equilibria with soil sorption surfaces and ligands. Much soluble
soil Ni2þ is chelated or complexed, but the free ion shuttles among sites
based on free energy and binding site specificity. As shown in Fig. 2, grasses
suffer an unusual symptom of Ni-induced Fe deficiency chlorosis in which
the severity follows a diurnal pattern (banded chlorosis). Phytosiderophores
(PSid) are secreted by young grass roots to dissolve soil Fe, and the Fe-PSid
is absorbed by a transport protein specific to the Fe-PSid. At low pH Ni fills
the PSid and can push Fe out by competition, but during the morning pulse
of PSid secretion, some Fe is dissolved and absorbed so part of the growing


Chelated to organic matter
humics and fulvics

Adsorbed on
Fe/Mn oxides

Occluded in
Fe/Mn oxides




Soil solution
Ni2+ + L



L = ligands


Figure 1 Equilibria of Ni in soils in relation to uptake by both dicots and grasses; note
formation of Ni-Al-LDH and Ni-silicate over time which reduces Ni phytoavailability.
PSid are phytosiderophores such as deoxymugineic acid secreted by wheat to chelate
soil Fe.


Kirk G. Scheckel et al.



Ni toxicity symptoms in oat and barley seedlings.

Figure 2 Unique symptoms of Ni-induced Fe deficiency (Ni-phytotoxicity) with diurnal variation in severity which results from Ni preventing Fe-phytosiderophore formation
in the rhizosphere except during morning pulse secretion of phytosiderophore by young
grass roots.

leaf blade receives Fe before it emerges from the culm. As pH is raised, Ni is
bound increasingly strongly by soil sorbents, and forms new solids
(Ni-LDH, Ni-silicates) such that insufficient Ni remains reactive to compete for filling the PSid in the rhizosphere (Kukier and Chaney, 2004).
Simply making soils calcareous can remediate Ni phytotoxicity potential for
species which are very sensitive at acidic pH (Siebielec et al., 2007).
Interestingly, Cu is more bound by organic matter than Fe and Mn oxides,
nor does it form LDH compounds in soils, so as pH is raised and Fe is less
available for chelation by PSid, Cu inhibits Fe uptake in a simple Fe
deficiency pattern (Michaud et al., 2007) rather than the banded chlorosis
caused by Ni and Co. Zn forms LDH compounds, and is readily converted
to lower phytoavailability forms in soil, so that making a high Zn soil
calcareous with reasonable soil fertility remediates Zn phytotoxicity to
sensitive plants (Li et al., 2000).
Unfortunately, others followed the toxicological approach to establish
limits for soil Ni, by adding soluble Ni salts followed by immediate cropping, failing to correct for the metal salt-induced drop in soil pH (Speir
et al., 1999) resulting in exaggerated soil solution Ni concentrations (Oorts
et al., 2006; Rooney et al., 2007; Thakali et al., 2006). Yet others studied
nutrient solutions but did not understand basic metal chelate equilibria in
nutrient solutions and observed apparently higher toxicity at higher pH
(Weng et al., 2003), in strong contrast with the real world (Kukier and
Chaney, 2004; Siebielec et al., 2007). McNear et al. (2007) examined the

Advances in Assessing Bioavailability of Metal(Loid)s in Contaminated Soils


speciation of Ni in Welland loam and Quarry muck soils around a refinery
and relate these findings to Ni mobility and bioavailability. XAFS and X-ray
fluorescence (XRF) showed that Ni–Al LDH phases were present in both
the limed and unlimed mineral soils, with a tendency toward more stable Ni
species in the limed soil, possibly aided by the solubilization of Si with
increasing pH.
Precipitation of some mineral phases in wetland sediments can potentially limit metal bioavailability through sequestration in low-solubility
compounds, such as metal sulfides. Analysis of XAFS data confirmed that
sulfide compounds dominated zinc speciation throughout the sediment in a
study by Peltier et al. (2003). Uptake of trace metals in Phragmites plants was
limited primarily to plant roots, while concentrations of both Pb and Zn in
other aquatic vegetation were significantly elevated, representing a potential
bioaccumulation hazard and possible food-chain transfer concern for local
Another example of synchrotron research to understand plant uptake of
contaminants was conducted by Punshon et al. (2005). Synchrotron XRF
(S-XRF) demonstrated changes in Ni and U distribution in wheat grown
on contaminated soil and the distribution of Ca, Mn, Fe, Ni, and U in roots
of willow growing on a former radiological settling pond, with U located
outside of the epidermis and Ni inside the cortex with confirmation by
microtomography. Further, XRF and XANES linked the elevated Se
concentrations in sediments of a coal fly ash settling pond at the site with
oral deformities of bullfrog tadpoles.

2.3. Risks to soil organisms
Toxicity to soil microbes and fauna has received much study, but often the
methods used suffered from serious artifacts much as noted earlier for
phytotoxicity. Addition of metal salts to soils is even more inappropriate
in study of soil organisms because the organism receives the shock of soluble
added elements rather than the metals equilibrated with the soil. Complexes
of the metals with anions can cause persistence of soluble ions, and high rates
of metal cations can drive pH several units lower greatly increasing soil
metal solubility. The effects of diverse soil properties on metal toxicity to
earthworms are considered by Lanno and Basta (2003), Bradham et al.
(2006), and Dayton et al. (2006).
In addition, remediation of phytotoxicity is often successful for remediation of toxicity to soil microbes and fauna (Brown et al., 2004, 2005, 2007;
Conder et al., 2001). As we have noted, when metals are present at
phytotoxic levels, the recommended remediation treatment would be to
make the soil calcareous to minimize metal phytoavailability and provide a
persistent remediation. Because these treatments give lower and lower metal
bioavailability over time, it generally provides effective protection of soil


Kirk G. Scheckel et al.

organisms. And consumers of soil organisms appear to be protected except
for soil ingestion risks (Pb, As, F) where earthworms can carry a high
fraction of dietary soil into diets of earthworm consumers.
Risks from soil Cd to earthworms and earthworm consumers have often
been overestimated (Brown et al., 2002a,b). In estimating bioaccumulation
ratios, one needs to take into account that the ratios are 10-fold higher for
background uncontaminated soils than for contaminated soils. Predictions
of risks to earthworm consumers have not been confirmed except for the
case of a Cu–Cd smelter at Prescott, UK. Because Zn was not present with
the Cd, earthworms accumulated high body burdens of Cd without injury
that would have occurred from Zn in most contamination cases. In mine
waste studies, Cd bioaccumulation was clearly limited by the presence of Zn
(Andrews et al., 1984).
Tolerance of soil microbes to metals is very complex, and traditional
methods of study by adding metal salts to soils clearly confound the tests.
Soils with deficient Zn have microbes which are less resistant to Zn additions than found in soils with Zn contamination. These findings led
McLaughlin and Smolders (2001) to introduce the concept of ‘‘metalloregion’’ to suggest that some soils may be much more resistant to additions of
Zn than other soils; that is, it would be an error to apply results from the
most sensitive soil to all soils. Although it is clear that white clover rhizobium is relatively sensitive to excessive soil Zn, it is also very sensitive to
simple soil acidity; causation in selection of ineffective nodulating strains
was more affected by low soil pH than by soil Cd or Zn levels (Ibekwe et al.,
1997). In our experience, sensitive plants are less resistant to excessive
bioavailable soil metals than are the microbes in the soil, such that
protection against phytotoxicity protects soil function.

3. Biological Metal Uptake
3.1. Risks through soil ingestion
For selected elements, the element in ingested soil can comprise a risk to
animals or humans and is especially well studied for Pb and As, but also
considered important for F, Hg, and other elements. Soil ingestion circumvents the soil-plant barrier whereby limited plant uptake limits significant
exposure. In soil ingestion, an element must have sufficient bioavailability/
solubility that it can be absorbed in the intestine to a greater extent than if
garden foods growing on the soil were consumed.
It has been recognized for decades that Pb deposited on the outside of
forages can cause adverse effects in grazing livestock. Then, as risks from Pb
in the urban environment were studied in more detail, it became apparent
that Pb-rich exterior soil and dust can be carried into homes and provides

Advances in Assessing Bioavailability of Metal(Loid)s in Contaminated Soils


exposure to young children who do not play outdoors. And that Pb paint
dust ingested by hand-to-mouth transfer could be the important pathway
for Pb exposure. Additional research eventually showed that interior paint
Pb comprised far greater risk than soil Pb (Lanphear et al., 1998). But a key
learning was that soil Pb was a greater risk through soil ingestion than
through uptake by garden food crops (Chaney and Ryan, 1994). Pb uptake
by plants can occur, but uptake of equilibrated soil Pb is small; soil adhering
on low-growing crops is a more important source of Pb risk than is Pb
uptake by plants. Gardening in urban soils is a difficult issue; if gardeners
avoid growing low-growing leafy and root vegetables, and take care to
exclude soil from their homes, gardening can be a safe practice until soils
exceed levels, which comprise a clear risk by soil/dust ingestion.
Soil Pb became a worrisome source of risk to children because Pb has
become widely dispersed in urban soils (Mielke et al., 1983, 2007) as well as
at industrial and DOD sites. Paint, building demolition dispersing interior
paint (Farfel et al., 2005a,b), stack emissions, and automotive exhaust
emissions contributed to urban soil Pb loadings. Center city soils are
considerably more contaminated than suburban soils, although exterior
Pb-paint scrapped to soil can cause massive soil contamination wherever it
occurs, easily causing soil to exceed 10,000 mg Pb kgÀ 1.

3.2. How much soil do children ingest?
Several studies have been conducted to estimate soil ingestion by young
children. Some investigators measured soil on hands of children and how
long after starting play it took for their hands to become contaminated. The
most widely accepted estimate of chronic soil ingestion by young children
were reported by the team of Calabrese, Barnes, and Stanek at University of
Massachusetts. They used ICP-AES and later ICP-MS to measure tracer
elements in feces of children recovered from diapers. They analyzed diets to
allow correction for dietary intake of elements, and provided toothpaste low
in Ti so that fecal Ti might measure soil. Over time, they discovered that
some of the elements they originally used as tracers were present at lower
levels in the fine soil fraction (<250 mm) than in bulk soil (<2 mm), and
thus they had to reassess their whole calculation method (Calabrese et al.,
1996; Stanek et al., 1999). In the end, the data for two populations they
investigated are reported by Stanek et al. (2001). These final estimates of the
distribution of soil ingestion by young children are considerably lower than
the original estimates (final median ¼ 24 mg dÀ 1; SD ¼ 16 mg dÀ 1; 95th
percentile ¼ 91 mg dÀ 1. These data were the original source of information
for the development of the 200 mg soil dÀ 1 assumed soil ingestion by
children used in Superfund Risk Assessment. The original estimate (based
on 2 mm soil, and a different set of elements than used in later estimates) is


Kirk G. Scheckel et al.

now known to be an overestimate of high end normal soil ingestion by
exposed children.

3.3. Food-chain transfer and risks
Plant uptake is essentially the inverse of leaching, with further limitation by
plant processes and tissue barriers to element transport. Chaney (1980)
introduced the soil–plant barrier concept to describe why nearly all animals
are protected from food-chain transfer of nearly all elements in soils. Most
elements are so insoluble or so strongly adsorbed in soils or in plants roots that
they do not reach plant shoots in levels, which comprise risk to highly exposed
individuals. Examples include Au, Ag, Hg, Pb, Cr3þ, Ce, Sn, Ti, Zr, etc.
Another group of elements does not comprise food-chain risk because they are
phytotoxic to plants before the concentration in the plant comprises risk to
consumers; Zn, Cu, Ni, Mn, F, and As are included in this group.
One key group does comprise potential risk to ruminant livestock
consuming forages grown on alkaline soils: Mo and Se. Both of these
elements are less strongly adsorbed in alkaline soils, so that if the alkaline
soil is Mo or Se enriched, plants may accumulate higher concentrations.
Under worst case conditions, plants accumulate high levels without
suffering phytotoxicity, and ruminant livestock are sensitive to Mo. Excessive Mo intake inhibits absorption or use of Cu in ruminants. Cu deficiency
has commonly occurred when forages contained excessive Mo. The Mo
case is focused on ruminants because monogastric animals are much less
sensitive to dietary Mo. Plants are essentially insensitive to Mo at levels
which are already toxic to ruminants. Soil and biosolids Mo risks were
reviewed by O’Connor et al. (2001) and a limit of 40 mg Mo kgÀ 1 dry
biosolids suggested as a regulatory limit. This suggested limit considered the
mixture of grass and legume crops normally consumed by ruminants, and
the usual mixture of feedstuffs provided by producers, and that forage
production on an alkaline soil which promoted Mo uptake also promotes
Mo leaching over time which reduces Mo risks. McBride and Cherney
(2004) considered this much Mo in biosolids to be a significant risk but
focused on feeding legumes only, a ridiculously impractical diet, and
assumed all feed was grown on alkaline soils enriched in Mo. Ruminants
must be kept on the high Mo diet for some months to deplete body Cu
reserves before an actual adverse effect occurs, and simple Cu supplementation counteracts the Mo toxic effect.
Se is potentially toxic to both monogastrics and ruminants, but because
the leaves contain higher concentrations than grain or other storage tissues,
grazing livestock are usually the most sensitive to excessive bioavailable soil
Se. Under rare conditions, humans have suffered Se toxicity when normal
crops could not be grown due to inadequate rainfall and the alternative food
crops accumulated higher Se than the usual food crop, such as rice (Yang
et al., 1983).

Advances in Assessing Bioavailability of Metal(Loid)s in Contaminated Soils


The principal exception to protection of humans by the soil–plant
barrier is Cd. Cd can be accumulated by plants to levels that harm animals
which chronically ingest the crops. Longer-lived animals are at greater risk,
and humans have experienced Cd disease from crops grown on contaminated soils (Reeves and Chaney, 2008). Another possible exception is Co
which can be accumulated to about 25 mg kgÀ 1 before phytotoxicity is
evident, but ruminant livestock can tolerate only about 10 mg kgÀ 1 in
chronic diets. No case of food-chain Co toxicity to cattle or sheep has been
reported perhaps because Co contamination is so rare.
Cd has caused renal tubular dysfunction and osteomalacia in farm
families who ingested rice for decades from fields contaminated by mine
and smelter discharges of Zn and Cd. The osteomalacia effect, ‘‘itai-itai’’
disease with repeated bone fractures, has occurred at several locations but in
only a small fraction of the population with severe renal tubular disease.
Thus, renal tubular dysfunction is the first adverse effect which must be
prevented by regulatory controls. Other possible adverse effects of dietary
Cd have been suspected, but not observed. Chaney et al. (2004) and Reeves
and Chaney (2008) have discussed the key role of Zn, Fe, and Ca deficiency
in subsistence rice diets which promote human absorption of Cd. These
nutritional deficiencies due to subsistence rice diets are a significant international malnutrition problem for which agronomists and nutritionists are
seeking rice cultivars with improved grain bioavailable Fe and Zn to prevent
widespread adverse effects (Graham et al., 2007). A paper by Reeves and
Chaney (2004) showed that the kinetics of Cd movement through the
intestine was significantly altered in rats with marginal Fe, Zn, and Ca
diets such that net Cd retention was increased 10-fold compared to rats
with adequate nutrition. Growing rice in flooded soils, but draining the soil
at flowering to improve yields allows CdS formed during flooding to be
oxidized and Cd (but not Zn) to be readily absorbed and translocated to
grain. In soils with the normal geogenic ratio of Cd to Zn (about 1 mg Cd
per 200 mg Zn), and crops other than rice, Zn inhibits Cd uptake by the
crop and reduces the bioavailability of Cd in the crop. No adverse Cd effects
have been shown for agricultural food-chains other than subsistence rice.
Biosolids can increase both Cd and Zn in crops such that when Cd in Swiss
chard was increased fivefold, no increase in kidney or liver Cd occurred in
guinea pigs fed the chard (Chaney et al., 1978b); while high increase in
lettuce Cd from a high Cd:Zn ratio biosolids caused large increase in kidney
Cd (Chaney et al., 1978a). Interestingly, Cd in spinach was significantly less
bioavailable than Cd in lettuce to Japanese quail or rats consuming these
foods (Buhler, 1985), and increased plant Zn significantly reduced retention
of plant Cd by quail (McKenna et al., 1992).
Oysters accumulate high levels of Cd, but also accumulate Zn and Fe
which reduce risk from Cd in shellfish. A few sources of Cd are of especially
high potential risk, those without normal Zn cocontamination (Ni–Cd


Kirk G. Scheckel et al.

batteries, Cd pigments, Cd plastic stabilizers, Cd plating wastes, Cd–Cu
smelter emissions). Failure to find Cd-induced renal tubular dysfunction at a
number of sites (UK, US, Germany, and France) where smelter emissions or
mine wastes have caused garden soils to contain 100 mg Cd and
10,000 mg Zn kgÀ 1, while finding 80% incidence of renal disease in older
persons ingesting home-grown rice, highlights the role of rice diets increasing
dietary Cd bioavailability in human risk (reviewed in Chaney et al., 2004).
Consuming threefold normal daily Cd intakes from shellfish diets did not
increase blood Cd in Swedish young women (Vahter et al., 1996), nor did
consumption of high amounts of high Cd oysters by New Zealand
residents cause Cd disease (McKenzie-Parnell and Eynon, 1987; Sharma
et al., 1983).
Several northern European populations have been reported to possibly
suffer adverse effect of dietary Cd at much lower dietary Cd, blood Cd, and
urinary Cd (Buchet et al., 1990; Ja¨rup et al., 2000) than other populations
without identified adverse Cd effects (Ikeda et al., 2003). These reports are
not explicable in terms of known aspects of Cd metabolism, and remain
debated among scientists.

4. Metal Extractability to Predict Availability
4.1. In vitro bioaccessibility
As mentioned, one of the major exposure pathways of heavy metals to
humans is through the incidental ingestion of soil. This is of special concern
for children due to their increased hand-to-mouth activity and enhanced
pharmacokinetics. The availability of metals to the target organism is a
function of many factors, including the way in which the contaminant is
held within the soil matrix and the source of the contaminant. These
concerns have driven the development of in vitro bioaccessibility assays
which introduce a soil sample to a reaction environment similar to that of
the human GI system to mimic bioavailability and typically under worst case
situations (i.e., fasting conditions, low pH, etc.). However, often overlooked in the scenario for incidental soil ingestion is the way in which the
human (animal) GI system treats a contaminant in either a synergistic or
antagonistic manner.
The first step for in vitro studies is to determine the ‘‘total’’ metal content
of a particular sample. However, what researchers call ‘‘total’’ and what is
total can be two distinctly different numbers. Whether using EPA Method
3050B (Acid Digestion of Sediments, Sludges, and Soils) or EPA Method
3051A (Microwave Assisted Acid Digestion of Sediments, Sludges, Soils,
and Oils) (see http://www.epa.gov/epawaste/hazard/testmethods/sw846/

Tài liệu bạn tìm kiếm đã sẵn sàng tải về

Tải bản đầy đủ ngay